In 2005, USGS published the results from a systematic study of the Nation’s potentially active volcanoes in a report called “An assessment of volcanic threat and monitoring capabilities in the United States: Framework for a National Volcano Early Warning System” (Ewert and others, 2005). Updated in 2018 (Ewert and others, 2018), the “NVEWS” assessment determined that the Nation’s volcano monitoring infrastructure is, in the aggregate, about 25 percent of what is needed. Closing this monitoring gap is one of the VSC’s top priorities. We view the widespread adoption of cloud services and machine learning technology as an effective and relatively inexpensive way to do more with what we already have, while also preparing us for the likely future growth of the Nation’s volcano monitoring infrastructure.

Nearly all of the volcano imagery collected and (or) aggregated by the VSC is available to the general public with little or no latency. A variety of external users take advantage of this capability for a number of different purposes, including scientific research, meteorology, search and rescue, outdoor recreation, and aviation. The VSC’s investment in cloud and machine learning technology for the aggregation, analysis, and distribution of volcano imagery will thus serve not only its own needs, but also the larger community of potential end users.

Emergency management officials and the general public rely on the USGS for prompt and accurate forecasts of volcanic activity. Instrumental data such as those from seismometers or gas sensors can be ambiguous, especially if the number of instruments is small. Real-time imagery of a volcano’s summit can, in many cases quickly resolve such ambiguities and thus improve both the timeliness and accuracy of forecasts and warnings.

At present, VSC personnel manually assess individual images for their potential volcanological significance, usually during routinely scheduled daily checks. The gaps between checks are sufficiently large (several hours) that in some cases the onset of a serious change in activity isn’t detected until well after the fact. By flagging sudden or otherwise remarkable changes, continuous algorithmic assessment of incoming imagery has the potential to reduce the latency between the onset of unusual volcanic activity and its detection. Once proven, such technology could also relieve the monotony of regularly scheduled checks, thereby improving employee morale and efficiency while also freeing up time for more interesting and productive endeavors.

Another potential benefit of algorithm assessment is the ability to identify and tag notable characteristics within images so they can be easily searched and retrieved at a later date. In principle, once such algorithms have been trained, they could be used on historical imagery, thus greatly improving the accessibility and usefulness of existing photo archives.

The proposed project leverages two existing resources: (1) the VSC’s 15 years of experience operating telemetered cameras on volcanoes, and (2) the FY2020 collaboration between VSC/CHS to build a cloud-based system for collecting and storing camera imagery. Thus, the project’s primary technical and scientific challenges consist of designing and testing a suite of machine learning algorithms that can scan through a series of images in real-time to identify scenes of potential interest. Although many image classification algorithms have been described in the literature and are in use operationally, the proposed project’s functional requirements are sufficiently complex and novel, that an “off-the-shelf” solution is unlikely.

Most machine learning algorithms must be trained on existing datasets, and this can be a time-consuming process that often involves considerable human interaction. An additional complication is that cameras are deployed in a variety of environments ranging from arctic tundra to tropical rainforests—we may discover that each locality needs a separate training process.

The VSC already has a mature and comprehensive volcano monitoring capability spanning four major facilities in Alaska, California, Hawaii, and Washington. Integrating the results from the proposed project into our existing operational environment should be straightforward, with one possible exception: we may need to hire a new computer scientist to maintain and operate machine learning algorithms.

The VSC had been sustaining its own system of data acquisition, storage and analysis for decades, thus we already have the system administrators and other information technology (IT) specialists necessary to keep the system running. We also have sufficient scientific expertise for the timely analysis and interpretation of incoming data required to fulfill our volcano monitoring mission. Thus, if the proposed project is successful, the longevity and sustainability of continuing its operation will not be in doubt. The uncertainty lies in whether or not the project succeeds, and if so, to what degree.

Primary Contact: Peter Cervelli, pcervelli@usgs.gov

Coastal villages on the north and western shores of Alaska consistently identify declining sea ice and the corresponding increase in severe storm flooding as the most significant risk to infrastructure and the long-term sustainability of their community. At least 60 of Alaska’s 144 threatened communities are located immediately adjacent to the ocean and are frequently impacted by Arctic storms and extra-tropical cyclones. The economic challenge to prevent failure of existing infrastructure in these communities is prohibitively high at an estimated cost exceeding $4.3 billion over several decades. Concurrently, these communities face multiple barriers that make potential mitigation strategies even more challenging than in the contiguous United States, including: small population densities with no road system, no tax base, and limited safety and public health services.

In the absence of expensive engineering solutions, Alaska needs an expanded scientific plan of collaboration to pursue increasingly urgent priorities in addressing the overwhelming magnitude of this problem. The scope of work transcends the purview of any single agency, disciplinary focus, data stewardship or trust responsibility. The proposed Use Case therefore seeks to strategically integrate the most urgent data, forecast delivery, and impact mitigation services needed from partner agencies such as USGS, National Oceanic and Atmospheric Administration (NOAA) Office of Coastal Management, NOAA National Weather Service, Alaska Ocean Observing System (AOOS), Bureau of Indian Affairs (BIA), U.S. Army Corps of Engineers (USACE), the Denali Commission, the State of Alaska Coastal Hazards Program, and the Alaska Native Tribal Health Consortium. Anchorage offers a promising location to reinvigorate this effort of national significance because of the shared administrative space of these multiple institutions within a common geography and jurisdictional authority.

The collaborations identified here directly advance the goals of the 2020 Alaska Coastal Mapping Strategy (following a 2019 Presidential Memorandum), whose implementation plan is currently under development. Likewise, this Use Case directly advances data gaps already identified by the Alaska Water Level Watch, which aims to expand water level observations necessary for flood modeling and storm surge forecasting. The Use Case is also well timed to leverage upcoming workshop discussions at the 3rd Alaska Coastal Mapping Summit on December 9–10, 2020, jointly sponsored by the Alaska Department of Natural Resources, USGS, NOAA, and AOOS.

This Use Case seeks to improve decision support in multiple ways. One line of effort will involve accelerating installation of water level sensors and monitoring equipment for all high-priority coastal communities, ports and harbors to comparable degrees of coverage. Accurate water level and ice observations are basic building blocks for storm-surge forecasting and informed emergency response, yet the conventional technologies are not well suited to the remote and low-infrastructure conditions of Alaska. NOAA is actively investing in a range of technologies and community training activities to fill the gaps in water level and ice observations across the state. More investment is needed both to accelerate the coverage and to standardize the data.

Another line of effort will involve completion of shoreline erosion assessment of both historical rates and future projections. The USGS Pacific and Marine Coastal Science Center has completed most of the historical work in Alaska, but projecting near-term shoreline change requires more attention. The USGS Coastal Storm Modeling System (CoSMoS) is a dynamic modeling approach recently developed to allow more detailed predictions of storm induced coastal flooding, erosion, and cliff failures over large geographic scales. CoSMoS projections are currently available for regions in California, but have not yet been applied to conditions in Alaska.

A third major line of effort will refine stakeholder mapping priorities, costs, data standards and integrate and upgrade nearshore bathymetric and topographic data at the coast. Improved knowledge of precise beach elevation through strategic deployment of tide stations and bathymetric LiDAR acquisition will also yield more accurate erosion measurements and storm surge estimates. All of these lines of effort will be necessary to develop highly useful end products of coastal storm inundation risk assessment maps. These derived new products can then inform and dramatically improve the needed community planning and development of timely mitigation solutions.

The USGS/NOAA Total Water Level and Coastal Change forecasts and the USGS Coastal Change Hazards Portal are the only national scale products for assessing coastal-change impacts of storms on coastal environments. While these state-of-the-art products begin to address a pressing national demand for coastal hazards information in some locations, they are not yet able to reliably and efficiently provide agile delivery consistent with user requirements. For example, users such as NOAA National Weather Service (NWS), National Park Service (NPS), USGS, and local emergency management officials depend on our existing storm forecast products. However, the USGS does not presently have the capability to rapidly update coastal mapping data and ingest real-time observations to provide the best-possible products. The accuracy of the forecast as well as the utility of assessments may be degraded by out-of-date information. The situation is even more patchy and underserved in low-infrastructure locations such as rural Alaska, where significant data gaps still exist.

Coastal flooding forecasts are currently only available at a regional level and only take into account storm surge. Local forecasts are needed at individual communities to effectively respond to and avoid impacts to people and infrastructure. Storm forecasts must also include wave run-up to predict total flooding. Anticipating flood impacts requires combining multiple baseline datasets to convert modeled storm water levels to local elevations. In order to plan for where the coast will be or how high floods will reach, long-term models of flooding and erosion are necessary which tie together all the near-shore and marine processes that affect coastal landforms. With further refinements, the conceptual model can be used to identify vulnerable locations and weather events to assess relative vulnerabilities of specific coastal facilities and systems. This information feeds directly into climate change adaptation planning, community infrastructure placement, and engineering projects in each nearshore community.

Alaska's extensive shorelines are incompletely mapped and poorly monitored for the evaluation of coastal dynamics. Improved coastal mapping data are critical for a multitude of reasons, including informed management of coastal lands and Alaska Native trust resources; safe navigation in an ice-diminished Arctic; and responsible resource and energy exploration. Community resilience to coastal hazards such as flooding, erosion, and tsunami begins with integrated mapping data to establish baseline conditions and forecast change. With renewed focus and funding resources likely to be associated with the Alaska Coastal Mapping Strategy, Alaska’s Federal, State, regional, tribal, academic and private sector partners have the opportunity and capacity to secure a targeted, collaborative, and innovative approach to comprehensive coastal mapping. The return on investment is apparent with the vast potential for improvements to life, safety, and economic opportunities that will accrue to Alaska and the Nation from a solid coastal mapping foundation. With corresponding targeted new investments, USGS could transform this Alaska-based Use Case into a pioneering demonstration of coastal hazard forecast delivery for application at a national scale.

Primary Contact: Dee Williams, dmwilliams@usgs.gov

Understanding the short- and long-distance movements of wildlife is critical for a wide variety of ecological research questions and management decisions. Tagging and tracking of animals is often used to determine locations of animals throughout their annual cycles, understand patterns of habitat use, quantify time spent on certain behaviors, and identify geographic areas repeatedly used by wildlife that may indicate sites of importance to species and populations. Satellite tags offer the ability to monitor movements and locations of animals across vast landscapes, across international boundaries, and throughout all seasons of the year. Satellite tags are used by scientists across throughout USGS and partner agencies and data management and documentation is complicated and data are not currently served through a unified portal. Expanding use of the ASC WTDC will increase efficiency and service of wildlife tracking data used by management agencies such as U.S. Fish and Wildlife Service (FWS), Bureau of Land Management (BLM), Bureau of Ocean Energy Management (BOEM), National Park Service (NPS), U.S. Forest Service (USFS), and state and local agencies to make land and resource management decisions and demonstrate responsive action to Secretarial Order 3362.

Wildlife tracking data collected using satellite reporting tags are either distributed through a range of different and unconnected data servers or are not available due to lack of resources to manage, document, and serve these complex and intensive data. Expanding use of the ASC WTDC will allow for improved opportunities to serve data and provide data for a range of analyses including habitat use, distribution, and the timing of migration for use by scientists, managers, and the public. The ASC WTDC currently serves data for eighteen species of birds and mammals (fig. 5.2).

Satellite tagging is a mature method of tracking wildlife and has been used broadly to track movements of a range of species. Data management and delivery, however, is not consistent among users of these tags. As a result, it is difficult to summarize data or conduct further analyses using these data. The ASC has developed a workflow to ease the burden of data documentation (metadata) and delivery of satellite tagging data. Expanding use of the ASC WTDC will provide benefits to both scientist and users of these data.

The ASC will provide data management and service for all USGS and Department of the Interior (DOI) scientist using satellite tags through the ASC WTDC. Working with Core Science Systems, ASC will seek to serve these data through the Cloud so that they are broadly available for analysis in an EarthMAP analysis environment.

The ASC is a trusted data repository and will continue to serve data in perpetuity. By expanding this service to wildlife researchers across the USGS and DOI, ASC will commit to ensuring these data are continually served as data and computing platforms evolve.

Primary Contact: Christian Zimmerman, czimmerman@usgs.gov

The increasing relevance of Arctic issues to national and global affairs requires a more professional and functional set of maps and shapefiles than was created over a decade ago as an interim representation of significant geospatial and political boundaries. A fundamental problem is that US agencies continue to hold differing definitions and area measurements for what constitutes the U.S. Arctic under ARPA. In fact, geographers at the State Department, NOAA, and the U.S. Coast Guard have recently expressed consternation over how imprecise mapping tools contribute to ongoing problems as basic as quantifying the size and extent of the U.S. Arctic. Making a standard shapefile of the US Arctic using the ARPA definition for the U.S. boundaries and extending to the Exclusive economic zone (EEZ) will help standardize the way US agencies represent the Arctic and calculate the various resources within it. Further, the existing Boundary maps provide no detail whatsoever about terrestrial features of Arctic Alaska, thus contributing to perpetual misunderstanding about the basic geography, scope and scale of Arctic matters. The fact that the Interagency Arctic Research Policy Committee is currently updating the National Arctic Science 5-year plan (2022–26) creates a high-profile opportunity for USGS to apply its geospatial mapping expertise to work with external partners to update and improve on previous tools for timely inclusion of new products in the next five-year plan. The maps and shapefiles will provide a widely used platform to deliver updated and integrated situational awareness and relevant modeling information for all stakeholders concerned about the features of a rapidly changing Arctic.

In total there will be at least five new map products as well as a newly standardized shapefile. The first map depicts the U.S. ARPA Boundary as it relates to Alaska and the Bering Sea. The second map depicts the Boundary from a circumpolar perspective. The third one depicts in poster size detail the Boundary as it relates to the terrestrial continental features of Arctic Alaska. The fourth one depicts in poster size detail the Boundary as it relates to the Aleutian chain terrestrial features of Arctic Alaska. A fifth map will represent the U.S. Arctic boundaries as a smaller subset of the ARPA Boundary map. A final sixth product will be an interactive web application that users can visit to interact with the maps in a selective and dynamic manner with automatically updated source files. A derivative product not owned or controlled by USGS will be a revised ARPA boundary shapefile with updated metadata information and guidance about how to reconstruct the boundary polygon for additional internal use by other agencies and stakeholders.

The updated Boundary maps will improve both understanding and decision making concerning Arctic Alaska in many substantial ways. First, the new products will integrate multiple sets of observational data across disciplines and agencies into a pan-optic format. For example, the maps integrate refined Alaska state waters and shoreline vectors from NOAA, demographic information from US census data and State of Alaska Department of Natural Resources, sea ice coverage to show maximum extent from the National Snow and Ice Data Center, gridded international bathymetry data of the Arctic Ocean, terrestrial land cover from USGS, conservation areas by DOI land management agency, circumpolar July average 10 degree isotherm from the National Weather Service, and major circumpolar transit routes from the U.S. Coast Guard. Second, the new products will provide a basis from which stakeholders can readily grasp essential details necessary to understand fundamental Arctic scales, relationships and issues, and to easily monitor annual changes relevant to Arctic Alaska, such as sea ice extent, land cover, population clusters, and shifting climatological boundary lines. Third, the new products will juxtapose, in a single location, the various alternative scientific and political definitions of Arctic, including: the Arctic Circle, the circumpolar July average 10 degree isotherm, the boreal tree cover limit, and the U.S. Arctic Research Policy Act of 1984.

The proposed maps and web applications under construction will integrate relevant data collected by many different agencies. The maps themselves will not involve interpretive science products nor present the results of otherwise unpublished scientific findings. This effort does not advance the state of science so much as the pan-optic integration of relevant features for more convenient comprehension and use by decision-makers, as well as partners, stakeholders and the general public. Creation of the web application will provide a novel platform for delivery of additional layers of information that will undoubtedly grow over time as EarthMAP services gain momentum.

This Use Case was first conceived and launched in May of 2020 and has gained momentum and further definition through the summer and fall. Cooperation between the Alaska Regional Office and the National Geospatial Technical Operations Center of the Core Science Systems Mission Area has yielded a first round of draft map products by October. These products will be shared with external mapping specialists from partner agencies through a documented peer review process. It is anticipated that final products will be forthcoming and cleared for public dissemination in the fall of 2020.

Once created and finalized, these new map products and web applications will exist in perpetuity with a built-in mechanism for convenient updates. The Use Case will yield a very targeted and relatively high-profile success in delivery of new products and services that are directly responsive to expressed needs of state, federal, and international partners. These products will thereby also play a valuable role in promoting momentum for the EarthMAP vision and brand as USGS increasingly disseminates the concept externally.

Primary Contact: Dee Williams, dmwilliams@usgs.gov

Assessments of climate-change effects on ecosystem processes and services in high-latitude regions are hindered by a lack of decision-support tools capable of forecasting and visualizing possible future landscapes under different scenarios. The increasing threat and occurrence of extensive wildfire in Alaska creates a compelling demand for improved weather observation and forecast (modeled) data. Direct beneficiaries of enhanced model output include the BLM Alaska Fire Service, State of Alaska Division of Forestry, NPS, U.S. Department of Agriculture (USDA) Forest Service, regional and local planning departments, and Alaska Native communities. This Use Case will move relevant climate information out of archival storage and into management decision spaces, and thereby help gage future fire hazard, reduce future wildfire risk, promote ecological resilience and wildlife habitat management, and demonstrate responsive action to Executive Order 13855 and Secretarial Order 3372.

IEM output products are currently partially served on public databases, but largely in formats requiring considerable analytical expertise to access and process. Serving, sub-setting, and summarizing IEM/ALFRESCO (Alaska Frame-Based Ecosystem Code) output for the specific regions, processes, and species of interest to AKCASC and its collaborators would provide better opportunities for leveraged research on vegetation and wildlife responses. Specific tasks include: (1) summarizing the range and timing of vegetation changes across available future climate scenarios (RCPs) with an emphasis on imaging the most characteristic changes by geographical landscape or watershed; (2) identifying the landscapes likely to undergo the most rapid and severe changes as indicators of potential vulnerability and likely management challenges; (3) identifying the landscapes with habitat most likely to become rare and (or) more isolated; and (4) characterizing and comparing the landscape modeling versus climate scenario uncertainties across the replicates of model output to determine which scenario input is the most dominant driver of model output.

The IEM uses the model ALFRESCO to simulate spatially explicit vegetation and fire responses to climate. Future climate scenarios downscaled from global climate models (Meteorological Research Institute CGCM3 and National Center for Atmospheric Research CCSM4) run under different greenhouse gas scenarios have already been used as input to ALFRESCO to develop a large library of replicated future landscape simulations (200 replicates per scenario account for stochasticity in the vegetation model). Summary variables comparing historical (1900–99) and current-to-future (2000–99) landscape characteristics, such as the change in number of fires per century or change in number of vegetation transitions, illustrate the relative changes expected across different parts of Alaska. However, the detailed projections associated with these simulations also allow exploration of other problems, such as habitat changes associated with different vegetation types important to wildlife (for example, changes from spruce to deciduous forest, increases in shrub-tundra, or changes in tree line). To date, these datasets are largely unexplored.

The AKCASC will provide UAF with $450,000 in continuing support of IEM, while the AK Regional Office (with contributions from the Southeast, Midcontinent, and Rocky Mountain Regions) will provide $300,000 to support a work order modification to launch the Use Case. The summary output maps and narratives will benefit from a separately funded work order to develop an improved web interface tool for more seamless interactive geographic display. Data currently exists in NetCDF format at data centers in Fairbanks. In exploration of additional national utility, the Use Case will also actively test a potential university-USGS-Tableau interface built on an Amazon Web Services platform.

Upon completion of the work order modification, a targeted stakeholder workshop is intended to occur in 2021 to evaluate next steps. The workshop will assess the utility of new model summary output products and evaluate prospects to sustain additional effort over the lifetime of the management-action framework, including data storage and delivery obstacles.

Primary Contact: Steve Gray, sgray@usgs.gov

Decision(s) this EarthMAP Use Case will inform—How national parks implement their Resource Stewardship Strategies (RSS). Gauge the value of RSS-relevant information to park natural resource managers, devise plans for USGS delivery of valuable information for RSS implementation, and provide systems of analysis to judge efficacy of RSS implementation.

Relevant decision-makers, stakeholders, and intended information recipients; their interests in and need for improved science information—NPS natural resource managers. The RSS sets resource management goals but implementation is limited by the complexities of dealing with interacting threats, poor understanding of vulnerabilities and species adaptive capacity, and inconsistent monitoring of management effects.

Integrated science and information synthesis necessary to address the problem in context of EarthMAP and the decision(s) to be made. Use Indiana Dunes National Park (INDU) as a prototype for RSS-relevant information delivery by USGS. INDU has the range of landscapes for which RSS-relevant information is needed—from the waters of Lake Michigan to coastal dunal ecosystems to coastal and inland wetlands to rivers, forests, savannas, and prairies. There is a need to synthesize information for a similar range of landscapes across the national park system, providing continuing information on rates of change of a variety of geomorphological (for example, coastal erosion at INDU), hydrological (for example, lake level effects on ecosystem processes at INDU), land use change (for example, Land Change Monitoring, Assessment, and Projection [LCMAP]), and visitor impact trends (for example, remote sensing of social trail development) as well as ecological characteristics such as diversity, resilience, productivity, vulnerability, and connectivity. This information would need to be delivered in at least yearly timesteps. Although implementation of an RSS has park specific aspects, the physical and ecological characteristics listed are representative of a suite of management relevant characteristics that could be honed through Use Cases and made relevant to all parks. Through this Use Case, data gathering and maintenance burden on the parks for RSS relevant information should be minimized if this system of information provision is to be expandable to other parks for the long term.

The RSS is a primary management document for a national park providing “an objective basis for assessing the condition of natural and cultural resources relative to the “desired conditions” established in the park General Management Plan, and (2) to document science- and scholarship-based comprehensive strategies to achieve and maintain these desired conditions.” The process linking USGS capabilities in risk management to RSS needs would emphasize multiple EarthMAP priorities especially “Modeling and forecasting species distribution, abundance, and activity, including drivers and impacts, across multiple spatial and temporal scales” since resource managers often understand their jobs in terms of the species they have responsibility for (FY2021 Annual Program Guidance Ecosystems Mission Area, p 7).

The proposed process, although initially focused on providing a prototype national park more automated delivery of information on status of species and physical processes through time, should improve in resolution as model quality improves and should be applicable across parks if the metrics of landscape quality are honed to be of significant management relevance and generality regardless of the specific circumstances of a given park. Careful selection of metrics can increase the likelihood that they will all be useful for decision making or at least all parks will find useful metrics served from among the metrics presented.

We propose a system of data collection, analysis and prediction that will be generally relevant to the RSS process because it will provide feedback on primary features of importance for the mission of conservation of desired landscape and species conditions and restoration of those conditions when they are not presently met. For example, we anticipate that managers will generally be able to act on RSS goals better if they understand the status of landscape characteristics such as species diversity, resistance of biotic communities and physical attributes to change, productivity, vulnerability, and connectivity. However, that is an assumption to be investigated. Therefore, the first USGS skill set to bring to this protype process of melding RSS and EarthMAP is decision support expertise. DOI decision support experts would have an upfront role investigating how NPS resource managers use information—what is valued information in their decision-making process? Can we create a consistent universe of updatable high value characteristics of NPS managed landscapes to be delivered by EarthMAP?

The Indiana Dunes landscape is often considered to be the birthplace of the modern science of ecology—“Ecological pioneers Henry Chandler Cowles and Victor Shelford discovered a time machine here” since they noted that the arrangement of habitats across the landscape was consistent with habitat succession through time (Jackson, 2011). It may well be the most degraded of the 62 current national parks, in part due to being embedded in one of the most heavily industrialized and populated landscapes in the U.S. Today in the park, USGS explores species composition of offshore communities in the park’s Lake Michigan waters with autonomous underwater vehicles and research vessel-based surveys, remotely monitors animal calls, monitors species, native and invasive, in waterways using eDNA, and is publishing a detailed flora of one of the smallest (6000 ha) yet biodiverse (~1200 native plant species) national parks. The USGS has worked with INDU’s resource management staff in creation of their RSS and with the Great Lakes Research and Education Center that coordinates research across the Great Lakes National Parks and is a key participant in bringing relevant science to the parks.

This prototype RSS-EarthMAP Use Case has the possibility of forging an important proof of concept of the utility of EarthMAP for guiding more effective restoration and conservation decisions in national parks. National parks can broadly be classified as “conservation parks” and “restoration parks,” that is, parks whose management efforts are more aimed at maintaining a status quo by building in resistance to change and parks that are managed with the intent of changing the landscape. As a highly degraded yet biodiverse national park Indiana Dunes could best serve as a prototype of the interaction between the restoration goals emphasized in the Indiana Dunes RSS and the guiding information provided by EarthMAP.

Indiana Dunes is an extensively studied and monitored park with a long history of ecological scientific inquiry and a robust commitment to fostering scientific inquiry that was foundational to the creation of the park’s RSS. Nonetheless, implementation of the RSS, and the validity of the RSS goals, are still only weakly supported by ongoing understanding of status and rates of change of the underlying biological communities and physical environment. There are many USGS datasets that could be consistently useful for updating the managers’ toobox for Indiana Dunes and other national parks. For instance, as NLCD becomes more frequently updated and as USGS products such as LCMAP become more adept at documenting land use change, we can regularly update the status of connectivity across a park and predict future changes more accurately in the park and in the greater ecosystem surrounding the park. We can use emerging models of microclimatic prediction, along with climate projections as an entrée into developing and serving metrics of landscape resilience, such as ability of landscapes to retain species as microclimatic patterns change. Metrics of this sort can be combined to give the RSS implementor access to a compendium of landscape characterization, for example, how connected and resilient is a particular property.

However, perhaps the most important USGS dataset that could be stood up in an improved way to be useful to RSS implementation is the USGS GAP models of species distribution. There are thousands of terrestrial and aquatic models available through this program. In their current state, many could be useful for the RSS implementation process by, for example, predicting likely patterns of diversity across parks. However, these models need to be supplanted with better, perhaps more biophysically based models that can provide more realistic estimates of distribution and likely species abundance and distribution change with environmental change. Currently these are mainly statistical models that compare observations of a species with basic environmental datasets, such as NLCD, to produce correlative models of habitat associations versus providing an understanding of the biophysical niche of the species and how variations and changes in the environmental underpinnings of that biophysical niche might affect distribution. Trying to use the current GAP models to estimate biotic metrics of utility to the RSS, such as biodiversity distributional patterns, will likely highlight the weaknesses of the GAP models but underscore the likely utility of better species distribution models for resource management use. The USGS is a world leader in species distribution modeling and national parks are relatively well documented repositories of species distribution observations and locales where automated species monitoring may relatively quickly become standardized and implemented. As such they should be good proving grounds for general ways to improve species distribution models in a manner that would help managers predict the effects of their management actions more accurately. In this RSS framework we could imagine a system of species distribution model improvement and management scenario automation coming together to guide how park resource managers game out their management options.

One reason to suggest Indiana Dunes National Park as a prototype for interfacing RSS and EarthMAP is that the USGS has produced park-specific preliminary manifestations of some of the likely RSS-relevant EarthMAP outputs—maps of INDU wildlife corridors USGS are an example. However, while USGS has a significant history of responding to NPS information needs at this park, it does not have the history, or current capacity, of delivering a system of analysis and prediction to the park that would provide readily updateable, management relevant metrics that the NPS would receive with minimal effort on their part. The process of vetting the value of information in a systematic way, creating or upgrading predictive models of species distribution and environmental change, and serving this information in a manner that doesn’t require unattainable amounts of continued work from park resource management staff will probably move approximately at the speed of funding but that’s the challenge of the EarthMAP program. Even if we focus the management relevant information delivery to ecological information mainly related to species distribution, we are proposing this as a testbed for significantly improving management relevant species distribution modeling and that part alone will require new resources. There will be a premium on managing dataflows feeding updated models and new metrics in an automated manner that can respond flexibly to NPS management scenarios, that is, systems that provide answers when managers ask “what if.”

RSS-EarthMAP will sustain its relevance if it becomes foundational to the management of parks and has components that will examine how managers use information. If successful, the system would be updateable and would have to be maintained at a national level. The system would be applicable across multiple park RSS so would become national in scope. Because the RSS are not completed nationally, it is likely that the two programs would influence each other’s growth so that the success of EarthMAP in providing actionable intelligence for the RSS implementation, and documentation of the effects of RSS implementation, will likely affect how the RSS are designed in subsequent parks. The RSS and EarthMAP products would interact in an adaptive management framework. For instance, a possible EarthMAP product might include resilience measures, such as how remote sensed primary productivity is affected by variations in climate and how management of different areas of a park affect the degree to which productivity changes as weather varies. As the parks learn more about how their management does, or does not, impart productivity resilience, RSS may be modified or created to consider how successful management is in producing landscapes that resist changes in productivity.

Primary Contact: Ralph Grundel, rgrundel@usgs.gov

Scientific and management communities increasingly recognize the vital role of hydrologic and water quality variability as fundamental drivers of ecosystems and the numerous services they provide to society. Of particular concern are the Nation’s rapidly changing and declining water supplies, the emergence of extreme events under warmer conditions, and alteration of land use and landcover as major stressors to human and ecological systems. Hazards from water quantity (both low and high) represent the second (drought) and fourth (flood) most costly weather and climate disasters in the United States. However, understanding the processes influencing the availability, movement, and quality of water across landscapes is currently hindered by the location, extent, and scale of current hydrologic monitoring. Additionally, there is a significant misalignment of hydrological, physical, chemical, and biological data across space and time to understand and predict how changes in the quantity and quality of water affect important ecosystem services and how these changes might be mitigated. Consequently, there is an urgent need to advance USGS integrated water and ecosystem science to inform decision making across the United States.

Ecohydrology, the interdisciplinary scientific field focused on interactions and associated feedbacks in both space and time between water and ecological systems, presents a unique opportunity to advance EarthMAP by developing, integrating, and applying methods and tools for water observation and prediction while conducting underlying interdisciplinary ecohydrological science to quantify ecosystem services and address stakeholder needs. This effort requires integration of USGS expertise and capabilities and will allow USGS to bring its full science capacity to bear on complex systems where biological, ecological, and water-related factors interact, affecting species, habitats, and ecosystem services important to society. This effort will directly benefit key stakeholders concerned with water-related issues from all regions of the United States, including Federal regulatory and management agencies, such as the FWS, BLM, Bureau of Reclamation (BoR), NPS, USFS, USDA, U.S. Environmental Protection Agency (EPA), USACE, National Oceanic and Atmospheric Administration etc.; State agencies, such as departments of natural resources, fish and wildlife, public health, and water resources; cities and counties; hydropower companies; farmers and landowners; and Tribes.

This EarthMAP project seeks to provide managers, decision makers, and the public with actionable information at multiple relevant spatial and temporal scales about surface-water availability, stream-network connectivity, groundwater level fluctuations, water quality, and ecosystem effects. By integrating these types of data, USGS can provide actionable information to stakeholders so they make more informed decisions regarding water resources and ecosystems that impact issues such as fishery condition, irrigation potential, land productivity, vegetation changes, conservation practices, and recreational resources. Irrespective of the basin and its overarching focus, development of specific water and ecosystem-service modules will be informed by diverse stakeholder engagement (for example, interagency team) and co-production of stakeholder-relevant data products. Thus, we envision a dynamic system that provides managers with timely water quantity, quality, connectivity, and extent information, and associated cascading impacts on vegetation patterns, ecosystem services, recreational use, socioeconomic stability, and others. Specifically, the goal of our project is to produce daily current, short-, mid-, and long-range predictions of streamflow discharge, probabilistic risk of hazard (for example, flood or drought), stream-network connectivity, wetland extent, and relative concentrations of ecologically relevant contaminants, presented at multiple relevant ecohydrologic scales (for example, reach, basin, region) for all locations in the Midcontinent Region. Further, biological data (for example, growth, survival, distribution, abundance, diversity, eDNA data for aquatic and terrestrial species/communities) that align with the probabilistic outputs from our model will be collected and (or) aggregated and served through our visualization platform. These combined products will facilitate integrated analyses to better understand the ecohydrologic connections and feedbacks in the system, ultimately providing more informed resource conservation and management.

This effort represents a great opportunity to integrate USGS expertise and capabilities within and across mission areas, including Water Resources, Energy and Minerals, Core Science Systems, and Natural Hazards (fig. 5.3). USGS has well-established, but separate, research themes focused on water resources including the quantity and quality of water, long-term trends in water availability, and improving abilities to forecast water availability; contaminant sources and prevalence, fate/transport, exposure, aquatic toxicity; and the biology and habitat needs of aquatic-dependent animals and plants.

This new path for USGS to advance coordinated and integrated multidisciplinary science will also contribute to broader Bureau-level priorities, including other applications of EarthMAP and the USGS 21st Century Science Strategy (U.S. Geological Survey, 2021), both of which focus on data integration and modeling, and the production and delivery of actionable information. The USGS possesses considerable resources and capabilities in regard to geology/vadose/soil mapping, water monitoring, water flow and quality modeling, ecosystem-service modeling, land-use/landcover mapping, vegetation sampling and mapping, and animal population dynamics and habitat needs that will make this project successful. For example, by leveraging USGS strengths (for example, Integrated Water Availability Assessments [IWAAs], Next Generation Water Observing System [NGWOS], USGS Ecological Drought Program, Risk Community of Practice [CoP], Community for Data Integration [CDI], Science Analytics and Synthesis, USGS Climate Adaptation Science Centers [CASCs], Great lakes Restoration Initiative [GLRI]) this effort could benefit the EarthMAP pilot project focused on drought in the Colorado River Basin or could support the emerging USGS Initiative focused on post-fire hazards, risk assessments, and decision making in western US watersheds. It will also create opportunities to strengthen Congressional support across administrations, and to enhance and forge collaborative relationships across the Bureau and with other agencies. This multi-Mission pilot project could be broadened to consider the full spectrum of ecohydrological issues relevant to stakeholders, from floods to droughts, and from stream-bottoms to terrestrial systems.

USGS Ecohydrological Science provides an adaptive science framework to observe, synthesize, model, forecast, and deliver decision-support guidance to address current and future management needs, while monitoring, revising, and improving tools and technologies to advance water and ecosystem science in a changing world. The first step of implementing this approach is to assess existing key data and platforms, including surface water, groundwater, water quality, water temperature, water use, soil moisture, snowpack, ecology and the supporting national hydro-geospatial framework. For example, modeling products for water quantity (PROSPER model, Precipitation-Runoff Modeling System, Monthly Water Balance Model, Soil-Water-Balance Model, Operational Simplified Surface Energy Balance, WRF-hydro) and water quality (SPARROW, RHyME2 [Regional Hydrologic Modeling for Environmental Evaluation], SWAT [Soil and Water Assessment Tool]) will be scoured from the newly developed USGS Model Catalog, compiled and synchronized. Second, strengthening and broadening data observation networks and predictions are needed to spatial and temporal coverage of space-based, airborne, aquatic, and terrestrial data observation networks that collect information at multiple spatial scales. Some examples of existing platforms include the USGS National Water Information System, the National Water Quality Portal, eDNA databases, dynamic citizen science efforts (for example, StreamTracker, FLOwPER [FLOwPERmanence]), various State- and watershed-specific databases, and others. A state-of-the-art data-exchange, storage, and delivery infrastructure will be important to support these new robust networks and deliver near- or real-time data. This project will coordinate with efforts like the Water Resources Mission Area NWIS [National Water Information System) Modernization project to ensure we are at the forefront of delivery mechanisms. These models and data will be aggregated by appropriate spatial and temporal scales, and subsequently linked to the National Hydrography Dataset (NHD). Next, coordination of these strengthened data observations will enable USGS scientists and partners to synthesize datasets and models to improve understanding of drivers, responses, and interactions of water-related issues facing ecosystems and society. This information will improve our ability to forecast vulnerable and resilient systems. Finally, these steps in the process to strengthen, synthesize, and predict, will require increased coordination across USGS Mission Areas and partners, but will ultimately be essential to build and deliver decision-support to address important decision making and stakeholder needs. This new integrated and predictive science will help Federal, State, Tribal, and local resource managers with improved planning, strategic implementation and mitigation strategies.

The final output from this project will be an interactive Web interface that provides users with daily current, short-, mid-, and long-range predictions of streamflow discharge, probabilistic risk of hazard (for example, flood or drought), stream-network connectivity, wetland extent, and relative concentrations of ecologically relevant contaminants, presented at multiple relevant ecohydrologic scales (for example, reach, basin, region) for all locations in the Midcontinent Region. Additionally, biological data (for example, aquatic productivity, threatened and species of special concern, invasive species, food web dynamics, eDNA) that align with the probabilistic outputs from our model will be collected and (or) aggregated and served through our visualization platform. Hydrologic outputs could also be used as inputs to risk models to predict the probability of aquatic invasive species (for example, Dreissenid mussels, Carp, Eurasian watermilfoil, smallmouth bass, invasive trout and char, northern pike etc.) invasion, establishment, and spread.

In the first phase of this work, the integrated ecohydrology framework will be applied to three pilot basins in the Midcontinent Region, representing different ecohydrologic contexts and issues (Upper Missouri River basin, Mississippi River basin, and The Great Lakes). These basins will be selected based on the availability of data, the diversity and representativity of ecohydrologic conditions, socioeconomic importance, and partner engagement. Synopsis of potential ecoregion foci, including ecohydrologic issue addressed:

Water-related stress has been studied by the USGS for over a century and humans have benefited from associated ecosystem services for millennia. The relevance of this EarthMAP application will be permanent as long as humans interact and use the landscape since foresters, agriculturalists, land managers and the public will need to know about, mitigate, and adapt to impacts to water supply/quality and ecosystem services that affect their work, livelihood and recreation. The methods and results of this Use Case will be applicable to any watershed or landscape. Further, the integrated data, models, and outputs will be archived in the cloud and independent of the lifecycle associated with locally established servers. Miminal upkeep will be required annually to ensure the operational success of the outputs. Effectiveness of the effort will be evaluated and quantified by the extent of stakeholder engagement, public and manager use of the tools, the effectiveness of tools to accurately portray conditions, and the level of synthesis/integration among earth systems.

Primary Contact: Clint Muhlfeld,cmuhlfeld@usgs.gov

There is a need to integrate the data being collected to better understand triggers for bloom development, persistence, toxicity, and to improve predictive capabilities. Cyanobacteria harmful algal blooms (cyanoHmonABs) and nuisance algal overgrowth cause a variety of environmental health issues that can also have significant socioeconomic consequences in rivers, lakes, wetlands, estuaries, and oceans. We have limited ability to predict variation in cyanotoxin concentrations at large spatial or temporal scales and limited ability to predict HAB-related toxins at all spatial and temporal scales. Land managers and managers of waters for municipal, industrial, and recreational purposes could use predictions of the location and timing of high toxicity HABs and nuisance washup events to mitigate or minimize their environmental health and economic impacts (for example, by switching water sources, posting warnings, etc.).

USGS collects and analyzes hydrological, chemical, environmental, biological, geological, land-use and many other types of data from multiple locations, at a wide range of spatial and temporal scales. Each of these types of data have an impact on the development and character of HABs. The EarthMAP concept can be used to connect this information together in the context of HABs (HABMAP) to not only help USGS provide answers to one of societies critical problems but also to develop the procedures and mechanism needed to apply EarthMAP to other integrated science questions.

Managers and users of waters for municipal, industrial, and recreational purposes could use predictions of the location and timing of high toxicity and nuisance washup events to minimize the environmental health and economic impacts of harmful algal blooms (for example, by adjusting or switching water sources, triggering sample collection, posting warnings, etc.). Improvements in predictability and minimization of harmful effects of algal blooms have been identified as high priority issues in USGS Ecosystems and Water Resources Mission Areas and have been highlighted as part of the Great Lakes Restoration Initiative (GLRI).

Meeting this demand could integrate information to understand, manage, and mitigate factors that lead to HABs with the ultimate goal of building predictive models for a wide variety of ecosystems across North America. The information could be used to develop a well-documented and approved methodology for incorporating available predicted effects of future climate-change with current USGS site-specific streamflow data and information (current hydrographs, flow-duration hydrographs, streamflow statistics, etc.) into a nation-wide decision support tool for planners to use. The proposed HABMAP could incorporate indicators of uncertainty in results and use automated data processing to continuously update available climate scenarios into streamflow information and apply machine learning to develop flow-related predictions of loads and concentrations for selected constituents. The integration of data could also be used to estimate cyanobacteria cell abundance in larger lakes and reservoirs of the United States to assist federal agencies, tribes, states, municipalities, and drinking water utilities with surface water management decision support. Managers and users of waters for municipal, industrial, and recreational purposes could use predictions of the location and timing of high toxicity events to minimize the environmental health and economic impacts of harmful algal blooms (for example, by switching water sources, posting warnings, etc.).

Potential collaborators include USGS Mission Areas­ ­­­Ecosystems, Water Resources, Natural Hazards; Public health officials, municipal planners, agronomists, and the public; Federal Agencies (USACE, FEMA [Federal Emergency Management Agency], EPA, NASA [National Aeronautics and Space Administration], FWS, DOD [Department of Defense], DOT [U.S. Department of Transportation], HUD [U.S. Housing and Urban Development], etc.); State Agencies (Water supply, water quality [DEQ; Department of Environmental Quality], DNR [Department of Natural Resources], economic development, etc.); Local and Tribal Agencies (Water Departments, Waste-water/stormwater planners, Public Works, etc.); University researchers, Engineering companies, Environmental groups, etc.

Federal, State and local health and environmental agencies are highly interested in using real-time data and information to improve HAB identification and reduce the cost of identifying HABs, sampling, determining cell abundance/biovolume, and estimating toxin concentrations to provide meaningful and actionable information related to protection of health and the environment.

Our partners know they need a decision support tool for information related to HABs. They also provide great help with respect to site-specific information, water-quality data, land use data, water use as well as other pertinent information needed to improve our ability to predict HABs. This kind of decision support tool will provide an expanded opportunity to involve our stakeholders in USGS science and this has already occurred in studies where we are already developing predictive models and tools.

The expected impact of this decision support tool is use by managers and users of waters for municipal, industrial, and recreational purposes throughout the nation to better predict the location and timing of high toxicity events to minimize the environmental health and economic impacts of harmful algal blooms.

We will build on a long history of local and regional research including in situ data collection, remote sensing and imagery, data analysis and evaluation, and statistical and process models to develop a field to rivers, lakes, wetlands, estuaries, and oceans understanding of the factors controlling HABs formation.

USGS will use existing or new sites to collect discrete and instantaneously measured water-quality data, field fluorescence and lab values of algal pigment data, phytoplankton community composition and abundance data, field-based hyperspectral imaging, and other relevant data such as wind speed and direction and photosynthetically active radiation (PAR). Surrogate (proxy) models will be developed to relate continuously measured data to discretely sampled data and once verified, data from days and time of scheduled satellite overflights will be used to verify and calibrate the relation between satellite data and site-based data using machine learning and automated data processing. At recreational and drinking-water treatment plant intake sites where data are available, statistical models will be developed to estimate toxin concentrations.

Existing infrastructure in USGS could be integrated for accurate bloom predictions, including real-time measurements of nutrient inputs and eDNA of harmful algae using the next-generation streamgage network, applied in both stream and lake settings, and geospatial mapping of bloom biomass through the Cyanobacteria Assessment Network (CYAN) project. Additional capacity, for example, in remote sensing of benthic algae, hyperspectral imaging of algal types, Autonomous Underwater Vehicle (AUV) mapping, and hydrodynamic models for nutrient and algal transport, could be developed within USGS or through collaborations to complete the analysis and prediction capabilities. We are using molecular tools to measure genetic potential, toxin production intermediaries and toxin composition in a range of laboratory and wild cyanobacterial populations. Molecular data also is used in statistical models to provide advanced warning of a toxic HAB.

Partnerships could form between USGS Ecosystems Mission Area, Environmental Health Program, Water Resources Mission Area, Natural Hazards Mission Area, Great Lakes Restoration Initiative (GLRI); Public health officials, municipal planners, agronomists, and the public; Federal Agencies (USACE, FEMA, EPA, NASA, FWS, DOD, DOT, HUD, etc.); State Agencies (Water supply, water quality [DEQ], DNR, economic development, etc.); Local and Tribal Agencies (Water Departments, Waste-water/stormwater planners, Public Works, etc.); University researchers, engineering companies, environmental groups, and others.

USGS has broad and expanding capabilities across the Region and Science Centers to facilitate the Use Case for HABMAP. With established programs in nutrient loading, streamflow, genomics, and survey assessments across the fields of hydrology, limnology, hydrodynamics, molecular biology, remote sensing, data analysis and modeling. Ongoing research activities include an interagency effort (EPA, NASA, NOAA, USGS) and targeted efforts (for example, GLRI) to develop the Cyanobacteria Assessment Network (CyAN) that utilizes remote sensing to estimate cyanobacteria cell abundance in larger lakes and reservoirs. Another Center is using sophisticated molecular tools to measure genetic potential, toxin production intermediaries, and toxin composition in cyanobacterial composition. Ongoing basinwide research exploring the drivers of nuisance algae (Cladophora) can also be leveraged to identify simultaneously drivers of multiple toxic and nuisance algal blooms, along with partnering agencies and academia (Michigan Tech, Michigan State Univ.). The Great Lakes NowCast, which currently provides estimates of E. coli concentrations at recreational sites using predictive models, could be expanded to include toxin estimates at recreational drinking-water intake sites throughout the region.

Without a better understanding of the triggers for bloom development, persistence, and toxicity at a large scale, and with potential future impacts of climate-change, the HABMAP Use Case has long-term relevance for the nation as HAB-related issues continue to impact users of waters for municipal, industrial, and recreational purposes.

Data from our large monitoring network will be used to evaluate and improve the effectiveness of decision support tool prediction and modelling efforts, and define metrics of performance and success.

Leveraging of research programs will assist in long-term viability of the HABMAP Use Case. USGS has an established reputation examining bloom drivers and predictive models and has established a large network of partners and stakeholders. Partners have an interest in maintaining a scale approach to HABs that will inform their management of local and regional actions that can be taken to improve algal blooms. This scaling approach could incorporate local streamgages and directed sample collection. The approach taken in HABMAP will benefit research in Great Lakes and other waterbodies that have stressors driven by similar conditions: toxic and nuisance algae, agricultural runoff, municipal nutrient loading, urban ecosystems, drinking water, and shoreline recreational activities.

Primary Contact: Victoria Christensen, vglenn@usgs.gov

“I’m using these costly management practices to keep the nutrients out of my water, and you’re telling me that my water quality hasn’t improved?!” This or similar questions are frequently voiced by stakeholders frustrated that investment in nutrient management, as best management practices, has shown minimal effect. We hypothesize that part of why significant change hasn’t been observed is that legacy nutrient sources, which reflect more than a century of agricultural production and purposeful application of fertilizer and manure, provide a steady flux of both nitrogen (N) and phosphorus (P) which has accumulated in groundwater, upland soil, and sediment stored in streams and water bodies. Communicating the longevity and magnitude of legacy nutrients to stakeholders is key to preventing a loss of confidence in their ability to manage and improve their water quality.

Just as predicting streamflow requires an understanding of antecedant conditions combined with recent events, so does the prediction of nutrient concentrations and loads. Nutrient concentrations are linked with recent land management, thus the interest in edge-of-field monitoring, which directly links field management and fluctuations in nutrient and sediment concentrations. In streams, seasonal fluctuations are also observable; however, new additions of nutrients are supplemented by contributions during baseflow and stream-energy driven erosion and redistribution of in-stream sediment. Similarly, nutrients stored in bottom sediments of water bodies can be released when water temperatures increase or bottom waters become depleted in oxygen—common occurrences in summer and early fall. Therefore, legacy sources can directly drive summer and early fall concentrations that foster hypoxia and harmful algal blooms (HABs).

How will nitrate concentrations in groundwater decrease? Will this be spatially consistent? A function of local aquifer properties? Will mixing outpace exfiltration as baseflow?

How will orthophosphate (ortho-P) and sediment-bound P (sed-P) interact? Recent research shows that suspended sediment is enriched in sed-P relative to upland sources; however, it is unknown if this is true in streams of all sizes and how this varies as a function of ortho-P concentrations.

How much of the legacy P pool is bioavailable, specifically the sed-P component which likely varies in grain size, age, and submersion depth along individual stream networks?

How much does stream condition (temperature, discharge, and depth) affect bioavailability of legacy P and how will this change as seasonality of precipitation and nighttime temperatures change as a function of projected climate?

How important is nutrient release from sediment stored in stream channels and lakes to the overall nutrient budget of the downstream receiving waterbodies?

How does a better understanding of legacy nutrients change our anticipated changes in water quality based on current management actions taken throughout the watershed?

Discernable improvement in water quality of impaired waters with large sources of legacy nutrients requires a toolbox of monitoring and management strategies that address both current and legacy sources. It is also important to know which tool in that toolbox is most appropriate for meeting management goals for a given source (fig. 5.4). Legacy pools of soluble nutrients (principally nitrate and phosphate) in groundwater can strongly influence drinking water supplies and baseline conditions and will inherently cause a lag in the observed water-quality benefits of nutrient reductions to upland surfaces. The legacy pools of sediment-bound nutrients (principally phosphorus) in upland soil, streambanks, streambeds, lake sediments, and overbank sediment act as a source and sink for nutrients over different time scales, seasons, and flow conditions. Management techniques may range from on-field conservation practices that reduce nutrient transport and nutrient accumulation in soils to riparian buffers and stream restoration measures that trap nutrients and increase microbially-mediated nitrogen removal to inlake management actions. However, real-time, geospatially-integrated quantification and understanding of these different sources are needed to inform decisions for the prioritization of management options and to provide an understanding of when to expect improvement.

A better understanding of the importance of legacy nutrients is needed by groups such as the Gulf of Mexico Hypoxia Task Force, the Mississippi River Cities and Towns Initiative, parties to the Great Lakes Water Quality Agreement, the Groundwater Foundation, and many other groups that are working to improve water quality across the country.

USGS has developed large-scale surface-water models (for example, SPARROW) and regional groundwater models (for example, Modflow) to describe generalized nutrient transport over large areas, whereas small-scale studies have generally focused on specific transport mechanisms and differentiation of new sources of nutrients and legacy sources (for example, SWAT, sediment fingerprinting, empirical to hydrodynamic-ecological models). However, we currently lack watershed models and communication strategies that couple these approaches and integrate our understanding of nutrient transport processes, especially with respect to legacy pools of nutrients, to enable large-scale predictions in water quality that drive HABs, drinking water restrictions, and hypoxia. Only with a better understanding of all of the sources of the problem—nutrient loading from both current and legacy sources—can we expect to describe variability and forecast change in HABs in the Great Lakes and other water supplies as well as hypoxia in lakes, estuaries, and the Gulf of Mexico.

Current science methods can be advanced to provide full watershed descriptions of current and legacy pools of nutrients, including age, chemical profiles, nutrient gradients and stoichiometry, areal distributions, and transport processes.

Evaluating water quality as a function of both current land practices and legacy nutrient sources will facilitate development of a full ecosystem approach to better forecast changes in the water quality in downstream receiving waterbodies. This information will help to inform decisions about where, when, and how to implement the most effective best management practices and stream restoration measures to improve water quality.

Datasets describing nutrient storage and processes that impact nutrient transport should be identified and catalogued for a specified area to allow for high performance computing and mapping. Identifying and translating these data to cloud optimized geo-tiffs, that could be readily displayed and queried, would improve both access and communication, as well as identifying geographic gaps in our understanding of nutrient pools, transport, stoichiometry, and bioavailability. Data gaps can be filled through field sampling by USGS and partner agencies. Researchers may need to partner with software developers to develop interactive maps to display nutrient pools and transport. Available data resources include ScienceBase, the National Water Information System and Water Quality Portal, and the National Cooperative Soil Survey Characterization Database. Existing watershed mapping tools that could be incorporated include SPARROW Mappers, SWATviewer, and Agricultural Conservation Planning Framework (ACPF) Tool.

Legacy sources reflect an accumulation of excess nutrients in the environment, much of which was purposely applied in agricultural production. Consequently, it is logical that as new nutrient additions are minimized and responsibly managed that the potential mobilization of legacy nutrients may change. It is widely recognized that this is a long-term issue and will require long-term plans for management and assessment of the conditions. Quantifying these legacy nutrient pools is a first step; however, ongoing estimation of the magnitude of these pools in the context of new nutrient contributions, in-stream conditions, and receiving water-body conditions is critical to measuring how effective targeted actions are to reaching nutrient goals. Requirements to sustain this effort include:

Primary Contact: Tanja N. Williamson, tnwillia@usgs.gov

Local, state and federal partners have been using the NowCast system since 2006 to determine when E. coli advisories or closures should be issued using models to predict E. coli concentrations, as studies have consistently shown that elevated E. coli levels, especially from human/anthropogenic sources, relate to increased levels of waterborne illnesses (for example, gastroenteritis) in contaminated beach waters. This system provides a framework to build additional remote monitoring and forecasting of other human and wildlife (for example, bird botulism, an endemic problem in the lower Great lakes) pathogens, as well as non-indigenous species using DNA-based (eDNA) technologies. By incorporating eDNA sampling of species that pose a threat to public health, aquatic communities, and ecosystem services into established NowCast sites and USGS real-time streamgaging stations, we can map locations where these organisms pose the greatest threat; thus, remediation or restoration of degraded habitats can be better coordinated by expanding the existing NowCast system. Incorporating eDNA collection of non-indigenous species and pathogens at current NowCast locations, while adding additional Great Lakes sites and USGS streamgaging locations, will expand the forecasting and tracking of species that indicate a threat or pose the greatest threat to public health and aquatic communities such as E. coli, avian botulism (caused by Clostridium botulinum), grass carp (Ctenopharyngodon Idella), Dreissena spp. mussels, lampreys, milfoils (Myriophyllum spp.) and round goby (Neogobius melanostomus). The product of the expanded NowCast could be used by local, state, federal and academic agencies to provide data for decision making on topics ranging from non-indigenous species control to public health advisories.

Early detection of environmental threats is critical to improve stakeholder decision making. For instance, USGS researchers recently demonstrated the use of a robotic water sampler to collect, process, and preserve environmental DNA (eDNA) from water samples for near real-time stream assessments. The DNA from such collections offer opportunities for targeting multiple species or targets of interest: from microbial pathogens to aquatic invertebrates to fish species over a defined timeline. It is important to note that autonomous, robotic water sampling technologies represent the cutting-edge science in environmental sampling that can overcome the temporal and human (manual) resource demands often associated with traditional eDNA sampling. Thus, near real-time results can potentially be achieved for species of choice (single or multiple), as needed (that is, site- or location-specific monitoring; Asian Carp, dreissenid mussels, sea lamprey etc.). Additionally, the flexibility of eDNA methods allows for resource managers to effectively answer temporal questions that may arise at later dates, along with providing near real-time detection and monitoring of the current species of interest even when conditions are unfavorable for sampling. Because eDNA has broad applications, we can work with decision makers to identify needs/priorities (for example, high risk watersheds to identify threats in Western US).

With the increase of this much needed information which allows for improved decision making, expected impacts include direct funding for restoration of degraded habitats, improvements to infrastructure, implementation of best management practices (for example, alterations to shoreline to improve water circulation to reduce contaminant loadings at embayed beaches); citizen engagement (shoreline water quality, bird botulism; nuisance algae etc.); more effective early detection-rapid response actions.

NowCast has been in operation since 2006 and has a total of 22 sites on Lakes Erie and Ontario. The USGS has strong cooperator ties with many local and state agencies who are collaborators on the NowCast. Additionally, the Wisconsin Department of Natural Resources has partnered with EPA and USGS to collect E. coli samples and map the results at another 310 sites within the Lake Michigan and Lake Superior watersheds. Michigan Department of Environment, Great Lake and Energy collects E. coli samples along their Great Lakes and inland beaches as well. The USGS also has thousands of real-time streamgaging stations throughout the Midcontinent Region.

Collection of eDNA or other biological samples (viruses, bacteria, nuisance algae and cyanobacteria, etc.) at selected streamgages by trained USGS personnel during their regular site visits requires little additional investment. USGS Science Centers have staff with experience utilizing robotic autosamplers and the expertise to combine real-time and high-throughput molecular data with environmental data (for example, water quality and weather) into an expedient data science pipeline to make predictions about species occurrence. Integrating predictive capability of NowCast water quality parameters (from streamgages) with eDNA studies provide opportunities to develop new predictive/forecasting models for current and emerging invasive species and integration with the USGS NAS database, which now reports eDNA results. Additional databases are also currently available using robotic and manual sampling, including Sepulveda (2020).

With the addition of the described methodology, biological data could be collected at sites with streamflow and water quality data for concomitant analysis and delivered using continuously updated maps showing distribution of species of interest in relation to area, time, water quality, season, etc. The use of machine learning algorithms can also be employed to estimate changes in species distribution as it relates to all other data collected.

The NowCast is currently operating as a standalone database and visual tool that is used by local, state and federal partners (fig. 5.5). It could be used as a baseline for further EarthMAP utilities and Use Cases. By incorporating additional tracking and methodology for environmental threats and utilizing existing USGS expertise, NowCast could expand to incorporate the entire Midcontinent Region and eventually the whole US and territories. Using existing data harvesters, NowCast is able to pull data from multiple sources including, but not limited to USGS, NOAA, and other online servers, which could be expanded to partner labs, universities, and state and local databases provided the data is available online.

Autosamplers, including highly advanced systems such as the Environmental Sample Processor (ESP), provide opportunities to collect samples at predetermined times and can hold such samples in situ until they are picked up for further processing. Moreover, certain advanced ESPs have built-in capabilities for collecting and processing samples for pathogens and (or) their indicators, as well as harmful algal blooms (HABs). Such technologies can be transferred to invasive species based on location-specific needs. With automated sampling, we can better coordinate time of sampling (for example, catching spawning events; disease outbreaks), reduce manual training, and sample processing needs. Because eDNA has broad applications, we can work with decision makers to identify needs/priorities (for example, high risk watersheds to identify threats in Western US).

Bringing together the database and visualization capabilities of NowCast and the remote collection capabilities of the USGS requires incredible computing power, but when implemented properly would provide a robust and user-friendly web-based tool that is adaptive to the needs of the stakeholders.

The eDNA technologies are evolving at a rapid rate. The Use Case summary provided here may easily be transferred to novel and emerging invasive species, pathogens or microbes, consistent with USGS priorities and partners’ needs. Some western states are already using existing streamgages for biological threats which can be further expanded to meet regional or national priorities. The use of eDNA for detection of invasive species, pathogens or microbes in conjunction with water quality parameters from streamgages will identify hotspots of potential threats while protecting human health and identifying sensitive aquatic native/endangered species and habitats within the Nation’s lakes and waterways. There will continue to be emerging threats to human health and natural resources; the NowCast tool is equipped to be the powerful early detection system for these novel threats.

Primary Contacts: Amie Brady, amgbrady@usgs.gov; Amanda Bell, ahbell@usgs.gov

Following a recent study estimating greater than 3 billion birds lost since 1970, there has been a call for the scientific community to collaborate broadly across geographic locations and jurisdictions to identify the drivers of declines in bird populations throughout their range, and to forecast changes in population size and distribution, given uncertainty in future environmental and management scenarios. The North American Breeding Bird Survey (BBS) is the primary source of the population change information that went into the 3 billion birds assessment, and the BBS is also a primary dataset for many other modeling activities. New opportunities exist for developing integrated models for BBS and eBird, as well as other datasets (that is, Christmas Bird Counts, Alaska Landbird Monitoring Surveys, Integrated Monitoring in Bird Conservation Regions) that can (a) enhance the value of the BBS for estimation of population change (it's primary historical use); (b) expand inference from the BBS to allow for estimation of change on and off roadsides; (c) allow for estimation of total abundance; and (d) allow for development of spatially-explicit models that directly connect change in land use and management to changes in bird populations. Integrated models that allow for spatially-explicit integration of eBird and other datasets collected with traditional survey designs provide a critical step forward for meeting the spatial information needs for bird conservation and management over the next decade.

The bird conservation community has devoted a huge amount of effort into using BBS and other datasets to inform management at local scales via spatial models. The “Avian Knowledge Network” is one example of this, and multiple modeling efforts that Partners in Flight and the North American Bird Conservation Initiative have sponsored, are also part of that effort. These efforts have been limited in scale and applicability by available data, in that BBS has issues that limit it as an omnibus spatial dataset (for example, roadside frame). While eBird, on the other hand, has recently provided spatially-explicit maps, but they are descriptive rather than providing models that facilitate management decisions by producing predictions of management actions. We propose extending the geographic and operational scope of a recently developed integrated modeling framework to leverage eBird and Breeding Bird Survey (BBS) data, thereby contributing to the BBS mission of remaining a leading source of data for bird management and conservation, supporting management and regulatory decision making of external partners (USFWS, Canadian Wildlife Service, National Park Service, Forest Service, and Bureau of Land Management, Mexican National Commission for the Knowledge and Use of Biodiversity), and strengthening the ties between government (USGS, FWS, State agencies) and nongovernment (that is, Cornell Lab of Ornithology, National Audubon Society, Partners in Flight) collaborators. This framework will expand the scope of taxa and geographic locations (for example, North Atlantic-Appalachian and nationally) for which we can make reliable inference and improve management decision making with more precise estimates of population trends for species of management concern. The integrated model can be used to support annual decision making by state and federal agencies, and to forecast changes in bird populations over longer time scales in response to climate and land-use change. The science and products developed from this effort align with the priorities of DOI (Create a conservation stewardship legacy second only to Teddy Roosevelt), the USGS Ecosystems Mission Area (species at risk; migratory birds; and, modeling and forecasting species distribution, abundance, and activity), and the North American Breeding Bird Survey program (Advance research on model development, including integrating complementary data, to better meet stakeholder needs; improve availability, utility, and visual appeal of analytical products; and, develop stronger partnerships with other organizations to enhance the mission of the BBS).

We recently developed an integrated population model, in collaboration with the Cornell Lab or Ornithology and FWS, using eBird and traditional aerial surveys to estimate the 2020 breeding population size of mottled ducks on the western Gulf Coast (fig. 5.6). The COVID-19 pandemic disrupted aerial surveys, however, FWS still had a critical need for population size estimates to make regulatory decisions. Our estimates are being used by the Service Regulations Committee to inform the current and future management of this species. This Use Case will improve the decision-making process for external cooperators (FWS, State agencies, BLM, USFS) by a) increasing the geographic and operational scope of recently developed integrated population models, b) improving precision of bird population estimates, and c) forecasting at multiple temporal scales relevant to the Department of Interior. This framework will be used to conduct population assessments at multiple spatial scales (that is, regional to continental), and for priority species identified by external partners. For example, USFWS has identified waterbirds as a guild that has typically received less research attention, yet has many species experiencing large declines in population size throughout their range. Forecasts on an annual scale will support management of species regulated by USFWS, and forecasts made on decadal scales will be relevant to understanding the potential impact of climate and land-use change on species of concern.

The Use Case relies on the extensive historical data provided by the BBS program as well as more recent community science data provided by the eBird program, and auxiliary data which drive bird population and distribution dynamics such as land-use change data from MODIS and NLCD. We bring these data resources together using modern methods of integrated modeling, which control for specific sources of bias and error in each of the contributing programs. This project will rely on existing USGS expertise in model development, data integration, and ecological forecasting. Scientists with USGS will provide expertise in using BBS data and external cooperators with the Cornell Lab of Ornithology will support the use of eBird. Guided by preliminary model development, this effort will extend the operational and geographic scope of integrated population models for a) conducting population assessments at regional and continental scales to guide annual decision making by USFWS, b) forecasting the impact of changing climate and land use on bird populations of conservation concern, and c) filling critical knowledge gaps for species that are of concern but have been relatively under-studied. This effort also has the potential to support and advance partnerships with other government agencies (that is, NPS, USFS, and BLM. For example, the NPS Inventory and Monitoring Program conducts annual bird population monitoring. Integrating these data with eBird and BBS will support population assessments and regulatory decision making within national parks.

USGS currently performs annual population assessments of migratory bird populations from the North American BBS program and the technical capacity to initiate the Use Case exists at Leetown/Patuxent. A number of on-going collaborative efforts between USGS and FWS provide annual assessments of species managed for take and this partnership is a key element of the proposed Use Case. Moreover, by advancing collaborations with the Cornell Lab of Ornithology we have established a multi-organizational team of experts in bird conservation and data science laying the technical groundwork and possessing the expertise to advance delivery of the Use Case. Key to initiation of the project is bringing together the data resources of the North American BBS program, maintained by the USGS, and those of the eBird program operated by the Cornell Lab of Ornithology, which has been demonstrated in one ongoing partnership between USGS, FWS and the Cornell Lab of Ornithology. Operational implementation and sustainability of the initiative will require additional staff resources in the form of one or more data scientists on staff associated with the BBS program.

For more than 50 years, the North American Breeding Bird Survey has helped keep common birds common by providing breeding population data for more than 500 species and serving as a quantitative cornerstone for avian conservation and management in the U.S. and Canada. Integrated models, that allow for spatially-explicit integration of the extensive eBird dataset and other datasets based on traditional survey designs, provide a critical step forward for USGS to continue meeting the spatial information needs for North American bird conservation and management. There will be an increasing need for short term predictions and long-term forecasts of bird population status by the bird conservation community as bird populations are increasingly affected by climate and landuse change. EarthMAP represents a unique opportunity as it provides a framework for integration of USGS land, water, and geological mapping products with biological datasets. Strong partnerships established between the entities (that is, USGS BBS, USGS Science Analytics and Synthesis, Cornell Lab of Ornithology) maintaining and curating data, developing modeling frameworks and presenting output to decision-makers and the public will ensure the sustainability of this project long-term. A successful project will be measured by the use of population assessments by external cooperators (for example, FWS, USFS, NPS, BLM, PIF [Partners in Flight], NABCI [North American Bird Conservation Initiative], and state wildlife agencies) in management-decision making and filling critical knowledge gaps for species of concern.

Primary Contact: Andy Royle, aroyle@usgs.gov; Paige Howell, phowell@usgs.gov

Long Island Sound provides social, recreational, and commercial benefits to millions of residents in Connecticut, New York, and surrounding States. However, four centuries of settlement and industrialization in the Sound’s watershed have led to widespread low dissolved oxygen, toxic contamination, pathogens, and floating debris. Seventy percent of all freshwater entering the Sound is from the Connecticut River, New England’s largest river system (fig. 5.7). The Connecticut River drains 11,250 mi² in five States and Quebec and flows 410 miles through forests, rich agricultural lands, and urban centers. The river’s watershed provides important habitat for a diversity of fish, wildlife, and plants, and is home to several federally threatened and endangered species including the shortnose sturgeon, piping plover, and dwarf wedgemussel. Nutrient loading (primarily nitrogen) from point and nonpoint sources within the Connecticut River Basin has been a priority environmental concern for Connecticut, Massachusetts, New York, and Long Island Sound stakeholders for many years. In response, the USGS New England Water Science Center, with collaborators at the USGS New York Water Science Center, Woods Hole Coastal and Marine Science Center, Leetown Science Center, and the Land Change Science Program, has been engaged in monitoring, modeling, and understanding of watershed processes that influence water quality and ecosystem services within the basin and forecasting environmental outcomes from potential land-use changes and pollutant-management actions.

The Center operates 65 streamgages, 19 temperature gages, and 17 water-quality stations in the basin. This data network provides critical real-time information on nutrient conditions in the watershed and nutrient delivery to the Sound. The Center also conducts nested environmental assessments and modeling studies that range from watershed-scale field research at the Sleepers River Watershed in Vermont to regional-scale modeling of nitrogen, phosphorus, and carbon throughout the basin. These monitoring, analysis, and modeling activities are done in partnership with the Connecticut Department of Energy and Environmental Protection (CTDEEP), Massachusetts Department of Environmental Protection, U.S. Environmental Protection Agency, the Long Island Sound Study, and many others. The Center has a robust program of stakeholder engagement in the basin and continually works to understand and address stakeholder needs.

In collaboration with its several partners, the New England Water Science Center will develop an integrated monitoring, analysis, and modeling framework that advances the understanding of nutrient sources, watershed transport processes, and trends within the Connecticut River Basin and the predictive capability of nutrient loading to Long Island Sound by explicitly incorporating dynamic (seasonal) transport processes and nonpoint-source loading of nutrients from groundwater into the modeling and analysis framework.

Nutrient management in the Connecticut River Basin has been focused on reducing point sources of contamination to rivers within the watershed yet monitoring at USGS streamgaging sites indicates that nonpoint sources are a substantial part of the total nutrient load to the river. Although nutrient loading from groundwater discharge has been shown to be an important, long-term nonpoint source to many freshwater and marine ecosystems, groundwater contributions to the Connecticut River have not been quantified.

A fundamental goal of EarthMAP is to provide actionable intelligence of direct value to USGS partners, stakeholders, and the public. To accomplish this goal, we will engage our collaborators and stakeholders early in the project-development process, as well as throughout the full life of the project, to ensure that the data, scientific information, and predictions that result from this EarthMAP initiative meet their resource-management needs. The monitoring and modeling framework developed as part of this initiative will provide the basis for improved seasonal and long-term forecasts of land-use changes, climate change, and management actions that address both point and nonpoint sources of nutrient loading. The knowledge gained from this predictive capability will provide managers much needed guidance on how, where, and when to conduct mitigation measures aimed at improving water quality throughout the watershed, as well as information needed to understand and manage the impacts of nutrient loading to freshwater and marine wildlife, fisheries, vegetation, and dynamics of harmful algal blooms.

The Center’s EarthMAP initiative will consist of two key components that will support each other and provide feedbacks for overall system improvement: a streamflow and water-quality monitoring network and an integrated modeling framework.

Dynamic SPARROW models: Two sets of SPARROW models have been developed to aid in the understanding of the long-term sources, fate, and transport of nitrogen, phosphorus, and suspended sediment across the Connecticut River Basin (Moore and others, 2004; Ator, 2019). SPARROW—SPAtially Referenced Regression on Watershed attributes—is a hybrid statistical and mechanistic model for estimating the load of a target constituent moving through the landscape. SPARROW was initially designed to simulate long-term, steady-state conditions (Schwarz and others, 2006). Recently however, the SPARROW methodology has been extended to simulate dynamic (transient) conditions with an initial emphasis on seasonal time periods (Greg Schwarz and Richard Smith, USGS, written commun., 2020). The New England Water Science Center has contributed to the testing of the new dynamic SPARROW modeling approach by modifying the original, steady-state model developed by Moore and others (2004) to simulate nutrient transport and loads for 35 seasons extending from winter 2001 through summer 2009 (Richard Moore, USGS, written commun., 2020). The models of Moore and others (2004) and Ator (2019) will form the basis for new, dynamic SPARROW models developed for the basin.

Groundwater/surface-water modeling: The Center will use a cutout of the National Groundwater Model (NGWM; Zell and Sanford, 2020) to simulate transient groundwater flow, groundwater/surface-water interactions, and nitrogen loading from groundwater discharge to streams. The NGWM is an innovative, MODFLOW 6 based model of the shallow groundwater system for the contiguous United States. Detailed simulation of the basin’s groundwater system is needed to accurately represent transport of nonpoint sources of nitrogen throughout the basin, as well as the time lags that occur between the entry of nitrogen into the soil zone and discharge of groundwater nutrient loads to streams and coastal waters. Subsurface nitrogen transport will be simulated using methods developed by Center staff as part of an ongoing regional groundwater-modeling project along the north shore of the Sound (Janet Barclay and John Mullaney, Fall 2020 AGU abstract). Model-simulated groundwater discharge and nitrogen loads will be directly coupled with the dynamic SPARROW models through a consistent surface-water drainage network.

Landscape modeling: The Forecasting Scenarios of land use (FORE-SCE) model will be used to reconstruct historical landscapes in the basin and to simulate future landscape change. FORE-SCE is designed to leverage USGS land-change monitoring data using information from the Earth Resources Observation and Science (EROS) National Land Cover Database (NLCD) and Land Change Monitoring, Assessment, and Projection (LCMAP) projects (Sohl and others, 2019). LCMAP uses a national-scale, high-resolution (30-meter) parcel-based modeling framework to simulate past, present, and future landscape change.

The Center will work closely with the USGS Water Resources Mission Area (WMA) developers of the dynamic SPARROW software and NGWM to ensure mutually beneficial feedback among the development and application teams. Moreover, the Center will design the modeling framework in consultation and partnership with the WMA’s Integrated Water Prediction Program and biologists with the USGS Leetown Science Center (Ecosystems Mission Area). Because of the computational demands of the proposed integrated modeling system, it is anticipated that the system will need to be deployed on one of the USGS high-performance computer systems, such as the Denali system housed at the Earth Resources Observation and Science (EROS) Center in Sioux Falls, South Dakota.

Because nutrient enrichment of surface waters from both point and nonpoint sources is such a widespread environmental concern and because the SPARROW and MODFLOW modeling systems have been applied extensively across all geographies of the United States, it is anticipated that the integrated modeling system developed as part of this EarthMAP initiative will have substantial transferability (for example, basins contributing to the Great Lakes). Moreover, the long history and institutional knowledgebase of the SPARROW and MODFLOW modeling systems suggests that they will remain viable modeling environments within the USGS for the foreseeable future.

Primary Contact: Johnathan R. Bumgarner, jbumgarner@usgs.gov

Sediment is a major pollutant nationally, and in the Delaware River basin (DRB), affecting reservoir storage and degrading aquatic habitat, as well as a transport vector for delivering attached contaminants to downstream water supply and ecosystems. Sediment and sediment-associated constituents can contribute substantially to water-quality impairments and infrastructure problems that can lead to water-availability impairments. Infrastructure problems associated with sediment include reservoir sedimentation, drinking supply intakes, erosion of critical infrastructure, and navigation channel maintenance. Ecological problems associated with sediment include light attenuation, burial of channel substrate, reducing primary productivity, impairment of benthic macroinvertebrates, clogging of fish gills, altering oxygen demand, and effect on egg gestation. Identifying the sources of this sediment and quantifying its flux in streams is information needed by management agencies to alleviate this sediment problem.

In general, the two main sources of sediment are upland erosion sources (soil erosion from various land uses) and streambanks. If sediment-source results indicate that streambanks are an important source of sediment, modeling sediment transport and delivery must include this as a term in their models. Sediment-trend analysis should include identifying channel change over time by either human or natural causes. Unfortunately, many of the models used to estimate sediment flux do not include a robust streambank erosion process. In the Chesapeake Bay watershed, over a billion dollars is spent each year on restoration, and the efficiency and effectiveness of that spending directly follows from the availability of the models and best science to predict pollutant sources in the landscape.

The major stakeholders interested in managing sediment pollution include industrial, agricultural, and municipal, and commercial resources managers who are tasked with treating intake water or preventing sediment movement related to their particular water-uses. Natural-resource managers including agencies that managed habitat, wildlife, fishes, and ecosystem responses (that is, eutrophication) are particularly interested in understanding the role of sediment in impairments and identifying the most effective management strategies and locations for reducing sediment pollution. To that end, the Sediment Use Case expects involvement of the USGS Water Resources Mission Area (WMA), USGS Ecosystems Mission Area, the USGS Integrated Water Availability and Assessments (IWAAs) program, the Integrated Water Prediction (IWP) program, and the Next Generation Water Observing System (NGWOS) program, all having interest in understanding sediment pollution for Integrated Water Science in the Delaware River Basin. The NGWOS program has invested resources during Federal Fiscal Year’s 2020–22 to pilot improved monitoring, rapid development of site-specific continuous models of sediment concentration, and landscape & channel sourcing of sediment in the DRB. In addition, funding from Ecosystems Mission Area, the Penn Foundation, and Smithsonian Institution is supporting an effort to improve modelling of floodplain and channel sources of sediment using the Floodplain and Channel Evaluation Tool (FACET, a toolkit to automate stream geomorphometry estimation) and USGS Chesapeake and Delaware Floodplain Network (CDFN, a measurement and modeling effort to predict streambank erosion and floodplain deposition).

Through the current work of NGWOS, Sediment Use Case envisions NGWOS-IWAA-IWP working together to monitor, assess, improve, and operationalize sediment prediction models over a large geographic extent (such as the DRB, the NAAR[the North Atlantic-Appalachian Region], or even larger regions of the U.S. depending on the extent of data collection). If the FACET-CDFN model can be improved in these ways, with closed sediment budgets, refined sediment sources, ancillary measurements of nutrients, bacteria and other sediment associated contaminates then FACET-CDFN can serve to feed and improve the accuracy, precision, and spatial extent of SPARROW models, upland sediment models, and machine learning tools under development in IWP. These goals align with DOI Mission Area goals and related USGS goals for conserving land, water, species, and habitat; managing DOI water storage and delivery; and informing land-use planning processes for public use and access with decision support tools. The goals of this Sediment Use Case also align with USGS WMA goals to provide data to society about the quality and quantity of water, improved understanding of processes (sediment transport) that determine water availability and predict changes in the quality or quantity of water related to external factors.

The Sediment Use Case that we are proposing would lead to improvements of sediment flux estimates within streams (erosion and deposition) and improved understanding of sediment sources in the DRB. The results of this study will meet WMA’s objectives of understating and predicting sediment flux from source to sink. Currently, our team is engaged with the Delaware River Basin Commission (DRBC), the Chester County Water Resources Authority (CCWRA), the USGS Office of the Delaware River Master (ODRM), the USGS WMA, the USGS Integrated Water Science team for the Delaware River Basin (NGWOS, IWAAs, IWP), four USGS Water Science Centers (Pennsylvania [PA], New Jersey [NJ], New York [NY], South Atlantic [SA]), Stroud Water Resources Institute, the Penn Foundation, and several academics from the University of Delaware and the University of Minnesota. We are currently working with EPA to incorporate our predictions of streambank and floodplain flux of sediment, N, and P into their next-generation regulatory model for the Chesapeake. Using improved reach-scale flux predictions and sediment source information, managers will be able to better predict problem areas, direct best management practices, more effectively treat water for public consumption, and aid in the restoration of ecological impairments.

Currently, NGWOS is assessing the rapid development of fluvial suspended sediment surrogate models using continuous turbidity or acoustic backscatter measurements. By deploying and selectively sampling based on streamflow statistics and turbidity/backscatter measurements robust predictions of suspended sediment concentration can be developed in less than 18 months. These predictions are on a 15-minute or shorter timestep and lead to highly accurate computations of hourly, daily, monthly and annual flux. These flux estimates can be used to greatly improve other models which attempt to estimate sediment in un-monitored areas.

Sediment fingerprinting is a method that compares the radiochemistry of upland, flood-plain and channel sediment sources to suspended sediment in streams using radionuclides common in rainfall, allows discrimination of the likely source (topsoil, streambanks, channel bed sediment). This radiochemistry can also provide a relative age of the sediment. If sediment is relatively young (that is, less than 5 years) then monitoring programs may show a decrease in sediment related to these management actions. If sediment is older (greater than 10 years), it may take longer until we see a reduction in sediment from these older sources. Combining sediment source analysis with age dating, provides managers and modelers with data need to control and understand sediment flux.

The proposed statistical model will build upon the CDFN model (https://www2.usgs.gov/water/southatlantic/projects/floodplains/) developed by Greg Noe and others using monitoring sites in the DRB and FACET model. Measurements of dendrochronology, field surveying, and sediment physio-chemistry were used to calculate changes in floodplain deposition and streambank erosion over time to create a quasi-sediment budget for each site. FACET incorporates high resolution airborne LiDAR to estimate channel and floodplain morphology, and along with characteristics of the upstream drainage area, was used to extrapolate the monitoring station results to the entire DRB. Statistical analysis including random forest regression (a form of machine learning) was used to develop statistical models of sediment flux with predictions of floodplain and streambank flux and geomorphic change for each NHDPlusV2 reach. By comparing floodplain and streambank fluxes (stream valley sediment fluxes) to SPARROW prediction of suspended sediment export from the catchment, a residual (other) sediment source was calculated using mass balance. The size of the modeled streambank vs. residual sediment source for any given reach estimates the contribution of channel vs. uplands to sediment transport. By expanding to monitoring stations across the DRB (or other IWS basins), the spatial resolution and accuracy of FACET-CDFN will be improved in new geographic areas. New reach-scale flux estimates could be used to improve SPARROW models by providing a new prediction variable (FACET flux).

This Sediment Use Case in the DRB will develop a scalable approach that can be implemented anywhere where appropriate fluvial sediment and sediment sourcing data is being collected. IWS basins (DRB, Upper Colorado, Illinois, and future basins) offer the opportunity to test not only the science, but the operation of NGWOS-IWAAs-IWP coordination. Scientifically, rapid sediment surrogate models from continuous monitoring, sediment fingerprinting and age-dating, FACET, random forest, and SPARROW models will improve the accuracy, timeliness, and spatial extent of sediment flux estimates and sources in unmonitored areas across the country. This sediment information could then be used directly by resource managers or be further incorporated into regional or national models to improve estimates of water availability.

Currently, the USGS has the technical capacity, interest, appropriate structure (NGWOS-IWAAs-IWP), and existing models and tools to operationalize a sediment flux and source workflow from a pilot in the DRB. Once the pilot in the DRB is complete, the developed template can be used in each other IWS basin across the country by gradually improving and standardizing the linkages between observation, assessment, and prediction of sediment flux, sources, and age. With advancements in real-time monitoring, improved workflows for sediment surrogate flux estimation, assessment of sources, geomorphic changes, high resolution LiDAR coverage, dendrochronology, FACET, and machine learning we have the tools to predict sediment flux more accurately and partition the relative sources (upland, floodplain, in-channel; ag, urban, storm water, etc.). However, the DRB Sediment pilot is currently under-resourced. We need resources from the IWAAs and IWP programs and more resources from the NGWOS in our pilot study for complete data collection, data assessments, prediction validation tests, and improvements to existing predictions. Effective delivery of this information could be done as a static data service, or more usefully, as a national web service that could deliver estimates of sediment sources, and real-time (at a daily or lower) estimates of sediment flux and apportionment of sediment sources on an annual or lower time-step (if applied to active sediment data stations). Data archives, sediment flux station site-specific models, and FACET meta data can all be served in a traceable format to the public from this same web service. By establishing a network of long-term monitoring locations through the NGWOS program, where continuous flux is estimated, routine drone LiDAR surveys and geomorphic measurements are conducted, and spatial variability is measured with synoptic observations we can create a data pipeline to feed assessment and prediction of sediment, at reach-scale, across the country. Data would be processed by the IWAAs and information products specific to local and regional needs can be delivered to local stakeholders, and IWP can establish computation pipelines involving FACET-CDFN, random forest, and SPARROW to provide highly accurate predictions of sediment & co-contaminant flux and sediment sources.

By generating this template and applying it to existing and new IWS, in less than 10 years we could have vastly improved sediment flux estimations, identified sediment sources for resource managers, and improved on the predictive capability of national, regional, and local water models. The coordination of resources across USGS Water Resources Mission Area programs, along with Ecosystems Mission Area expertise, and the scaling of pilot projects over time—potentially to operational scales in new IWS basins—would help to ensure the sustainability of the Sediment Use Case. Once the workflow, data requirements, appropriate assessments, and data flow for prediction are established, sustainability would require data scientists and IT infrastructure to serve information at various spatial and temporal scales. This technique should be largely adaptable to different IWS basins, and also to interested science centers to support stakeholder needs. Potential problems could exist with computational power for proper scalability but should be feasible with existing cloud infrastructure. Model predictions should be approved at the work-flow level, and reproducible data pipelines should be used to expedite data delivery in a timely manner. Metrics of success should be evaluated on references to our data and hits on our data services.

Primary Contact: Joe Duris, jwduris@usgs.gov

Similar to the Chesapeake Bay, research in the Delaware River Basin has focused on improving water quality with limited research on living resources and ecosystem science. This limited research has focused on the population trends of key species, such as Atlantic Sturgeon, white perch, striped bass, weakfish, American eel, American shad, brook trout, blue crab, horseshoe crab, eastern oyster, freshwater mussels, and other macroinvertebrates (https://www.state.nj.us/drbc/library/documents/SOTBreport_july2019.pdf). Understanding the response of living resources and their habitats to management actions and ecosystem change is the next stage in the scientific priorities identified by partners for the Chesapeake and Delaware watersheds. The goal of this project is to create an USGS interdisciplinary science team focused on conducting integrated science to inform the management of recreational and commercially important fish species, waterbirds, benthic macroinvertebrates, other aquatic and semi-aquatic species, pollinators, and their habitats. This team would establish an integrated research framework similar to Chesapeake Bay to provide a multidisciplinary approach to the management of the Delaware River Basin. Essentially what are the drivers of the ecosystem, how do the stressors impact those drivers, what are the cumulative impacts of these drivers and stressors on the habitat quality and availability on the fish, wildlife, and people that use the watershed. Ultimately, providing recommendations to the Delaware River Basin Restoration Program and other stakeholders on the conservation, management, and restoration of these resources/habitats to sustain the living resources in the ecosystem. In addition, identifying gaps in our knowledge that would be needed to better inform the management actions for living resources. Ultimately, with similar research efforts being conducted in both watersheds (Chesapeake and Delaware), management decisions and restoration recommendations could be made on a broader regional scale.

Stakeholders are essential to this project and examples of local, state and federal stakeholders include, but are not limited to, Delaware River Basin Commission, Partnership for the Delaware Estuary (PDE), Delaware Department of Natural Resources and Environmental Control (DNREC), Department of Fish and Wildlife (DE DFW), United States Fish and Wildlife Service (USFWS), Delaware Water Resources Center, Institute for Public Administration, University of Delaware, USGS New Jersey [NJ] Water Science Center, Maryland-Delaware-D.C. [MD-DE-DC] Water Science Center, Lower Mississippi-Gulf [Lower MS Gulf] Water Science Center, and Woods Hole Coastal Marine Science Center, Atlantic States Marine Fisheries Commission, Migratory Bird Initiative and Migratory Connectivity Project: Smithsonian Migratory Bird Center, Georgetown University, National Audubon Society, and Cornell Lab of Ornithology. Many of these partners are already actively collaborating on research projects for the Chesapeake Bay. Table 5.1 highlights the alignment of the Delaware Basin Plan goals, with USGS priorities and Mission Areas, DOI Priorities, and the Chesapeake Watershed Agreement goals.

One activity identified by the Delaware River Basin Restoration Program is to increase scientific capacity to support planning, monitoring, and research activities necessary to carry out coordinated restoration and conservation activities in the basin (https://www.fws.gov/northeast/delawareriver/partnership/index.html). The goal of this Use Case activity is to increase coordination of scientific resources among partners in the basin, supporting project planning, decision making, and measuring improvements to natural resources after project implementation, mirroring ongoing efforts in the Chesapeake Bay. Over the last decade, PWRC/LSC researchers have made considerable progress implementing interdisciplinary integrated science in the Chesapeake Bay. With the additional research capabilities of this USGS science team, we could provide cutting-edge modeling efforts and decision support tools to the Delaware River Basin Restoration Program and other key stakeholders. For example, using structured decision making to create spatial value models and visualizations to identify important landscapes for conservation and (or) optimizing salt marsh management.

PWRC/LSC researchers have been working in the Delaware River Basin for many years now and have compiled quite an extensive multidisciplinary dataset on, but not limited to, amphibian distribution, fish community ecology, habitat suitability, decision support tools, listed species ecology and management, fish health data, terrestrial/freshwater organismal records, harmful algal bloom data, avian population abundance and distribution data, influences of river flow and temperature on fish communities, freshwater mussel data, pollinator distributions, avian disease data and linkages to poultry industry, avian health data, invasive species, and water contaminant data. Much of this research is in collaboration with other science centers through the USGS Chesapeake Bay Science team (USGS NJ Water Science Center, MD-DE-DC Water Science Center, Lower MS Gulf Water Science Center, and Woods Hole Coastal Marine Science Center). New approaches are needed for this project to adapt the Chesapeake Bay integrated modeling framework for the DRB. For example, the NHD-HR Hydrolink proposed Use Case will develop the capability to link biological data to the medium resolution National Hydrography Dataset Plus (NHDPlus, 1:100K) and newly developed National Hydrography Dataset Plus High Resolution (NHDPlus HR, 1:24K). This step is necessary in order to summarize landscape variables in catchments surrounding stream reaches for aquatic habitat assessment and align with efforts to build out the National Hydrography Infrastructure (NHI) to provide information at the highest resolution, decision-relevant scales that stakeholders have requested.

There are two potential avenues of implementation to achieve this project goal depending on the resources available: 1) Leverage off the existing PWRC/LSC USGS Chesapeake Bay research activities that could be adapted or expanded into the Delaware River Basin with minimal additional resources or 2) develop a new USGS Delaware River Basin Science Team. This USGS Science Team could maintain the data storage, access and delivery of datasets and products to internal and external partners via a project site that could be linked to the Delaware River Basin Restoration Program website to encourage stakeholder engagement.

Managers and decision-makers will continue to need scientific data and modeling to make informed decisions for managing human and natural communities in an ever-changing climate. The USGS Delaware River Basin Science Team would continually evolve, similarly to the Chesapeake Bay Science Team, to ensure USGS fulfills the needs of the stakeholders as they adapt to better manage the system. For instance, the structured decision-making models can evolve as the decisions needed change and the associated framework is scalable so the resolution of the output can change based on the scale of the input data.

Primary Contact: Tom O’Connell, toconnell@usgs.gov; Alicia M. Berlin, aberlin@usgs.gov

The National Fish Habitat Partnership (NFHP) is a national effort that brings together partner organizations from Federal and State agencies and NGOs with the purpose of identifying, enhancing, and restoring fish habitats. USGS is an active partner in NFHP with leadership roles served by Science, Analytics and Synthesis (SAS) and Ecosystem Mission Area (EMA). A major project of this partnership was an assessment of the status of fish habitats in the US, last conducted in 2015, using nationally consistent environmental data and fish community observations from partnering agencies. While an incredible resource for national and regional comparisons, this assessment was necessarily limited to data available nationwide, and was limited to the regional-scale (1:100,000). However, stakeholders and natural resource managers have consistently identified the critical need for information at finer spatial scales (for example, 1:24,000) in order to more effectively design and implement habitat assessments and associated conservation, mitigation, and restoration activities and better inform decisions.

However, current efforts to integrate high resolution data (for example, the National Hydrography Dataset NHDPlus High Resolution [NHDPlus HR] Beta version at 1:24,000) into decision making have revealed issues and limitations of existing information that must be addressed before these efforts can move forward. These include resolving issues with catchment boundary delineations which hinder summarizing landscape stressors associated with stream reaches and connecting to existing biological data. For example, current data collected by partner agencies from 1:24K mapped stream features have no equivalent representation at the coarser spatial scale losing up to 10 percent of fish observations that cannot be used resulting in loss of data integrity, increased uncertainty, and limiting decision relevance.

A pilot project that includes scientists with Core Science Systems, Ecosystem Mission Area, and USGS Water Resources Mission Area is addressing the technical challenges connecting NHDPlus HR data to biological data to improve modeling landscape-aquatic habitat interactions at management relevant scales in the Chesapeake Bay Watershed (mapped in USGS NHD HydroRegion 02). Correcting issues with the NHDPlus HR will allow for more effective and accurate monitoring, assessment, and prediction—key elements of the EarthMAP vision.

This Use Case will expand this effort to align geospatial and biological data for fish habitat assessments in additional river systems in NHD HydroRegion 02, particularly the Delaware River Basin (DRB), which has been designated a national treasure of great cultural, environmental, ecological, and economic importance. The capability to link biological data to the medium resolution National Hydrography Dataset Plus (NHDPlus, 1:100K) and newly developed National Hydrography Dataset Plus High Resolution (NHDPlus HR, 1:24K) is a necessary step in order to summarize landscape variables in catchments surrounding stream reaches for aquatic habitat assessment. and align with efforts to build out the National Hydrography Infrastructure (NHI). Achieving this capability will help USGS provide information at the highest resolution, decision-relevant scales that stakeholders have requested.

The DRB contains 23,700 linear miles of streams and rivers and expansive groundwater systems supporting nearly 180 species of fish and wildlife that are considered special status species in the Basin due to habitat loss and degradation, particularly sturgeon, Eastern oyster, horseshoe crabs, and Red Knots. In addition, the DRB provides habitat for over 200 resident and migrant fish species, includes significant recreational and commercial fisheries including Eastern brook trout, American and hickory shad, river herring, and American eel. Fishery managers are no longer able to sustain our nations fish, fisheries and aquatic habitats solely by regulating harvest, but need to better understand how the connections between landscape changes and aquatic habitats affect ecologically and socio-economical important fish and aquatic resources. USGS EMA is currently conducting habitat, population, and stock assessment research on several of these species in support of partners including the Atlantic States Marine Fisheries Commission (ASMFC) and NPS.

The approximately 135 major stakeholders in the DRB have identified a suite of conservation and restoration needs that require scientific information at fine spatial scales in order to achieve their goals and outcomes. These include:

The goals of the Delaware River Basin Restoration Partnership Program Framework include 1) sustaining and enhancing fish and wildlife habitat restoration and conservation activities and 2) increasing coordination of scientific resources among partners in the basin to support decision making for project planning and measure improvements to natural resources after project implementation. Their Focus Areas for Conservation Prioritization and Design include:

Results of the this Use Case project will support those conservation and restoration goals and priorities. This methodology, including alignment of data to the National Hydrography Infrastructure, can serve as the framework for national efforts to enrich geospatial datasets and to provide the basis for enhanced information discovery and management relevant data delivery to USGS stakeholders, such as through the EarthMAP platform.

The USGS Core Science Systems Mission Area National Geospatial Program (NGP) is developing the National Hydrography Infrastructure (NHI) to support the referencing and discovery of water-related information, which in turn, will help support hydrography-based modeling efforts. The hydrographic framework for the NHI is the NHDPlus HR, which represents stream networks at 1:24K scale and provides intelligent stream flow routing and landscape attributes similar to the highly successful medium resolution NHDPlus used in numerous science and assessment activities, including the NFHP. Currently in a Beta release, the NHDPlus HR will serve as a critical framework for integrating a wide variety of USGS and partner agencies’ observational information to fine-scale stream reaches as well as for summarizing environmental factor data in local and upstream drainage areas and catchments.

The Delaware River watershed is being used as a pilot for USGS integrated science activities, including WMA’s Integrated Water Prediction and Next Generation Water Observing Systems implementation. Key to translating these efforts to National implementation are framework geospatial datasets, including the NHD Plus HR. In addition to USGS WMA data on water quality and quantity, USGS EMA (Leetown Science Center) scientists have previously conducted extensive eastern brook trout, freshwater mussel, stream temperature, stream habitat, and submerged aquatic vegetation studies in the basin. Additionally, they have developed bathymetric and topographic LiDAR datasets for the mainstem Delaware River that can provide the basis for 2D hydrodynamic modeling of river flow velocity. These data can be used to further enrich the NHD HR framework to add value to integrated modeling activities. The stakeholders and partners noted under 2 have extensive biological and ecological data that is available for this effort. This Use Case will leverage all of these existing data sources.

This Use Case will build on the existing multidisciplinary, integrated Chesapeake Bay science team to expand efforts in the DRB. The Chesapeake Bay Program already supports extensive data frameworks to store and deliver datasets and products to internal and external partners. USGS is already working with most of the major stakeholders identified under 2 and will further those collaborations as part of this project.

Developing the capabilities to update the NHDPlus HR data such that it is useable for modeling landscape-aquatic habitat interactions at management relevant scales will support more effective and accurate monitoring, assessment, and prediction. It is anticipated that the methodologies developed in this project will be transferable to other areas nationwide.

Primary Contact: Tom O’Connell, toconnell@usgs.gov; John Young, jyoung@usgs.gov

Population growth and increased standards of living have accelerated societal demand for mineral resources. Technological advances and the clean energy transformation will require increased quantities of minerals critical for their function. Decisions affecting the future supply of mineral resources must balance sustainability, environmental impacts, national security, and economic factors. The overarching question that this Use Case addresses is: where should mineral resources be extracted to 1) maximize economic efficiency, sustainability, and security of supply and 2) minimize hazards to human health and the environment and risks of supply disruption?

Stakeholders include State Geological Surveys, elected officials in Congress and State legislatures, decision makers at the Federal to local level, including such agencies as the Bureau of Land Management (BLM), the U.S. Forest Service (USFS), National Park Service (NPS), Department of Energy, Department of Defense, Department of Commerce, and the Environmental Protection Agency. Additionally, the information should be made publicly available for use by industry, nongovernmental organizations such as the World Bank, international organizations such as the United Nations International Resource Panel, and academic researchers. Furthermore, access to mineral resource information contributes to the Department of Interior goals of stewardship of natural resources and conservation.

Relevant stakeholder questions include: Where should mineral exploration activity be focused? What areas should be off-limits to mineral activity? Where do mine wastes pose risks to the environment and human health? Where in the United States could a specific mineral commodity be economically mined? As some of these questions require evaluating competing land uses, mineral resource information should be presented alongside other data in EarthMAP such as protected areas, biodiversity hotspots, environmentally sensitive areas, water resources, and agriculturally important areas.

Geospatial information on where minerals have been mined in the past, the location of wastes associated with their production and processing, and prospective areas for future mineral production is crucial for resource policy decisions, economic planning, land use management, and environmental regulation. The USGS Mineral Resources Program provides data, expertise, and analysis necessary for addressing this question. In the past, regulatory and land management agencies have sought input and information directly from the USGS to inform land planning decision making, as required by the Federal policy. As a result, the USGS has produced a body of work contained in publications, maps, and data releases over many decades. While the scale and scope of each science varies depending on the goals of the project, much of this information remains relevant and useful for guiding future work.

Stakeholders weigh the economic value provided by mineral resources against the risks and costs associated with mineral extraction. To make decisions, they must clearly understand the uncertainties associated with each layer of information.

The Federal Land Policy and Management Act (FLPMA) requires land management agencies (BLM, NPS, USFS) to consider mineral resources in their long-term land use planning. Past assessment work has focused on specific areas, such as Wilderness Study Areas, proposed for withdrawal and been published at a relevant scale on an as-needed basis. A specific example is the Sagebrush Mineral-Resource Assessment (SAMiRA) assessment of mineral resource potential of Greater Sage-Grouse habitat on BLM lands. This assessment delineated where disturbance through mineral extraction of a critical habitat could pose further stress to a threatened species. This assessment provides the fundamental scientific information to enable BLM to make wise decisions regarding land use in Greater Sage-Grouse habitat. Leveraging this published data by making assessment results more accessible to stakeholders will facilitate prioritization of future assessments by identification of crucial focus areas that need to be assessed.

Other stakeholders, such as State Geological Surveys, currently work with the USGS to determine focus areas for geological investigations funded under EarthMRI (Hammarstrom and others, 2020). These focus areas are selected because they encompass geologic deposits that have significant potential for important mineral occurrences. Identification of these focus areas will help prioritize where research funding should be allocated to maximize the ability to make informed decisions regarding mineral resource extraction and its impact on the environment and human health.

Existing foundational data have been published by the USGS in a variety of formats and are currently hosted by several different platforms. MRDATA (Mineral Resources Online Spatial Data) contains much of the spatial data in a form amenable to EarthMap. Permissive tracts from previous mineral resource assessments exist as maps in the USGS publications warehouse and National Geologic Map Database (NGMDB) or GIS files in ScienceBase. Active mines and abandoned mines may be spatially associated with mineral sites present in the Mineral Resources Data System (MRDS) and USMIN mineral deposit database (USMIN) databases, accessible through MRDATA and ScienceBase. Externally, information on abandoned mine lands is maintained by the Office of Surface Mining, Reclamation, and Enforcement (OSMRE), the EPA, BLM, and an interagency effort (https://www.abandonedmines.gov/) that includes the USGS.

There is limited information technology (IT) support to facilitate integrating science products hosted in Science Base and other repositories into EarthMap. Currently, science products are published and disseminated through USGS publications warehouse, ScienceBase, and in the case of mineral resource assessment maps, the National Geologic Map Database (NGMDB). The simplest starting point for integrating this information into a spatial platform is to link shapefiles of study areas to their original publications. A next step would require the digitization of legacy datasets and reattribution of digital polygon files for permissive tracts. This work has been carried out for a few study areas, but not on a national scale. Work initiated through the National Geological and Geophysical Data Preservation Program has resulted in the curation and preservation of data in dozens of States and has forged partnerships with State Geological Surveys.

New approaches to disseminating assessment results may need to be developed to address decision-maker and stakeholder expectations. Past assessment results are a mix of quantitative and qualitative results depending on the project or client. Application of economic filters in resource assessments will provide new data that will be of interest to a variety of potential partners in the mining sector.

Mineral Resource Assessment constitutes a core function of the USGS Mineral Resources Program. Expertise in this area has enabled the training of several early- and mid-career scientists with the capacity to carry out assessment work and contribute toward new results.

There is limited existing operational support for the Use Case. In the past, mineral resource assessment work has been funded by appropriations from Congress and reimbursable funding provided by partner agencies. The foundational-level input data utilized to conduct mineral resource assessments exists.

In the process of integrating multiple data sources supporting mineral resources assessment work, it would be critical to define whether the data in use are readily available in terms of their level of processing, relative quality, storage formats (as is the case of remote sensing, for example), and whether it can be easily brought up to satisfy the user needs. For example, the Mineral Resource Data System (MRDS) was recently updated with a “grade” of record qualities where a category of record completeness for each mineral deposit was estimated (https://www.sciencebase.gov/catalog/item/5d712a5ae4b0c4f70cfd4e54). This information allows users to better query data based on quality to ensure the data meet their project needs. Partnerships are being developed with State Geological Surveys through Earth MRI to incorporate more and better-quality data into MRDS. Similar efforts to carefully assess and present information on record quality will be necessary to increase the usability of the integrated mapping platform. Furthermore, the user should be able to identify and access the technical or scientific support necessary to improve on the existing datasets, such as specialized software licensees who can develop and deliver specific products. An example of this is the development of mineral alteration maps from remote sensing data. Such maps can be useful for recognizing particular mineral deposits, which would draw in partners from the mining sector, or can pinpoint potential natural hazards associated with volcanoes, which would benefit emergency management and response agencies, at both the national and State level.

The U.S. Geological Survey has, since its inception, provided expertise in the area of mineral resources of the United States and will continue to meet the needs of Federal government agencies, fulfilling its original mission defined by the Organic Act of 1879. A tiered approach to implementing the mineral resource component of EarthMAP is envisioned. At a minimum, the existing publications and geospatial data currently hosted on USGS platforms (Science Base, MRDATA, NGMDB, and Publications Warehouse) could be linked to or added to the EarthMAP infrastructure. The next phase of implementation may require digital conversion of legacy dataset, for example, the type of work that was conducted for the SAMiRA project.

The implementation of a mineral resources Use Case can exist at virtually any scale. Typically, assessments have been conducted at 1:100,000 to 1:1,000,000 scales. The presentation of this level of detail at a national scale is envisioned. A smaller-scale Use Case, for example, at the state or regional level, may represent a good starting point for implementation. Eventually mineral resource capability of EarthMap could be scaled-up to a national level. Successful implementation can be measured soliciting stakeholder feedback to gauge ease of use while maintaining an adequate level of complexity needed to address policy and regulatory missions.

Primary Contact: Graham Lederer, glederer@usgs.gov

The goal of this Use Case is to provide science information for enhancing the resilience of natural and built environments facing significant and increasing changes in current baseline conditions and disturbance events. Predictive science on resilience of coastal regions will inform adaptations to greatly reduce the costs in lives, property, and infrastructure of anticipated chronic and catastrophic disturbances.

The NAAR Urban/Coastal Resilience Initiative (UCRI) will integrate goals from all the current USGS Mission Areas (MA), the 4 Water Centers, several programs including the Coastal and Marine Hazards Program, the Science and Decisions Center, Core Science data and mapping programs, and several Ecosystems MA Programs. Collaborations with the Southeast and Great Lakes Regions are likely as the Initiative advances. Interested outside partners include the New York City (NYC) Mayor’s Office of Resiliency (MOR) and Department of Environmental Protection (DEP), Parks & Recreation (Parks), and Transportation (NYDOT); NOAA; FEMA; USACE; USFWS; and The Nature Conservancy (TNC).

Multiple issues will be addressed by the UCRI using a meta-project, multi-scale, interdisciplinary approach that reflects the EarthMAP goals, but for this Use Case we will describe one example: development of coupled models of compound flooding and its effects on developed and natural coastal landscapes. Natural resource and infrastructure managers and policy makers will be the primary stakeholders in this Use Case.

Specific stakeholder interest in this Use Case:

What NYC Boroughs need adaptation to avoid compound flood damage? Where do we need to evacuate for storms, and where do we need to retreat? Where could wetlands, living shorelines, or natural or nature-based infrastructure be established to protect or enhance valued resources?

Coupled models of the key processes controlling resiliency based on whole-system responses to directional change and episodic disturbance (groundwater, surface water, SLR, and storm surge; and the infrastructure, ecological, and socio-economic effects). Coupled-Ocean-Atmosphere-Wave-Sediment Transport Modeling System (COAWST), Hydrologic Engineering Center’s River Analysis System (HEC-RAS), and MODFLOW will be coupled to allow groundwater dynamics to be influenced by surface water and inundation from coastal waters to be influenced by groundwater processes. Socio-economic and ecological models will be linked to the water models and applied to infrastructure maps to map resources and communities at risk.

Groundwater monitoring arrays will be linked to the Surge, Wave, and Tide Hydrodynamics (SWaTH) network, tidal wetland SET stations, tide monitoring (NOAA and USGS) and surface water gaging, creating an array of continuous fixed stations, rapid deployment arrays, periodic surveys, and remote sensing for whole-system, multi-scale (time and space) tracking.

City planners have a pressing need for compound-flood modeling capabilities, particularly as they relate to coastal adaptation, stormwater infrastructure, and transportation planning. The limited existing models for cities are not yet available to smaller coastal towns.

This integrated modeling framework could be deployed to evaluate risks over a planning time horizon and to account for issues that will occur with SLR and changing climate. It could also be deployed in a more operational framework to understand short term risk of disturbance.

As environmental disturbance events increase in number and magnitude, and human populations grow in coastal cities and towns, urban infrastructure undergoes multiple stressors and, in many instances, the damaging impacts are further exacerbated by the mere fact that many are aging and decaying already. Additionally, with increased environmental disturbances and human population growth, natural landscapes are continuously being degraded and wildlife populations in or near urban areas are being displaced. Developing coherent, longer-term, location-specific resiliency for urban areas requires a science strategy that produces interdisciplinary and multi-scale information from all available sources.  The USGS, with its multi-scale, interdisciplinary capabilities in data collection, management, and synthesis, will use this Urban/Coastal Resilience Initiative (UCRI) to provide the integrative leadership needed for delivering regionally-focused, systems-level environmental science for timely informing of management and policy decisions.  Specific goals for improved decision support include anticipation of hazards and best practice recommendations for mitigation or managed retreat, early warning, and enhanced accuracy, availability, timeliness, and utility of USGS monitoring and interpretation. Close collaboration with stakeholders will establish these milestones. The improved modeling will allow robust assessments of coastal adaptation pathways for managed retreat, better anticipation of evacuation requirements, and improved planning for infrastructure operation, maintenance, and replacement.

The UCRI will applied state-of-the-art modeling and monitoring network design (see Delaware NGWOS) for each issue addressed. Machine learning models, artificial intelligence, and remotely sensed data collection will be used where they afford better or more cost-effective and efficient assessment of the processes and conditions affecting resource decisions. Hydrologic models will be linked with socio-economic maps and models, ecosystem vulnerability assessments, and natural, nature-based, or grey infrastructure disturbance mitigation strategies for assessments of whole-system responses to change. Collaborations with stakeholders throughout the model development will guide products toward supporting resource and infrastructure management and planning decisions.

Existing capabilities in the Northeast are extensive both within USGS and our Federal, State, municipal, academic, and NGO partners. Monitoring records are available for mapping, modeling, and trend detection. Examples include: NYCDEP water monitoring, NYWSC observation wells, streamgages, tide gages, and water-quality networks; and the SWaTH rapid deployment network. State-of-the-art models for the coupled model strategy include COAWST, SHERA, CoSMoS, and the Long Island Regional groundwater and solute-transport models. Established collaborations with coastal modeling efforts by other agencies and academia will be enable further integration of modeling capabilities and potential products for national transferability (NOAA Coastal Coupling Community of Practice (CCCoP), National Science Foundation Coasts and People Program (CoPe). The initiative has access to significant computational and data management capabilities in the Water and Coastal and Marine science centers of the NAAR to initiate the modeling. The goal over time is to fully integrate the IT requirements for UCRI into the services being designed by the national EarthMAP initiative.

The Use Case will develop several new capabilities and approaches, incorporating new technology and better integrated strategies for applying existing technology when it results in more efficient or robust measurement and interpretation. This initiative will rely extensively on the leveraging of existing projects and capabilities within the NAAR. Potential for new funding from stakeholders is significant, but the issues being addressed will require a funding strategy of leveraging existing funds with supplemental financing to be cost effective. This strategy also leverages data collected by prior investigations, decreasing the time required to detect trends and accelerating delivery of actionable intelligence. Multiple existing datasets are available for many issues the UCRI will address, through USGS archives and partner databases. For example, detailed below-ground mapping of water infrastructure in New York City compiled by NYCDEP will allow a significant improvement in groundwater modeling to determine infiltration rates and saturation potential affecting compound flooding. Partnerships will depend on the issue being addressed, but the potential is significant because of a long history of successful USGS collaborations in the region. Partnerships for the compound-flood modeling in NYC include NYC MOR, DEP, Parks, and NYDOT; FEMA, NOAA, NYSDEC, and several academic institutions.

Primary Contacts: Peter S Murdoch, pmurdoch@usgs.gov; Chris Schubert, schubert@usgs.gov; Liv Herdman, lherdman@usgs.gov

Drought represents one of society’s grand challenges for the 21st century (Crausbay and others 2020). Impacts of drought are especially evident in the western United States, where precipitation deficits increasingly coincide with warmer temperatures, impacting human and ecological water availability. Managers seeking to address drought include those responsible for managing millions of acres of federal public lands in the western United States administered by the U.S. Forest Service, Bureau of Land Management, and National Park Service. These lands are critical for providing innumerable societal benefits and managers of them are in urgent need of reliable, accessible scientific information to inform difficult decisions ranging from water security to ecosystem management. To address these needs, the USGS has developed a coordinated and integrated drought science plan to address the issues of drought and build long-term resilience to drought at the local, regional, and national levels (Ostroff and others, 2017). Eco-Drought is the on-the-ground implementation of the USGS Integrated Drought Science Plan. Implementation of Eco-Drought has involved shared resources and collaboration across three USGS Mission Areas, including Water Resources, Ecosystems, and Core Science Systems. Work in the Use Case described herein is also part of 5 major new investments in the NOAA-NIDIS “Coping with Drought” program (National Oceanic and Atmospheric Administration).

Products from the USGS Eco-Drought effort are intended to address drought at scales and time periods that are salient for managers who face difficult on-the-ground decisions. Drought can manifest quickly (for example, “flash droughts”) or take place over much longer time frames to act more as a press disturbance that potentially leads to a fundamental ecological and potentially social transformation (Crausbay and others, 2020). Similarly, the spatial grain at which drought responses are considered can influence perceptions of impacts and appropriate management responses. Kovach and others (2019) provided the example of how USGS streamgages are often mis-aligned with ecological observations in drought-sensitive river networks. Addressing this alignment to decision makers needs is one of the central themes of the USGS Eco-Drought studies. In most of these studies, stakeholders have been directly engaged in providing feedback on relevant questions and often providing funds to leverage USGS investments. Tools that allow them to easily collect complementary information for tracking drought on the ground have also been developed (FLOwPERmanence [FLOwPER]; Jaeger and others, 2020). In short, stakeholders have been engaged to inform research direction and objectives, as well as to collect data in many cases. Regardless of the specific decision or resource addressed, this level of engagement is certain to ensure actionable outcomes (Dunham and others, 2018).

Existing scientific and technological capacity. Existing scientific and technological capacity is strong among the collaborators working in the Donner und Blitzen study, but additional capacity would likely be needed to support additional requirements of EarthMAP. Expertise on the scientific team presently engaged in this work includes aquatic and landscape ecology, remote sensing, Bayesian statistical modeling, hydrologic modeling, and modeling of soil moisture. Additional time and expertise to adapt emerging modeling platforms for prediction of hydrologic responses (for example, Read and others 2019), data processing and visualization, and time for stakeholder engagement (Delie and Biedenweg, 2019; Miller and others, 2020) is needed to ensure this effort is a success from data to decisions.

Potential for integration and existing datasets. This study includes most of the elements of other USGS projects, such as the Integrated Water Availability Assessments recently developed (Miller and others, 2020). Existing USGS and partner datasets are being assembled now for integration in the Eco-Drought and related studies ongoing in the Donner und Blitzen basin and could be integrated into EarthMAP. Partnerships are strong in the basin and would be significantly strengthened locally and regionally through this effort. In addition to the Eco-Drought work in the Donner und Blitzen, USGS is actively engaged in a long-term study of groundwater use and availability and persistence of springs within the broader extent of the Harney Basin (Steve Gingerich and Hank Johnson, personal communications, U.S. Geological Survey Oregon Water Science Center [ORWSC]). A recent study involving USGS (Forest and Rangeland Ecosystem Science Center [FRESC] and ORWSC) and USFWS (Pearson 2020) also provides information on flow-dependent persistence and restoration of wetlands in the Malheur National Wildlife Refuge that could be well-informed by implementation of EarthMAP. Finally, non-market economic valuation led by USGS (Lucas Bair, U.S. Geological Survey Southwest Biological Sciences Center, personal communication) could be integrated into this work, providing a full complement of tools for evaluating this case as a social-ecological system (Dunham and others, 2018).

Existing and potential partnerships. The Donner und Blitzen River watershed is part of the larger internally draining Harney Basin, located in southeastern Oregon, USA. This watershed is a cooperative management area administered by BLM and includes Wilderness and Wilderness Study areas, as well as designated multiple use lands. The river itself is a designated Wild and Scenic River and Redband Trout Preserve. The river provides critical flows that sustain wetlands downstream in the USFWS Malheur National Wildlife Refuge (Pearson, 2020). In addition to federal stakeholders, a host of stakeholders working under the High Desert Partnership are addressing restoration of wetlands on Malheur National Wildlife Refuge, as well as future management actions to cooperatively manage limited water resources in the basin. In short, there is a high level of “stakeholder readiness” (Halofsky and others, 2018) in the basin. In the framework of EarthMAP, this Use Case is very “demand-driven”: it is highly certain that stakeholders would place high value on products that EarthMAP could deliver, and that these products would be integrated into decisions that influence outcomes for decades in the Use Case area.

We envision products that EarthMAP can deliver for this Use Case would parallel other developments providing novel visualizations for science applications, for example, the National Institute of Visible Human Project. In short, application of EarthMAP in this Use Case could provide stakeholders with an analog of this project: the “visible riverscape.” This product would allow stakeholders to view spatially and temporally variable meterological, hydrologic, and ecological conditions, as well as view the current social landscape, and to more readily envision scenarios and outcomes associated with alternative futures and management decisions. Technical capabilities to execute the Use Case exist but are not currently supported by existing funds. Additional resources would be required to further application of EarthMAP to leverage existing USGS and cooperator investments in this effort. In addition to the conduct of the science itself, additional support for time to fully engage stakeholders would represent an important investment. As mentioned previously a strong stakeholder establishment is present but allocating time to fully engage them is an important aspect of this work that is not currently supported. In addition to the existing team of scientists and support needed for their time, allocation of time or direct funding to support is needed to meet required technical elements (for example, AI/ML scientist, data scientist, and support for physically-based models) for EarthMAP. Determination of exact needs would require more detailed consultation on specific plans for implementation.

The long-term viability of the science and technologies of this effort will depend on the initial investment in developing products, as well as periodic investments needed to provide ground-based and remote sensing information to update EarthMAP outputs, as well as data management and delivery of model-based products. Current decisions in the study area are being considered over longer (greater than 20 yr) timeframes (for example, drought adaptation planning), as well as seasonal to annual adjustments in more immediate actions (for example, decisions regarding agricultural water uses, livestock grazing, fish and wildlife management). If resources allow we could also incorporate an effort to quantify metrics of performance and success that specifically address the stakeholder community (for example, through interviews conducted by a social scientist (Delie and Biedenweg, 2019). More conventional measures of effectiveness (for example, assessments of model performance or validation) will also be conducted, but we consider success as judged by the stakeholder community as the ultimate measure of effectiveness. We strongly believe that technical approaches developed and lessons learned from application of EarthMAP in this Use Case can be adapted to the broader objective of applying EarthMAP across diverse social, ecological, and hydrologic settings across the Nation. The Use Case of the Donner und Blitzen River represents a unique opportunity to integrate all these elements and advance the grand challenges of the 21st century that EarthMAP seeks to address.

More broadly, over the next 3 to 5 years, the Use Case of the Donner und Blitzen river can provide lessons and a means of adapting EarthMAP to other basins currently under investigation across the Nation under the USGS Eco-Drought study, as well as within the next 10 years adapting these cases to extend across the western United States where transformational droughts are acting to influence ecosystem processes and human well-being (Crausbay and others 2020). Other agencies are supporting collaborative work with USGS and partners now to move in this direction and application of EarthMAP in this Use Case would provide a powerful means of enhancing these collaborations, relationships, and societal outcomes for adapting to the grand challenge of drought in the 21st century.

Primary Contact: Jason Dunham, jdunham@usgs.gov

Interactions between different components of global change underlie may important ecological and economic impacts. For example, invasive species create fuel for fires whose intensity is increased by climate change, and newly intensive rainstorms fall on increasingly impervious land cover that intensifies flooding. Mapping and modeling spatial patterns of anthropogenic change—including climate change, land and water use change, and species invasions—is key for understanding and predicting variation in ecological and hydrological condition and function, and for anticipating and mitigating associated socioeconomic impacts. USGS data products capture key elements of resource condition and global change but we still lack an overarching framework to integrate, understand, and anticipate the interactions among different elements of global change. We propose to synthesize existing data products to describe the intensity of anthropogenic change and predict its implications for natural communities and ecological services. We propose to develop our Use Case with an initial focus on ecohydrological conditions in the PNW, while recognizing its applicability to terrestrial ecological systems as ecoregion boundaries complement watershed perspectives.

We will develop a platform to compile existing ecohydrological data and deliver assessments of ecological and hydrological condition to inform spatial prioritization decisions for water resources and species. Specifically, we would provide up-to-date and in-the-cloud compilations of environmental indicators, which are typically compiled in a laborious process for many individual National Environmental Policy Act (NEPA) and regional assessments by partner resource-management agencies and other stakeholders. Delivery of these compilations would be scalable depending on the targeted use (for example, individual stream reaches to regional HUC 4–6 watersheds). Our approach would reveal where multiple dimensions of global change intersect, and conversely, could identify areas that act as ‘multidimensional refugia’ for conservation and management prioritization. These refugia could be defined as areas experiencing low rates of multiple types of global change and (or) where system attributes are expected to minimize sensitivity. We would apply machine-learning and statistical analyses to characterize the linkages among datasets, quantify key interactions between the different elements of global change, and provide predictions and uncertainty bounds for data sparse areas. We anticipate that our synthesis would inform issue-specific analyses, such as updating threat assessments for salmonids, predicting algal blooms and their impacts, selecting indicators used in monitoring and adaptive management, and assessing trends and current conditions as necessary for NEPA assessments. We would develop our approach within the PNW, while ensuring it is scalable to the CONUS and ideally to Pacific Islands and Alaska (contingent on focal questions and datasets). We envision an interdisciplinary, spatially-explicit, and interactive online framework that delivers up-to-date information on ecohydrological status (for example, clickable summaries within HUCs) and helps decision-makers anticipate the impacts of anthropogenic changes within watersheds.

Our synthesis can inform a range of resource management planning that extends from local municipalities to federal agencies. We will provide nested grain sizes, from local (reach-scale) to larger watersheds, and analyses will be tailored based on stakeholder needs. Our aim is to co-develop our work with key partners (for example, BLM, USFS), while consulting with other users to ensure broad applicability and utility. Currently available tools for water resource assessments remain relatively siloed. For example, USGS Water-Quality Trends provides trends on many water-quality constituents based on discrete time series datasets, and Climate Engine is a powerful, in-the-cloud web application that delivers plots and spatial summaries. However, resource managers typically evaluate multiple datasets that can be difficult to align. In addition, many land managers may lack the technical background or resources to evaluate multiple datasets to understand trends and conditions to support the NEPA process. Following the conceptual framework of Climate Engine’s analysis and visualization, we will integrate ecohydrological datasets for targeted stakeholder needs.

Our framework can inform local decisions by capturing the regional value of a given resource. For example, one application considers a National Forest developing a programmatic NEPA document for logging. Our models will have complied current conditions (land cover, presence of invasive species, water quantity and quality) and indicators of change (for example, rates of land-cover conversion; climate change; change in flow regimes, drought and fire hazard) to identify areas representing multi-dimensional refugia. These summaries would facilitate decision making within management units while also providing regional context, which represents an identified need and ongoing challenge for USFS partners. Moreover, decisions could be made rapidly, without the need to collect additional data or undertake laborious analysis.

Existing datasets can include, but are not limited to the following portals: landcover (for example, National Land Cover Database, LANDFIRE, and FOREcasting SCEnarios of Land-use Change); water quantity (for example, NWIS streamflow, PROSPER streamflow permanence, NWIS groundwater levels, U.S. Drought Monitor, The Climate Toolbox); water quality (for example, NorWeST stream temperature, national water quality trends); and the composition of biotic communities (for example, USGS Nonindigenous Aquatic Species [NAS], eDNAtlas, USGS BISON, and BLM Assessment, Inventory, Monitoring [AIM]).

We have reached out to partners at the BLM to initiate discussions about the deliverables and use of our proposed platform. In a next phase of this project, we would expand discussions to include agency partners with whom our team has established relationships (for example, USFS, Bureau of Reclamation, state water resource departments) and we would reach out through our networks to new partners in state and federal agencies (for example, National Marine Fisheries Service), land trusts, conservation districts, and municipalities.

Our proposed Use Case would draw from existing human and technological capacity within USGS and we would broaden our team as needed. Data science workflows would utilize the APIs of established products. We would serve summarized information via a web-based platform, and to do so we envision leveraging existing data science and computer programming expertise within USGS (for example, the Oklahoma-Texas Water Science Center Integrated Hydrology and Data Science). We would apply statistical and machine learning models to understand the linkages between multiple sources of global change and indicators of biological and hydrological condition. Our team includes expertise in quantitative tools including statistical modeling, trend analysis and machine learning methods. One challenge we anticipate is that while some datasets provide spatially continuous information along streams and rivers (for example, NorWeST, PROSPER), many other temporally rich datasets are constrained to a point location. We would work with stakeholders to identify useful summaries of these point datasets, and could evaluate whether artificial intelligence tools can reliably predict ecohydrological condition based on a combination of continuous and point data. Our research team has experience designing and delivering novel products online, and we would reach out to include additional expertise in data management, other analytical approaches (for example, process-based deep learning), and visualization frameworks.

Our Use Case will have long-term relevance because it can inform a broad set of management prioritization decisions, and can facilitate revisiting priorities as new information emerges. We aim to replace the piecemeal process of data aggregation from multiple sources by delivering up-to-date and easily accessible data summaries. We will draw from flagship USGS products that capture landcover (NLCD), streamflow (NWIS, PROSPER), and invasive species occurrences (NAS). By leveraging ongoing USGS investments, our Use Case will be more sustainable and scalable with extensions to the CONUS and Pacific Island systems. The emphasis on USGS products places our agency in a leadership role for providing aggregated information, and we will also draw from well-maintained additional sources of information (for example, for climate-change patterns and projections). We will measure the effectiveness of our effort by its use by stakeholders and by the impact on their decision making.

Primary Contact: Kristin L. Jaeger, kjaeger@usgs.gov

The sagebrush (Artemisia spp.) biome, which spans over 160 million acres across 13 western states and provides important ecosystem services and benefits to both wildlife and human communities, is widely recognized as being imperiled as it has contracted by over 50 percent since European settlement. Notably, over 350 plant and animal species depend on sagebrush rangelands for all or part of their life cycle, including sagebrush obligates like the greater sage-grouse whose declining populations and threats to vital habitat have prompted multiple status assessments under the Endangered Species Act. Approximately 70 percent of the biome is managed by federal and state agencies. Accordingly, land-use decisions affecting the biome largely drive complex conservation policy at national and state levels. Land managers spend millions of dollars annually on land treatments to counter the increasing threats from disturbances such as wildfire, invasive plants, drought, and woodland encroachment to sustain multiple use activities such as livestock grazing, mineral extraction, energy development, and recreation that can pose their own risks.

While contemporary research efforts steered by inter-agency collaborations such as the Integrated Rangeland Fire Management Strategy Actionable Science Plan (Integrated Rangeland Fire Management Strategy Actionable Science Plan Team, 2016) and resulting Science Frameworks have greatly improved understanding of sagebrush ecosystems’ responses to disturbance and corresponding effectiveness of conservation and restoration techniques, there are clear opportunities to better incorporate past, current, and future knowledge into management practices. These efforts would build on nearly two-decades of co-production among Federal (DOI and non-DOI), State, and Local agencies, nongovernmental organizations, and the public across the West. To do so, USGS scientists across 4 regions (Northwest-Pacific Islands, Rocky Mountain, Southwest, Midcontinent) and multiple Mission Areas (Ecosystems, Energy and Minerals, Core Science Systems) comprising a Sagebrush Working Group (SWG) are primed to leverage multidisciplinary expertise spanning restoration, wildlife ecology, ecohydrology, + science, remote sensing, human dimensions, and advanced quantitative modeling. The EarthMAP framework will allow more effective integration of volumes of data describing ecological processes, disturbance histories, and management actions across the entire biome and over time periods ranging from decades to centuries. Formerly disparate, these data will form the backbone for multiple integrated predictive science tools and actionable intelligence. The underlying “engine” features enhanced spatiotemporal integration across ecological gradients and management jurisdictions and includes inputs for multiple prediction tools including data mining, hierarchical parameter estimation, and machine learning. User-friendly displays for decision-support will include creation of various treatment scenarios and corresponding real-time predictions based on underlying parameters that represent a host of rangeland ecosystem processes and services.

The SWG is developing science to inform planning and prioritization, conservation design and delivery, monitoring, and adaptive management. The science behind this Use Case will improve decision making at multiple scales, including placement of treatments on the landscape to maximize benefit to high value resources, and reduce conflicts between land uses, and selection of plant materials and actions that will minimize cost and maximize the effectiveness of the treatments.

Multi-scale and co-produced SWG science products forming a foundation for the Use Case are in high demand and widely used. For example, the National Landcover Database (NLCD) Fractional Shrubland Component and related ‘Back-In-Time’ (1985 and 2018) products allow biome-wide assessments of long-term changes in landscape condition and treatment effectiveness and identify potential conflicts between areas of high energy potential and wildlife habitat. These products coupled with predictive sage-grouse population habitat models inform implementation of the BLM/USFS Sage-grouse Habitat Monitoring Framework and the Sage-grouse Habitat Assessment Framework and facilitate conservation planning for actions such as fuels reductions, post-fire restoration, and energy development. In collaboration with BLM, FWS, and State Wildlife Agencies across all 13 states, the SWG most recently developed a sophisticated Targeted Annual Warning System (TAWS) that allows managers to identify local sage-grouse populations in need of additional monitoring and management intervention when they fall out of synchrony with larger population aggregations. SWG scientists have also developed a suite of tools that provide the basis for modeled predictions and decision support for sagebrush restoration and land-use planning efforts. These include several web-based applications aimed at increasing data accessibility and co-production with managers. The Land Treatment Digital Library catalogs over 100 years of treatment information and is employed by the Land Treatment Exploration Tool (LTET) which mines its historic data to predict management outcomes in user-prescribed areas, generating a variety of spatial products that feed directly into planning documents. The Smart Energy web tool incorporates spatial information to better predict areas of potential fluid mineral and renewable energy development. Both the Smart Energy and LTET web tools display information from the Conservation Efforts Database (CED). The CED captures data from state and federal agencies and nongovernmental organizations describing the conservation and restoration efforts implemented in the sagebrush biome and can provide important context to evaluating restoration effectiveness.

The SWG remains strongly engaged with multiple working groups through regularly scheduled conference calls and ongoing conversations to refine and update science needs. The latter includes lead coordination for the Integrated Rangeland Fire Management Strategy Actionable Science Plan (IRFMS) update to ensure the most pressing science needs of the management community are being addressed. The original Plan was developed in 2015/16 and the 5-year revisit of the plan has been initiated.

Going forward, the multi-scale framework and engaged SMG facilitates next-generation science and decision support tools that will be directly relevant to managers working to maintain and improve the condition of the sagebrush biome. This framework can lead to tangible benefits for the people and species that depend on this landscape, increase the efficiency and effectiveness of treatments, and reduce the long-term costs of actions necessary to achieve land management goals.

The USGS Sage-Grouse and Sagebrush Ecosystem Research Program (which comprises the SWG) includes scientists located in 11 westerns states from 3 of the 7 USGS Mission Areas and is led out of the USGS National Center, Ecosystems Mission Area. Research is purposefully aligned with priority needs outlined in the IRFMS, with DOI priorities as outlined in Secretarial Orders 3353, 3362, and 3372, and other science needs as they are identified by partner agencies.

This research can roughly be organized into five thematic areas: fire; invasive species; restoration; sagebrush, sage-grouse, and other sagebrush-associated species; and weather and climate. USGS research on these themes has resulted in a large and growing body of science that is meeting ongoing and emerging challenges faced by our stakeholders. Much of this work deals with the complex interaction of these themes; for instance, restoration of the sagebrush vegetation community, as is necessary following rangeland fire and extensive invasive annual grass invasions, requires precise deployment of limited resources to achieve the best possible outcomes. Working across multiple disciplines and stakeholders, USGS scientists are creating restoration tools that give early warning to managers to target actions in areas with decline sage-grouse populations, tools that enhance understanding ecological drought, and web applications to track ongoing investments and inform future management actions.

The newly completed Sagebrush Conservation Strategy provides a road map for the next 10 years of science and management interactions to resolve the grand challenges facing the sagebrush ecosystem. Using USGS and 21st-century science capacities we can improve the efficiency and effectiveness of land management actions, reverse or stabilized wildlife population declines, and improve the overall condition of western rangelands for the use and enjoyment of the American public.

The SWG has a demonstrated capability to develop actionable science tools by working closely with management partners to co-produce science that confronts existing and evolving challenges. In the process, they have built the foundation for the next generation of decision support tools that are informed by ‘big data’ and aimed at increasing the capacity to predict management outcomes. The suite of tools described in section 2, and others, deliver USGS science to management agencies in user-friendly interfaces that directly inform their decision-making processes. Other more recent products require further development to bring the science into operational forms. For example, the NLCD Fractional Shrubland Components and associated ‘Back-In-Time’ data are critical sources of information for multi-scale modeling across the working group, but real-time visualization tools to support management decisions are lacking. If integrated with the newly developed TAWS in an interactive web-based format, wildlife managers would have a highly valuable tool to identify where changes in land cover associate with declining wildlife populations, and better target land treatments. Existing and rapidly developing tools could be brought together into a more holistic framework of access to information across the spectrum and use the EarthMAP framework to better deliver USGS science.

USGS, in coordination with BLM, USFWS, and the Western Association of Fish and Wildlife Agencies (WAFWA), has also been developing SageDAT, a mechanism for broader integration of data and mapping services that allows for broader participation and transparency in decision making among federal and non-federal partners. The SageDAT team is building upon USGS ScienceBase to allow effective sharing and discoverability of relevant data and tools while maintaining/preserving control by data owners. This makes a vital linkage to State, private, and nongovernmental data for the next generation of online tools. This system will also be an outlet for access and delivery of USGS data and tools. In addition, USGS web application development teams at FORT (Fort Collins Science Center) and FRESC (Forest and Rangeland Ecosystem Science Center) have successfully developed tools (see Section 2) to inform management actions. The team at FORT is pioneering the move of USGS to cloud-based web application development and brings significant capabilities to ongoing and future development. Despite these successes, teams will require additional resources to support development costs of the next generation of tools, as current support for ongoing operations and maintenance of the cloud-based tools remains a challenge and is an impediment to development of new operational tools. Most of the funding for these efforts is provided by management partners via project-base interagency agreements that have a limited time span. As USGS web-applications increase in complexity, programming capabilities and computational power will need to increase to facilitate advanced modeling, real-time processing, and the integration of larger data streams. Improved mechanisms to recruit and retain computer scientists and other IT-related positions will be critical for long-term success of the data and tool delivery component of this effort.

Overall, this effort is built on the continued advancement of science related to sagebrush ecosystem vegetation and wildlife responses to ongoing disturbances and management actions. The development of new tools will further hinge on improved field-based monitoring of vegetation conditions and wildlife populations. This could be through the development of new sensor technologies, refinement of auditory surveys for passerine birds, advances in genetic barcoding and eDNA technology for terrestrial species, ongoing mapping of human activities, and other important data streams to support the monitoring component. USGS research scientists continue to advance the assessment capabilities to help inform predictive models.

Demands for services and resources provided by the sagebrush ecosystem, and subsequent reliance on their effective management, are increasing. Urban and suburban population growth, demand for energy (renewable and non-renewable), ranching and other agricultural uses, and consumptive and non-consumptive recreation will continue (and in some cases rapidly expand) into the foreseeable future, while the spread of invasive plants, devastating wildfires and prolonged drought are compounding the complexity of the management decisions to sustain these landscapes. For DOI, other agencies and private landholders responsible for the management of millions of acres in the West, decisions regarding placement and selection of land treatments to maintain and or improve the sagebrush ecosystems and mitigate threats remain an ongoing challenge.

By building on long-standing partnerships with the federal and state agencies, ranchers, and other NGOs tasked with resolving the growing array of resource conflicts, USGS can sustain long-term coproduction of the science that addresses managers’ priority information needs. Success will require continued investment in the multiregional and inter-Mission Area SWG, engagement by USGS personnel at all levels of the agency with the appropriate DOI and state agencies, and stable fiscal support of the research and tool development identified by management partners. Reliable long-term fiscal support would facilitate research to develop the science to drive the next-generation tools and costs of web application development, operations, and maintenance. A sustained financial commitment from the USGS will demonstrate our commitment to our partners who have long supported this work, and who continue to request and use it.

Over the past decade, the land management agencies have put several monitoring systems in place, including the Sage-grouse Monitoring Framework, the BLM Assessment Inventory and Monitoring program, and National Resources Inventory, that can help track multiple metrics of success. These programs provide local- and landscape-level metrics of management effectiveness. Important metrics to follow will include changes in vegetation condition, costs of land treatments, size and intensity of rangeland fire, population size and viability of wildlife populations, and prevalence of requests to list species under the Endangered Species Act.

Development of tools to increase the success of treatments in the sagebrush ecosystems can be directly applicable to semi-arid and arid rangelands throughout the western U.S. (Mojave, Chihuahua, Colorado Plateau, Sonoran), and throughout the world (for example, Argentina, China, Mongolia, East Africa). The decision analysis frameworks proposed here will be scalable to other at-risk species and ecosystems. For example, the greater sage-grouse TAWS can provide a framework for any species with expansive range and consistent monitoring information; treatment evaluation tools can be established if information on prior treatments are available for a particular region. The partner engagement in the identification of science priorities as outlined in the IRFMS, the co-production of multi-agency and interdisciplinary research that feeds directly into management guidelines as demonstrated in the Science Frameworks, and the integration of massive amounts of data and information as is pursued in SageDAT are highly successful and noteworthy models that can inform similar efforts to evaluate the success of complex land and species management issues across the globe.

Primary Contact: Lief Wiechman, lwiechman@usgs.gov=

Warming temperatures as a result of climate change are predicted to increase drought severity and cause shifts to aridity across the United States (Martin and others 2020; Overpeck and Udall, 2020). It is increasingly recognized that understanding trends in persistence of surface-water features on the landscape will be crucial for water-resource managers. Surface-water features, such as headwater streams and springs, are particularly important as they serve as the primary interface between terrestrial and aquatic species and systems and can function as ecohydrologic refugia—areas that remain buffered from changing hydroclimatic conditions over space and time—in otherwise water-limited landscapes. However, the vulnerability of these ecohydrological refugia to changing climatic conditions are poorly understood. The paucity of existing, consistent datasets representing surface water-permanence has historically limited investigations of changes in surface-water availability over multiple time steps. Further, because of this data gap, scientists have previously been unable to investigate how these changes in surface-water availability and timing affect wildlife behavior and ecosystems services. Understanding this relation has implications for management strategies, including (i) areas to focus efforts for invasive species and pathogen preventions and (ii) elucidating the consequences of restoring connectivity among aquatic systems. While this Use Case will focus on studying the link between surface-water permanence and the spread of diseases and aquatic invasive species, the resulting data from this work will facilitate investigating the role of surface-water availability across a much broader array of ecohydrologic and socioeconomic systems (fig. 5.8).

This EarthMAP project seeks to provide managers, decision makers, and the public with actionable information at multiple relevant spatial and temporal scales about surface-water permanence, stream-network connectivity, and ecosystem effects. Surface-water permanence, which includes headwater streams and groundwater expressed on the surface as springs, drives many management decisions regarding water-quality standards, water rights, and land use. These management decisions are constrained by available tools that neither adequately capture the year-to-year variability in the location and duration of flowing streams, nor describe their connectivity between locations. Consequently, there is a need for up-to-date surface-water permanence classifications that reflect both average conditions, as well as the year-to-year changes across larger extents and at multiple scales. This information would be extremely informative for water-use management. For example, clean and abundant water is an identified need for livestock grazing, energy development and other multiple-use categories defined in Section 103 of the Federal Land Policy and Management Act of 1976 (U.S. Department of Interior, 2016. Therefore, improving understanding of surface-water permanence is necessary for effectively managing public lands and ecohydrologic resources for sustainable multiple use.

The USGS has developed multiple modeling techniques that are designed to predict the likelihood of streamflow permanence and spring permanence using state-of-the-art, data-driven methods including remote sensing and machine learning. For example, the PRObability of Streamflow PERmanence (PROSPER) is an empirical, machine-learning based model that predicts the probability that locations on streams will retain surface flow through each year. Additionally, the USGS has done work to develop a remote-sensing method for detecting the sensitivity of springs to climate change. Both of these approaches have been used at different scales in the Pacific Northwest. By combining output from these methods, or recalibrating the methods to incorporate aspects of both approaches into a more holistic watershed model, we are creating the opportunity to understand interannual changes in surface-water availability at an unprecedented temporal granularity and in a regionally consistent manner. Further, we plan to implement state-of-the-art sampling techniques, including the use environmental DNA (eDNA), which allows us the capacity to quickly and cost-effectively sample and detect the presence of species in a watershed.

A logical path forward with this work is to implement a three-phase plan to investigate surface water-permanence and its relationship with invasive species and pathogen spread. Phase 1a: Utilization of currently available datasets to develop perennial surface-water connectivity maps. In phase 1, we will collate existing USGS products, including the PROSPER model (Jaeger and others, 2019) and a remote-sensing approach for evaluating spring vulnerability to climate change (Cartwright and Johnson, 2018) to produce a probabilistic index of ecohydrologic refugia (surface-water permanence) at biannual time steps within a watershed (HUC-10; see fig. 1. Phase 1b: Development of risk maps based on current data and understanding of aquatic invasive species occurrence and pathogen transmission. Congruence between SWIPE and aquatic invasive species and pathogen occurrence data will inform machine learning algorithms, which aim to predict the probability of invasive species or pathogens introduction, existence, and spread. Risk predictions will be ground-truthed using strategic eDNA sampling methods to detect presence of the target species. Sampling will be performed twice per year, at high- and low-flow periods in aquatic areas that are predicted to be connected, as well as a subset of aquatic areas that are predicted to be isolated. We will demonstrate these methods through two use case assessments, the first addressing an invasive species (for example, grass carp, Ctenopharyngodon idella) and the second addressing an emerging infectious disease (for example, amphibian chytrid fungus, Batrachochytrium dendrobatidis). Phase 2: Cloud-based interactive mapping tool for managers with capacity to incorporate new datasets. Products from phase 1 will be implemented through a tool on a cloud-based platform, with intent to publish the tool for public use. Phase 3. Predicting future spread of pathogens and invasive species. Using the probabilistic index of surface-water permanence from phase 1, we will forecast stability of these surface-waters under three climate change scenarios: (i) increased temperature and stable precipitation, (ii) increased temperature and decreased precipitation, and (iii) increased temperature and increased precipitation. Using these forecasts for surface-water permanence, risk maps will be created to identify refugia and areas where disease and invasive species have potential to spread based on connected surface-waters.

This work will require key experts across multiple disciplines and USGS Mission Areas. Specifically, data scientists, remote sensing specialists, machine learning experts, aquatic ecologists, disease ecologists, hydrologists, and potentially others. Because of the amount of data that will be processed and analyzed, the majority of the work will be done on the USGS High-performance Computing systems (YETI, Tallgrass, Denali), and will also leverage a cooperative agreement that is being developed between the USGS, Google, and Desert Research Institute. Many of these resources/collaborations are already in place and can easily be leveraged.

The products from this project will be developed in a modular manner with the capacity to (i) be applied and (or) expanded to multiple scales ranging from local (stream reach) to regional, and (ii) incorporate new data streams—specifically, new invasive species or pathogen occurrence data—to create risk maps for novel and on-going threats. All data and models will be developed and based in the cloud to minimize maintenance and the potential for abandoned and broken services. Finally, as the rate of climatic change is predicted to be significantly higher than in the past, the threats from water-related stressors and their effect on invasive species and pathogen dynamics will increase. Because of this, we anticipate the products of this project to have immediate national application and interest that will only increase in value as the impacts of climate change, increased water use, and more frequent human/wildlife interaction are realized across the country.

Primary Contacts: Alynn Martin, alynnmartin@usgs.gov; Lindsey Thurman, lthurman@usgs.gov; Roy Sando, tsando@usgs.gov; Joseph Kennedy, jjkennedy@usgs.gov

Water timing and availability directly and indirectly influence terrestrial wildlife and their habitat, yet linking these explicitly is rare, even where data exists. Managers need to better understand these connections to protect, maintain, and restore key wildlife and plant populations and the ecosystems to which they belong. Subalpine and alpine systems in the mountainous western United States are highly vulnerable to climate change, but provide opportunities for water, vegetation, and disturbance management that could greatly affect multiple species both in these communities and downstream. By integrating existing datasets (for example, representing snowpack, surface water, evapotranspiration, glaciers, avalanches, fires, soils, vegetation, and wildlife), we will predict vegetation and wildlife responses to the effects of changes in water flow, availability, and seasonal timing in the subalpine and alpine regions of the northwestern United States. Our initial focus will be on alpine and subalpine National Park Service units in Washington and Montana, where our team has existing connections with decision-makers tasked with conserving and managing wildlife population and ecosystem function. However, we envision predictive modeling results to be relevant across multiple land management jurisdictions and ecoregions. We will map changes in subalpine and alpine meadow size, extent and vegetation composition as a function of shifting surface and subsurface water and evaluate subsequent effects on wildlife distribution and population dynamics. In this effort, we will integrate water, vegetation, and wildlife data (across multiple trophic levels) for new insights; synthesize patterns in National Park units across the northwest; and provide a template for broader use in connecting water to wildlife from local scale management questions to regional trends. This effort is unique in that prior links between water and wildlife have focused primarily on aquatic species.

National Parks have a mandate not only to preserve individual species, but also to maintain ecological functioning and integrity in changing ecosystems. As such, national parks have an immediate need for understanding relationships among water, vegetation, and wildlife in order to plan proposed infrastructure (for example, cell towers), backcountry visitor use, weed management, and prescribed fire, and to manage for species of concern including invasive species, disease, aquatic insects, bumblebees, pikas, marmots, mountain goats, and grizzly bears. Our team has close, long-term connections with National Park personnel that will ensure that such work will improve decision making and focus on key deliverables. Our products will provide park-specific recommendations while also determining patterns of change across mountain systems that will be relevant not only to park stakeholders but also to surrounding landowners and resource managers. Our products will help identify existing climate refugia in subalpine and alpine systems, as well as areas that may require active management. While the first phase of this work is focused on select National Parks, the need to manage wildlife and its interaction with humans is universal and a goal of most federal, tribal, and state agencies. Ultimately, the framework developed here will be readily applicable in other, lower elevation ecosystems where managers are faced with similar questions and management challenges. The results of the expanded study would be informative to issues such as human/predator conflicts, effects of development on migration routes, changes in ecosystem carrying capacity related to climate, changes in ecosystem services and economic value, and many others.

This Use Case will leverage existing but disparate datasets to produce a novel, cross-disciplinary analysis that explicitly links key water quantity and timing metrics to effects on vegetation and wildlife. First, we will work with wildlife, snow, and other locally collected datasets contributed by Park partners in the region to establish a proof-of-concept within select National Parks. We have initiated conversations with partners in multiple National Parks (Olympic, Mount Rainier, and Glacier) who are excited to integrate local datasets with remotely sensed and (or) regional datasets. Working collaboratively with stakeholders to develop the project more fully and incorporate their knowledge and available data will be key to designing and developing products useful to biologists and managers. Regional hydroclimatic datasets (for example, remotely sensed snow products, and gridded precipitation, temperature, evapotranspiration, etc.) will be utilized to capture the mechanisms driving the changes seen in vegetation and wildlife distribution and use. We will incorporate modeled (for example, PROSPER, PRMS, NWM, others) and observed (USGS streamgages, citizen science applications such as FLOwPER, others) hydrologic data to inform our models on the availability and timing of surface water flow. Finally, we will rely on remotely sensed data developed by USGS and other agency partners (for example, LEMMA on vegetation cover and photosynthetic activity, Berman and others (2020) on changing vegetation phenology and productivity, Manier and others on soil moisture /evapotranspiration, DHSVM [Distributed-Hydrology-Soil-Vegetation-Model]) to link water and snow data to vegetation. By linking these datasets, we can evaluate the direct effects of both water and vegetation on focal wildlife species to better understand mechanisms and potential management options for reaching wildlife management goals. Example focal species with existing datasets include pikas, Western bumble bees, mountain goats, Olympic marmots, and grizzly bears. While the initial project would be focused on select National Park units, we would ultimately reach out to other potential partners, as the results of this study would be broadly applicable, and explicit linkages between water and wildlife remain an important and understudied aspect of ecosystem function across the US.

This Use Case draws on our team’s varied expertise within the USGS and existing relationships with National Park unit scientists and managers. Our team includes scientists with technical expertise in wildlife and plant ecology, remote sensing applications for hydrology, surface water-groundwater interactions, and relevant statistical analyses and modeling. Our strong relationships with National Park partners will enable access to key datasets and has informed the Use Case development. We will work closely with stakeholders to ensure that our data products meet their needs. We further plan to broaden our team to include additional partners with technological (for example, machine learning) or ecosystems (for example, fire, glacier and avalanche, additional remote sensing) expertise as appropriate. Other areas we anticipate needing additional capacity include expertise in data management and visualization, certain analytical approaches (for example, some aspects of machine learning), and effective, accessible delivery of products over the long-term (for example, online portal).

We anticipate several products from this project. (1) We will develop models quantifying the relationships between water quantity and timing metrics, vegetation indices, and wildlife seasonal use and distribution. These model results will provide immediate insight to park stakeholders on direct and indirect drivers of wildlife occurrence, which will aid in current management decisions. (2) We will then build models and create maps showing predicted changes in the quantity, composition, and quality of key habitats such as subalpine and alpine meadows under different climate change scenarios (that is, modeled changes in key hydroclimatic drivers). This modeling and mapping effort will identify climate refugia and threatened habitats within the focal ecosystems, which can guide active management and conservation efforts. (3) We will extend these models to show predicted changes in indicator and flagship wildlife species use, distribution, and abundance in meadow habitats under these same scenarios. These models will allow parks to understand the spatiotemporal characteristics of wildlife occurrence in new and integrated ways. Once the integrated models and predictive algorithms have been developed for this National Park case study, they can be adapted to other ecosystems of interest for local to regional applications.

Understanding linkages and relationships among water, vegetation, and wildlife are key to projecting and assessing ecosystem integrity, resiliency, and adaptability in diverse ecosystems under climate scenarios that forecast increasing temperatures, changing precipitation regimes, and changes in the frequency and magnitude of extreme weather events. Our Use Case will provide immediate patterns and predictions relevant to DOI stakeholders in National Parks tasked with conserving, maintaining, and restoring viable populations of listed species and key habitats. The tools developed here will help with decision support in the near- and long-term in these fragile subalpine and alpine environments and new or updated datasets can be incorporated over time. More broadly, the tools developed here will be scalable and transferable to (1) high-elevation ecosystems found in other land-management jurisdictions, and (2) other ecosystems where understanding linkages between water availability, vegetation, and wildlife is critical to effective decision making under future climate change scenarios. Integrating data from water to wildlife will have broad application for understanding wildlife and vegetation connections with and responses to climate-related changes in fire regimes, snowpack, and precipitation.

Primary Contact: Rebecca McCaffery, rmccaffery@usgs.gov

Year after year western communities are made painfully aware of expensive and long-term economic and social impacts caused by wildfires, including the loss of property, livelihoods, and lives. Economic impacts alone amount to annual costs of $5 billion for fire suppression and more than $80 billion in direct and indirect losses. Yet fire in wildlands is essential to reduce future wildfire risk, improve wildlife habitat, maintain or restore ecological resilience, and manage fire-adapted species in many ecosystems across the Nation. Finding balance between wildfire benefits and risks requires information to support decision making and evaluate management alternatives. The Wildland Fire Leadership Council (executive level wildland fire representation of DOI, USDA, DHS (Department of Homeland Security), DoD (Department of Defense), States, Tribes and Communities) identified tools to assess vulnerabilities and mitigate direct and indirect fire risks before, during and after wildfires as one of the greatest unmet needs. A critical need is integrated predictive models for fire occurrence and severity, and potential for post-fire debris flows and changes in water quality and quantity. This need could be met by integrating fire science across USGS disciplines in a dynamic “fire risk” map that combines the:

Such integrated assessment and prediction capabilities would enhance the ability of the wildland fire and emergency management communities to protect the safety, health, and prosperity of communities, minimize economic losses and ecosystem damages, and enhance post-fire mitigation and recovery.

The USGS is perfectly positioned to meet this need. Our expertise in fire and hazard-related sciences is well recognized by stakeholders that work with USGS to address challenges in wildland fire science and management across the Nation. However, USGS’s many fire science capabilities require integration in order to support truly actionable management applications and decision making. We propose an interdisciplinary science integration effort for EarthMAP that crosses the Ecosystems, Core Science Systems, Natural Hazards, and Water Resources Mission Areas with external partners from the wildland fire community to align and improve USGS’s diverse risk assessment and prediction capabilities.

The work we propose to integrate is designed to address the needs of our diverse stakeholders and aligns with agency, department, and national research strategies and management priorities:

Our stakeholders include DOI, USDA, DHS, DoD, States, Tribes and Communities represented by the Wildland Fire Leadership Council and involved with managing and mitigating the impacts of wildfires. Our science-based fire risk map will provide stakeholders with actionable intelligence by producing spatially consistent and detailed information about the potential for post-fire risks that are not currently available. Creating this “EarthMAP on Fire” vision will require close collaboration with the fire management community, to ensure that the results are truly actionable, and with the fire science community, to ensure that the best available science, is incorporated. We will work with the Wildland Fire Leadership Council and stakeholders to co-develop our approach, ensure the approach is useful for their needs, and fully understand how it would inform decision-making processes. Examples of where our risk map could inform decision making include: (a) activities to reduce the likelihood and severity of fires; (b) real-time fire suppression priorities targeted to areas where the impacts are of most concern, or where fire recovery would be most difficult; and (c) assessment of post-fire risks and recovery challenges to speed the deployment of rapid responses, such as debris flow management or revegetation and invasive species management. The post-fire decisions of fire and emergency managers will be further supported by near-real time updating of the fire risk map based on remote sensing observations of changes in the burned areas and by later monitoring and observation.

The USGS has many fire science capabilities. These include monitoring efforts for land cover (Land Change Monitoring, Assessment, and Projection [LCMap] and National Land Cover Database [NLCD]), vegetation (Gap Analysis Project [GAP] and LANDFIRE), fuels (LANDFIRE), fires (Monitoring Trends in Burn Severity [MTBS] and Landsat Burned Area), fire occurrence, and burn severity modeling, as well as, efforts to model fire occurrence and smoke emissions and post-fire geological and hydrological impacts. However, these capabilities are not yet integrated in a way that allows us to produce a fire risk map or to answer the management challenges described above. Getting there—achieving the goal of predictive, integrated fire-climate-ecosystem-hazards models that link landcover, fire risk, fire behavior, post-fire hazards, and short-and long-term fire effects, including ecosystem resilience and future fire risk—requires everything that EarthMAP aims for. It requires existing and new monitoring, with data made available rapidly, integrated across disciplines, and accessible to scientists and decision makers alike. It requires integrated and predictive models to evaluate the potential consequences of fire as part of the fundamental “pre-fire” risk assessment, and models that can evaluate post-fire conditions and highlight effective recovery options.

This effort aims to assess and reduce the risk of post-wildfire hazards such as flooding, debris flows, and water-supply impairment. While there is much existing USGS research that could be built upon, central to this effort is the need for continued research to develop more sophisticated, complex, and informative models that improve assessment and prediction of potential fire occurrence, behavior, hazards, and risks, as well as post-fire effects and recovery. Improved modeling capability relies on integration of fire, water, and geomorphology models with complex, 3D, dynamic, multi-scale input data. This effort is organized around four objectives to provide an integrative simulation modeling framework.

Objective 1—Develop simulation models to predict fire behavior, burn area, and burn severity:

The first step of this effort would link models that simulate and predict fire ignitions, fire behavior and perimeters, and burn severity to provide necessary inputs for post-fire hazard models. To this end, we will first assess the degree to which existing fire behavior and prediction models can generate these inputs. In particular, fire intensity has been used as a surrogate for fire severity in a previous debris flow modeling effort, but the relationship between modeled fire intensity and observed burn severity has not yet been assessed; if this relationship is poor, new methods for predicting burn severity will be developed. Our burn severity models will predict both classified and continuous measures of burn severity, as well as modeled correlates of field-based severity to allow for field measurements and validation. Coupling our burn severity models with ignition probability and fire behavior models will allow us to generate thousands of simulated fires which can be aggregated to provide probabilistic risk estimates of fire hazards.

Objective 2—Link fire model predictions of burn area and severity to debris flows models:

We will link fire model predictions developed in Objective 1 with debris flow models following the same operational approach used by the USGS to assess debris-flow hazards after observed fires. The USGS currently delivers post-fire debris-flow hazard assessments at the request of state and federal Burned Area Emergency Response (BAER) teams. Observations of burn severity are a critical input to these models. To produce an assessment of debris-flow hazard and risk that can be used to inform management decisions before a wildfire occurs would require predicted burn severity. In our integrated simulation framework, predicted fire perimeters and burn severity will be used as inputs to empirical models to estimate debris-flow likelihood and volume. Linked wildfire-debris flow models will be used to identify drainage basins with high debris flow risk, information that can be used by land management agencies to target fuels reduction or fire management activities that could potentially reduce burn severity and related post-fire hazard.

Objective 3—Link fire model predictions of burn area and severity to water models:

We will initiate systems development necessary to link predictive fire products with the USGS StreamStats program, a web-based application that is used by stakeholders to obtain watershed areas and characteristics and streamflow statistics (for example, flood flows, mean flows, low flows). StreamStats is currently developing a national application which will incorporate historical and active fire data (burn severity and fire perimeters from the Monitoring Trends in Burn Severity Program and National Interagency Fire Center, respectively) to identify water supplies that may be affected by wildfire runoff and to enable assessment of post-fire peak flows based on existing regression equations derived from post-fire runoff studies. We propose to supply simulations of burned area and severity to StreamStats from the models developed in Objective 1 to enable the assessment of potential risks of flooding and water supply impairment before wildfires occur. Such predictions would be useful to BAER teams and could enable at-risk communities to mitigate flood risks preemptively. Objective 3 activities would also enable the WMA (Water Resources Mission Area) to identify potential locations for immediate installation of streamgages and water-quality monitoring instruments, should a fire occur. This would enable both an alert system to downstream communities to prepare for and mitigate against flooding and water-supply impacts, and also provide model input to ongoing WMA efforts to predict impacts of extreme events on water availability.

Objective 4—Developing decision support functionality:

Through stakeholder engagement, we will determine what decision support functionality is desired and add capabilities incrementally. Because our approach operates on individual simulated fire perimeters, we will work with the wildland fire management community's IT organization to integrate our framework into the Wildland Fire Decision Support System (WFDSS). This would allow WFDSS users to generate near real-time predictions based on known ignitions at watershed or basin scales, directly supporting BAER teams.

Sustained relevance of this Use Case is anticipated to be high as wildfire occurrence is expected to increase in the future given past legacies of fire suppression and projected patterns of development, and climate change. Furthermore, post-fire vegetation regrowth may also be limited or transition to novel ecosystem types because of continued drought conditions. Sustained long-term investment of this Use Case through USGS supported for research, stakeholder engagement, and computing infrastructure would enable:

Primary Contact: Paul Steblein, psteblein@usgs.gov

The growing population of the arid and semiarid Southwest relies on overallocated surface water resources and poorly quantified groundwater resources (fig. 5.9). To meet human water needs, water managers must meet water delivery requirements while planning for future water supplies. Recent studies have found that about 50 percent of the surface water at USGS stream gages in the UCRB is derived from groundwater as base flow. Water resource management has focused on securing the surface water supply, and current water resource models simplify groundwater processes such that surface water is disconnected from groundwater in aquifers. Such simplifications reduce the ability of the models to represent base flow and complex interactions between groundwater and surface water in response to evapotranspiration, groundwater pumping, and their feedbacks in a variable climate and during drought. As a result, simulations of water resources and the impacts of variable climate on groundwater and surface water remain uncertain. We aim to improve water resource tools by providing daily and seasonal forecasts of water availability using continuously updated simulations of physically-based groundwater flow and surface water as a single, integrated resource.

In 2018, the USGS Water Resources Mission Area (WMA) and the Water Availability and Use Science Program launched an investigation of groundwater and surface water as an integrated resource in the Upper Colorado River Basin and parts of the Lower Colorado River Basin (Rocky Mountain and Southwest Regions). This ongoing study, called the Colorado Plateaus Aquifers: A Water Availability Assessment of Groundwater and Surface Water as a Single Water Resource project, here called the “Colorado Plateaus Project,” will complete a model of integrated groundwater and surface water in Fiscal Year 2022 using the USGS software GSFLOW (Coupled Groundwater and Surface-Water Flow Model). An advantage of GSFLOW over hydrologic models by other agencies is that it simulates groundwater and surface water as an integrated resource. That is, it simulates how surface water is affected by changes in aquifer flow from groundwater pumping and human water use, changes in evapotranspiration, and changes in groundwater discharge as baseflow. GSFLOW also simulates runoff and snow melt contributions to groundwater recharge and to streamflow. These hydrologic couplings allow for rigorous representations of watershed hydrology with groundwater and surface water interactions. We are currently using the GSFLOW model to address the following questions.

The Colorado Plateaus project, upon completion, will have a daily, high spatial resolution (1 km²) model ready for operational expansion for EarthMAP to address the above questions and stakeholder needs. This expansion is not part of the existing project scope. Our model complements new USGS efforts aimed at innovative hydrologic monitoring and integrated hydrologic modeling and prediction capabilities in the Upper Colorado. The WMA’s new programs, Next Generation Water Observation Systems (NGWOS) and Integrated Water Availability Assessments (IWAAs), are in the initial phases of planning and implementation in the UCRB. Our model is planned to be a key tool in the water availability strategy in the IWAAs UCRB project, in collaboration with surface water modeling by the National Weather Service (NWS), National Center for Atmospheric Research (NCAR), and the Bureau of Reclamation (Reclamation). Our model could be used to improve estimates of salinity in streams by Reclamation's Colorado River Simulation System. Our forecasts could also be used by the Bureau of Land Management’s Regional watershed modeling efforts related to salinity and rangeland impacts with the APEX (Agricultural Policy/Environmental eXtender) model. The Colorado River Salinity Control Forum needs model predictions of surface flows to improve predictions of salinity transport mechanisms from groundwater. Simulated streamflow may benefit U.S. Fish and Wildlife in assessments of ecosystem services.

The Colorado Plateaus GSFLOW model can provide actionable and timely forecasts of integrated hydrological conditions across the UCRB, a product that is not currently available to decision makers or scientists. We will expand the capabilities of the GSFLOW model to provide forecasts from the National Oceanic and Atmospheric Administration (NOAA) Climate Forecast System which is updated daily (up to 28-day outlook) and monthly (6-month outlook). Such forecasts are needed for both water management and scientific understanding of hydrologic and ecological processes. Water use and demand forecasts may be provided by local and federal agencies (Reclamation) and the USGS Water Use model. The resolution of the forecasts will be daily through time and at spatial resolutions of 1 km². Daily temporal resolution is fine enough to represent the dynamics of fast hydrologic responses to large precipitation events from storms, the accumulation of snowpack, and rapid snowmelt and runoff to stream channels. The 1 km² spatial resolution resolves dynamic groundwater and surface interactions and baseflow throughout the UCRB.

Forecasts at these timeframes are highly sought by Federal, State, and local resource managers as there may be enough actionable lead time to design and implement management strategies to address periods of excess water or drought. For the science community, the GSFLOW model provides a conceptual understanding of the hydrologic system and a science-support tool for evaluating the possible effects of human water development and climatic stresses on ecohydrologic systems. For example, our GSFLOW model has been identified as a potential tool for identifying jurisdiction of intermittent water for the Clean Water Act by the interagency effort on the Waters of the United States which includes the USGS, EPA, and U.S. Army Corps of Engineers.

This Use Case may generate new collaborations with stakeholders such as Reclamation and NWS. Reclamation currently provides essential river forecast products for the UCRB. However, their forecast models contain simplified groundwater and baseflow processes. There is opportunity to collaborate directly with Reclamation to evaluate and possibly improve model performance. Our GSFLOW model can fill gaps in predictive capability by simulating temporal and spatial changes in low and intermittent streamflow during drought periods that are sustained mainly by baseflow. The model can resolve the locations of these vulnerable stream reaches at 1-km resolution and forecast how wet and dry reaches may change seasonally.

The Colorado Plateaus GSFLOW model provides a foundation for providing actionable forecasts of groundwater and surface water conditions. Extensive existing data, software, and infrastructure have already been developed by the USGS and specifically by the Colorado Plateaus project to construct the model. However, this Use Case has many opportunities for advancement, especially with monitoring, analysis, and modeling aspects of EarthMAP.

The existing data for constructing the model are multidisciplinary by necessity. This includes (in part) geology, surface water flows, water chemistry, groundwater levels, vegetation, land cover and use, snow cover, and water use data. Local scale and regional geology are used for defining soils and regional hydrogeologic units for the entire Upper Colorado River basin. The hydrogeologic units are defined in geologic framework models (already published as data releases) that define the spatial extent of permeable and impermeable units, as well as the connections between units. The watershed model is parameterized and calibrated with a wide range of vegetation, land use and land cover, snow cover, recharge, and surface hydrology data. Many of these data are derived from global and CONUS-wide satellite data products by NOAA and international hydrologic models (WaterGAP).

Many opportunities exist for new scientific advancement and capabilities through innovative data collection and monitoring and new modeling approaches. New collaborations with NWS and the National Center for Atmospheric Research (NCAR) aim to modernize the hydrologic modeling portfolio of the USGS. Innovative observation is the thrust of NGWOS, such as satellite-derived streamflow monitoring and inexpensive-but-widely-dispersed streamflow sensors across the landscape. Our model can immediately use new data types to resolve hydrologic dynamics, improve model calibration, assess model uncertainty, and implement data assimilation by directly combining model predictions with observations.

The expansion to an operational model can be done with existing models and computational infrastructure in the USGS but many aspects will require additional capacity expertise that might not be currently available in the USGS. The GSFLOW model will be complete and ready for transition to an operational framework in fiscal year 2022. This can be done by existing staff on the Colorado Plateaus project, by USGS developers of the software codes for GSFLOW and the code Precipitation-Runoff Modeling System (PRMS), and by staff on the former USGS project “WBEEP” (Water Budget Estimation and Evaluation Project; part of which is now contained in the National IWAAs Integrated Products project). The WBEEP project released an operational version of the National Hydrologic Model (NHM) in fiscal year 2020 that runs on the USGS supercomputer Denali. The NHM is based on the PRMS modeling code, the same code contained in GSFLOW for simulating watershed hydrology. It is feasible to implement existing tools from the WBEEP project to operationalize the GSFLOW model. The Colorado Plateaus GSFLOW model is already being run on Denali for calibration and evaluation.

Forecast input data from gridMET provides surface meteorological drivers. gridMET data are gridded datasets that are updated daily (data exist for prior day) at 4-km resolution for the entire CONUS. gridMET provides daily and seasonal forecast products from NOAA’s Climate Forecast System (CFSv2), which uses four daily ensemble forecasts that span the next six months. Both the NHM and the Colorado Plateaus GSFLOW model are driven by gridMET data. The operational NHM model acquires data for the prior day’s meteorological drivers (precipitation and temperature) and automatically runs the model daily.

Additional tools and capabilities needed to operationalize the model include data assimilation, and data storage and mechanisms for the public and cooperators. Data assimilation, in addition to calibration, constrains model states and predictions on a real-time basis to improve model fidelity. Data assimilation is rarely used in USGS hydrologic models but fully implemented in weather forecast models. Expanded data delivery tools will be needed by cooperators and the public (for example, Pangeo, Google Earth Engine, USGS GeoDataPortal).

This Use Case is likely to have many years or possibly decades of viability based on cooperator need and improved capacity and technology developed under EarthMAP. The issues of water availability under human and climatic stresses are likely to continue indefinitely. The need for this Use Case is demonstrated by the launch of new USGS Water Resources Mission Area priorities: new innovative observation networks, advancements in integrated water resource modeling and assessment, improved delivery of products to stakeholders, and modernization of our internet presence and product delivery. This Use Case addresses those priorities.

This Use Case outlines several advances in tools using existing USGS software and operational modifications that will continue to be in high demand and are likely to transfer to any hydrologic environment with integrated groundwater and surface water. The USGS and USGS Water Resources Mission Area aims to advance its hydrologic modeling software to new modeling paradigms of other agencies like NCAR. These paradigms include tight coupling of atmospheric and surface water models (for example, WRF-Hydro). However, tools that couple groundwater with atmospheric and land surface processes are not yet available; USGS GSFLOW is the state-of-the-art tool for integrated groundwater and surface water modeling. Ultimately, the longevity of this Use Case may be extended into the next decade through collaboration with NCAR by coupling with atmospheric models, and the authors of this Use Case are enthusiastic about this opportunity.

New observation networks by EarthMAP and NGWOS will be directly implemented in this Use Case. Extensive data networks will be needed to evaluate the performance of the operational model, assess forecast uncertainty, and to drive model predictions through data assimilation. There are many opportunities to collaborate directly with Reclamation, NWS, BLM, NCAR, and other agencies to develop performance metrics based on new and existing networks, and the forecasts are likely to improve through that engagement.

Primary Contacts: Jesse Dickinson, jdickins@usgs.gov; Melissa Masbruch, mmasbruch@usgs.gov

Natural resource management depends on accurate and timely environmental modeling and data integration to support a broad range of environmental and societal water issues across spatial and temporal scales. These applications require continued and evolving advances in watershed, ecological, and economic modeling to address concerns related to water availability (quality and quantity) and environmental and social impacts and to provide timely information to scientists, managers, and policy makers. Further, pressing water availability challenges warrant new strategies for disseminating model results that effectively educate the public and aid resource managers and policy makers. This use case will inform decisions surrounding impacts climate change and water, land, and resource management strategies may have on water availability, and environmental and economic sustainability. The initial focus is on the Upper Colorado River Basin (UCRB), although the work is easily scalable to regional and national scales.

Stakeholders include states, tribes, other Federal agencies including the Bureau of Reclamation, Fish & Wildlife, and the Bureau of Land Management. These stakeholders need information on the effects of environmental changes on a wide range of ecological and human systems such as water management including water rights, reservoir operations, hydroelectric power generation, managing wildlife habitat and water for ecosystems, allocating water for various uses and users. In the UCRB, these stakeholders have an immediate need for hydrologic information on the conditions that would triggering various elements of the Drought Continency Plan. The Salinity Control Forum is also an important stakeholder in the basin interested in information on how to manage water quality. Stakeholders also include other water users (agricultural, municipal, industrial, ecological) who use water for drinking, growing crops, and habitat. Information on water availability is important for these users to plan crops, recreation activities and businesses, and ecological habitat protection and restoration efforts.

In its more developed stage, this use case will integrate data and models of streamflow and water quality, regional climate, water use, land cover, geology, ecology, and economics to illustrate the multifaceted effects of changes in climate, water use, land use and land cover on water availability for human and ecological needs. The results will be packaged in an online interactive mapper decision support tool that will allow users to understand the effects of resource management and planning decisions on water availability, quality, ecological impact, and cost.

This work directly supports DOI, USGS, and USGS Water Resources and Ecosystems Mission Areas priorities to provide scientific information about natural resources to address societal challenges, deliver integrated observations and predictions of the future state of natural systems at regional (or national scales), and achieve sustainable management and conservation of its natural resources.

This work will improve information scope and value and provide results faster and in a more accessible and useable format than currently exists. The improved decision-making ability will allow users to make decisions that consider a wide range of potentially conflicting objectives and understand the tradeoffs of decisions. They will be able to understand the effects of decisions on multiple entities (for example, cost, tradeoffs among opposing stakeholders, etc.) and effects of changes on hydrologic, ecological, and human systems.

Existing data, science, models, and IT: Currently, long-term average models exist for streamflow, dissolved solids, total nitrogen, total phosphorous, suspended sediment. These are SPAtially Referenced Regression On Watershed attributes (SPARROW) models that cover the Southwestern U.S., including the UCRB. The most recent dissolved solids model is focused on the UCRB. There are also similar models covering the rest of the U.S. that would support national applications. SPARROW is a hybrid statistical and process-based model that estimates constituent loads in streams by linking monitoring data with information on watershed characteristics and contaminant sources, routed through a stream network. These models draw on a wide variety of existing spatial data including geology, topography, soils, land cover, land use, vegetation, human activity, climate, and water use to predict streamflow and water quality throughout the study area. Results of SPARROW models have been presented through online mappers (for example, the Southwest mapper). New developments in the RSPARROW code could also support interactive decision making (Alexander and Gorman Sanisaca, 2019). SPARROW is a useful tool for this use case because it easily accepts new data, has reduced computational costs relative to more physically-based models, is explicitly spatially referenced, has improved prediction accuracy over more physically-based models at larger scales, and is easily scalable to larger areas.

We are working on two existing USGS projects that this work aligns with. One project is applying projections of future climate and the surface water streamflow model to bracket the range of projected streamflow for the region. I am also working on a project to estimate historical time-varying salinity loads from irrigated agricultural lands throughout the UCRB using the UCRB dissolved solids model. Currently the model is a long-term average model, but it will be modified to represent temporal changes.

The USGS and other Federal agencies have developed a wide range of currently existing ecological information for the region that complements stream model results including data on wildlife, fish, invasive species, climate change refugia and adaptation science, and drought effects on ecosystems. The Bureau of Reclamation has extensive data and models to support reservoir management.

Current expertise in environmental economics exists within the USGS including the Social and Economic Analysis group in the Fort Collins Science Center, which could support incorporation of the social science and economic data and modeling aspects of the analysis.

Next steps-Modeling advancements: The next steps to developing this use case include linking the streamflow and water quality models so they could be run concurrently. This would support an integrated modeling approach to answer questions about how changes in land cover, land use, water use, climate, and other factors will jointly affect water quality and quantity. Results could then be integrated with information on regional ecology and economic data and models to understand the broader impacts of such changes on ecological and human systems.

Another advancement would be in convert long term average SPARRROW streamflow and water-quality models to work on an annual basis. This would provide more temporally-specific information and support understanding of current and future changes that may evolve with time.

Next steps-Development of an interactive mapper to support decision making: As a first step, the interactive mapper would allow the user to try out different scenarios and visualize results and quantify expected impacts. For example, the user could test how an increase in temperature would affect streamflow across the study are or in their area of interest. Depending on the modeling advancements, scenarios could become more complex. The user could download results and visualizations for their own uses.

A next step would be to integrate additional data to explore economic and ecological impacts of scenarios. The mapper would evolve into a more sophisticated decision support tool to help users to visualize and explore tradeoffs in scenario results using, for example, pareto sort techniques, a variety of plotting tools, and animations. The tool would facilitate exploring large multi-objective solution sets to aid in decision making, including the ability to optimize solutions based on stakeholder preference. In this way, the user can balance multiple conflicting goals and understand the tradeoffs associated with decisions. Such decision support tools will provide a clear idea of the optimal solutions for different combinations of objectives.

The USGS is well-positioned to initiate this Use Case and support its progression. We have extensive existing environmental data and models, and interactive mapper capabilities that support project initiation. The USGS Web Informatics and Mapping (WIM) team develops web-based tools using the latest technology to support data integration and visualization, including digital cartography, data management, data analysis, and decision support. This use case would likely require a cloud-hosting technology in order to make it easily accessible to internal and external users.

This use case could persist long-term, particularly given the anticipated high demand for information on water supply in the UCRB. In other regions, where questions of water quality or other environmental impacts may be more pressing, the use case will be relevant for answering different questions. Some of the driving input data would require updating as time progresses, science advances, and new understanding is developed, but the ability to test scenarios is real-time would allow the product to be persistently useful. Because many of the underlying data and models already exist for other geographies, the use case is easily transferrable and scalable.

Primary Contact: Olivia Miller, omiller@usgs.gov

The objective of this case study is to determine surface water bodies, including lakes and streams, that are at greater risk for trace metal contamination during wildfires with a focus on impacts to avian waterfowl and fisheries. Stakeholders for this effort include the USDA Forest Service (USFS), the National Park System (NPS), U.S. Fish and Wildlife Service (USFWS), tribal fishery programs, state fish and wildlife agencies, and municipal water suppliers. Having a holistic approach to understanding the drivers of trace metal contamination post-fire can assist these entities as well as the USGS in developing attainable post-fire monitoring goals and help to interpret current fish and wildlife toxicology data. This case study proposes to synthesize remote sensing data for fire history with records of legacy mines, atmospheric deposition, soil chemistry, water chemistry, and animal tissue studies to find linkages between fire, trace metal mobilization and ecotoxicology.

Time and funding resources limit monitoring targets that are needed to understand ecological and human health impacts from fires. Process-level information gained from studies on a small watershed scale are critical in understanding the impact of fire on trace metal mobilization from legacy mine lands and other sources to fisheries and water supply. On the larger scale proposed under the EarthMAP program, an integrated approach can be applied to identify watersheds or ecoregions that are more sensitive to trace metal mobilization from legacy mine lands and soils that have received atmospheric deposition of mercury. Input from stakeholders and land managers within the study area, including USFS resource managers, USFWS fisheries and waterfowl managers, and NPS resource and research offices, should be sought to refine natural resource management targets that are the end-goal of larger scale data evaluations.

Mercury is volatilized from forest soils during fire (Webster and others, 2016) with trace metals from legacy mine lands transported into waterways through storm runoff following fires (Murphy and others, 2020). However, a direct linkage between fire and impacts to species that rely on these waterways needs to more effectively established. Extant satellite imagery including AVRIS coverage would be used as a regional framework to determine burn history. Other spatial data available include soil data from the USGS Geochemical Landscapes project (Smith and others, 2019), National Atmospheric Deposition Data for mercury, mercury bioaccumulation data from dragonfly larvae (Eagles-Smith and others, 2020), and baseline water chemistry collected through the National Water-Quality Assessment (NAWQA) program. The Mineral Resources Program USMIN project is synthesizing information on mine locations and commodities across the United States. All of these data are valuable, but synthesizing large datasets across watersheds to regions can yield new information on challenging, integrative tasks that inform land use management decisions following wildfires.

Using a GIS platform to develop Expertise within the Geology, Geophysics, and Geochemistry Science Center (GGGSC) can carry out aspects of this work. Ray Kokaly (hyperspectral remote sensing) has worked with airborne visible/infrared imaging spectrometer (AVIRIS) datasets from prior fires, including the 1988 Yellowstone fires, the 2000 Cerro Grande Fire in New Mexico, and the 2009 Station Fire in Los Angeles County. Other information of fire extent and intensity can be provided by the Land Change Monitoring, Assessment, and Projection (LCMAP) developed in the Earth Resource Observation and Science (EROS) Center. JoAnn Holloway (biogeochemistry of trace metals) can develop the geographic extent of the proposed Use Case based on availability of adequate data coverage for fire history and burn intensity relevant to different NPS or USFS administrative units within the Rocky Mountain Region. The USMIN project led by Jeff Mauk and Carma San Juan in GGGSC has mine data coverage and GIS capacity to develop the spatial framework for these data. Maggie Goldman and Ben McGee work with visualization of data using a GIS platform, and Ben is also developing eDNA applications for detecting presence/absence of fish species. Expertise from other USGS programs can round this effort out, including aqueous geochemistry (Blaine McCleskey and Sheila Murphy, Water Resources Mission Area-Boulder), aquatic invertebrate ecotoxicology (Travis Schmidt, Wyoming-Montana Water Science Center [WSC]; Johanna Kraus, Columbia Environmental Research Center), and vertebrate ecotoxicology (Collin Eagles-Smith, Forest and Rangeland Ecosystem Science Center [FRESC]—Corvallis). David Krabbenhoft (Wisconsin WSC) is a leading expert on mercury contamination in aquatic ecosystems. Adding a component of economic loss of water quality from fire impacts through the Water Economics group at the Fort Collins Science Center [FCSC] can add an increased societal impact.

This effort does not replace USGS science in areas burned by wildfires. A regional EarthMAP effort can be augmented with smaller-scale studies in the USGS and by developing cooperative relationships with local, state, and federal land use managers.

A paleofire approach (Natalie Kehrwald, Geosciences and Environmental Change Science Center [GECSC]) to understand wildfire occurrences from the perspective of Holocene climate change in northern New Mexico can be augmented with data synthesis from the Cerro Grande fire. New samples can be collected in coordination with local water resource agencies for mercury, lead and other targeted metals in lakes that are sources for municipal and rural water supplies.

An ongoing survey in the Yellowstone National Park Research Office of loon biometrics, including lead and mercury blood chemistry can be augmented by a synoptic sampling of lake chemistry associated with the bird populations, with background information on historical fire intensity derived from AVIRIS data. Erin White, Park Hydrologist at Yellowstone National Park (YNP) has indicated an interest in coordinating YNP research interests with USGS work in relationship to a long-term monitoring of the common loon by YNP avian biologists. Environmental stressors for the common loon includes mercury and lead, which have been found in elevated concentration in YNP populations. Little work has been done of post-fire resource impacts in Yellowstone since the 1988 fire that burned 793,880 acres in the park. The capacity for lakes to act as long-term sinks for mercury released through wildfire and the availability of an immediate partnership with the YNP research office, and the public visibility of study outcomes would make the greater YNP ecosystem a good testing ground for an EarthMAP approach to interrogating a complex regional problem with a multidiscipline approach.

Increased data storage and computational resources will be needed for the synthesis of geospatial datasets, particularly at the scale of the Rocky Mountain Region. In the short term, it may be a more attainable goal to form partnership with different NPS and USFS units.

The changing climate regime for the western United States is yielding more persistent drought conditions, increased frequency of fires, and a longer fire season. Providing data synthesis approaches to understanding areas more at risk for water quality and ecosystem impacts from trace metal transport will provide a durable input for prioritizing monitoring sites and post-fire mitigation of risks to human and environmental health. This Use Case proposes to develop approaches to understanding geographic patterns of risk associated with wildfire. The approach can be scaled up to understand regional impacts or scaled down to interrogate process-level trace metal mobilization on a watershed scale.

Primary Contact: JoAnn Holloway, jholloway@usgs.gov

Numerous studies have observed increasing solute concentrations in mountain headwater catchments around the world that appear to be due to warming temperatures and shifting precipitation regimes caused by global climate change. Watersheds containing abundant sulfides (mineralized watersheds) may be particularly susceptible to these changes because of the sensitivity of sulfide weathering to subsurface oxygen transport and redox conditions, which are directly linked to climate-controlled factors such as groundwater recharge rate, water table depth, and extent of frozen ground. This susceptibility raises a significant concern regarding potential increases in future stream metal loads in mineralized headwaters, which could negatively impact stream ecosystems and downstream water resources. Mine tunnels and exposed waste piles from historical mining activities could further exacerbate the problem, as acid drainage from these features could be yet more sensitive to climate variability than natural background acid-rock drainage. The UCRB has many mountain headwater catchments containing extensive mineralization and legacy mining features. A possible Use Case for EarthMAP is therefore the development of a regional-scale modelling tool for predicting future water-quality changes in mineralized headwater catchments in the UCRB that could be useful to state and federal agencies who manage water resources, fish populations, and abandoned mine site cleanup efforts.

Two projects cooperatively funded by the USGS Mineral Resources Program and the Department of Energy’s Subsurface Biogeochemical Research Program are currently underway that focus on water-quality issues in mineralized headwater catchments in Colorado. This work is being performed collaboratively with scientists in the Colorado Water Science Center, with additional support from the USGS Toxics Program. The first project focuses on improving understanding of processes controlling acid-rock drainage generation and metal transport in mineralized mountain watersheds, and the sensitivity of these to climate warming. The second focuses on developing science-based approaches for screening legacy mine sites to aid in the process of prioritizing sites for remediation work. These projects have high stakeholder interest, involving active engagement with representatives of the USEPA Region 8, Colorado Division of Reclamation, Mining, and Safety, and with Trout Unlimited. We fully anticipate that the Use Case work proposed below will have the same high level of stakeholder interest, will be performed mostly by the same group of GGGSC and Colorado WSC scientists working on these current projects, and will also be considered priority work by the Mineral Resources and Toxics Programs.

A regional-scale modelling tool capable of indicating the likelihood of future water-quality degradation in different mineralized headwater catchments in the UCRB could be particularly beneficial for three groups of agencies. The first group includes agencies tasked with managing stream fisheries which are vital for tourism, primarily the U.S. Fish and Wildlife Service and, in Colorado, the Colorado Parks and Wildlife Department. Knowledge of which stream networks are more or less threatened by future increases in metal loads could help considerably in decisions regarding where to focus stocking and habitat improvement efforts to develop the most robust and sustainable populations. The second group includes agencies tasked with managing abandoned mine site cleanup efforts, primarily the USEPA and, in Colorado, the Colorado Division of Reclamation, Mining, and Safety. Knowledge of which mining-impacted watersheds may experience the most pronounced additional water-quality degradation in the future would be highly valuable in the process of deciding where to invest new or enhanced site clean-up efforts to maximize benefit to the environment. Furthermore, an awareness of the potential for increasing natural background metal concentration in watersheds where remedial work is underway is important in the process of establishing attainable clean-up standards, and developing plans for adaptable standards at sites where background concentrations are likely to be a moving target. The third group is municipal water resources managers in mountain and mountain-adjacent communities, where surface water is often the primary source for public supply. Knowledge of which catchments in their source stream networks may experience water-quality degradation, and to what degree, would help water managers anticipate future required modifications to water intake locations and treatment systems, and allow them to plan accordingly.

Two fundamentally different types of models could potentially be developed for the purpose of predicting changes in surface water quality. The first is a regional-scale numerical reactive solute transport model, but these complex process-based models require extensive parameter inputs and computational power that will likely limit their application beyond the small-watershed-scale for many years to come. The second is a much simpler empirical model that would utilize a set of observed correlations between mapped watershed physical characteristics and measured stream water quality changes (disregarding the ‘why’) to predict the latter when only the former is known. To our knowledge, a regional-scale empirical model for this application has not been developed and is thus aspirational. However, we believe that our current understanding of the factors controlling sulfide weathering, available map coverages, and available statistical tools for deriving necessary correlations are all sufficient to conceivably develop an empirical model in a few years.

We envision the following general approach for developing such an empirical model. Central Colorado could be used as a model ‘training ground’ given that a GIS model has already been built for the Central Colorado Assessment and Legacy Mine Lands projects (CCAP and LML), containing most of the required data coverages (for example, bedrock lithology, hydrothermal alteration mineral mapping from hyperspectral data, and mine features from USMIN). A comprehensive sampling program for stream water chemistry would be undertaken throughout the region, perhaps in cooperation with the Colorado WSC. Targets would include all mineralized watersheds where historical water chemistry data exist, including the numerous sites sampled ~15 years ago for CCAP. Changes in metal concentrations over recent years could then be assessed and compared to watershed physical features to identify correlations using principle component analysis and other advanced statistical tools. Artificial intelligence and machine learning could also be very useful for this purpose. The next phase of the work would involve expanding the GIS model to other parts of the UCRB where historical stream water-quality data are available, collecting new data from these sites, and testing the model’s capability to predict observed changes. Finally, if the model passes these tests, it could be expanded to the entire UCRB. Regarding partnerships with stakeholder agencies, these are already well established through our current project work and could be directly utilized to in the proposed Use Case for obtaining stakeholder input and adapting the model tool to best meet stakeholder needs.

It should be understood that success for this effort is admittedly uncertain, with it possibly yielding insufficiently robust correlations to enable development of an empirical model. However, the water chemistry data alone would still be highly valuable to establish regional baselines and trends, and to assess the regional extent of recent surface water-quality deterioration in mineralized watersheds to gauge the scale of the potential problem. Furthermore, the cost of this effort would be relatively low, heavily leveraging prior CCAP and LML project work and requiring only the collection of standard water chemistry data and application of standard GIS and statistical modeling tools. Regardless, this approach clearly presents the most tractable path forward if the development of a predictive capability for water quality on the scale of the UCRB is determined to be a priority for EarthMAP.

The research team from the GGGSC and Colorado WSC working on the two current USGS projects described above possesses most of the scientific capacity needed to implement the proposed Use Case. One key gap is the need for additional expertise with applying statistical analysis to GIS models, particularly if developing correlations is more challenging than anticipated and demands the application of AI/ML. Another operational need is additional staff to collect the extensive set of surface water-quality samples throughout Central Colorado, though, as noted above, this short-term need could possibly be met by the Colorado WSC, or perhaps temporary student hires. As far as the utilization of the empirical model-based tool by stakeholders once developed, we envision that this would require only minor initial training because it would function on a GIS platform with which stakeholders already have considerable experience.

The proposed Use Case should have considerable long-term relevance because climate prediction models consistently indicate that the climate will continue to warm through the 21st century, meaning that water quality degradation in mineralized watersheds poses an ongoing challenge that could persist for many decades. The empirical water-quality prediction tool we propose should thus also be applicable and useful for many decades. However, it would certainly be yet more useful if it were a living model that is updated and improved through time. A monitoring program involving ongoing collection of surface water chemistry samples from a select set of sites would enable continued testing of the model’s accuracy, and the statistical exercise of deriving correlations between observed water-quality changes and watershed characteristics could be relatively easily re-run periodically to refine and improve these correlations. In this way, the model would likely become ever more accurate and useful through time. The transferability of the UCRB-specific empirical model we develop to other regions is uncertain, presumably limited by the degree to which the other regions differ climatically and geologically from the UCRB. However, the process of empirical model development we propose should be entirely transferable to other regions and enable the creation of other region-specific empirical models for other mountainous areas of the U.S.

Primary Contact: Andrew Manning, amanning@usgs.gov

Traditional USGS oil and gas assessments provide estimates of the total undiscovered technically recoverable petroleum resource in a given region. While that information that is used by a wide range of decision makers in the public and private sectors, increasingly those stakeholders are also interested in more specific information regarding which areas within a given study region have greater potential for oil and gas development. Interested stakeholders include entities responsible for resource management such as state and federal land and wildlife managers, and individuals and groups with varied interests in resource use such as recreational interests, private property owners, and wildlife advocates. In addition, increasing emphasis in much of the western U.S. is focused on other energy resources, including geothermal energy and renewable resources such as wind and solar power, presenting a similar set of questions, as well as the potential complexity of conflict or synergy between energy resource development and existing uses and benefits provided by the landscape.

Accurately predicting energy resource development activity in space and in time is not a realistic goal, as actual energy industry activities depend on a wide array of economic, regulatory, and social factors in addition to geologic, environmental, and a suite of logistical considerations. However, it is possible to develop an approach that identifies locations where energy development is more likely based on a set of predictor variables, in order to address stakeholder questions on resource development potential and desirability. The task of addressing this stakeholder need will require synthesis of energy resource information (geologic and petroleum-related, as well as solar and wind potential), along with consideration of existing uses, ecological attributes, terrain, roadways, infrastructure, and associated factors. The result will not be a prediction of specific development that will occur, but rather a set of mapped variables that can be used to estimate the relative potential and suitability for resource development. This information will be useful to inform decisions regarding the optimal management of public lands.

The development of a decision support tool that addresses energy development potential will draw on the science products and expertise from multiple USGS programs and Mission Areas and collaborations with the National Renewable Energy Laboratory. Though the participating science centers are largely based in the USGS Rocky Mountain Region, the topic of interest is national in scope and potentially involves all USGS regions. The demand for this sort of decision support tool is long-standing, and interest in these products will likely increase as land-use decisions and energy resource concerns continue to gain prominence. Development of USGS products to address these needs will be a long-term activity, with the near-term objective of addressing immediate stakeholder interest, followed by opportunity to hone these products and improve their applicability to stakeholder decision-making needs.

Due to the challenges of estimating energy resource development potential in a meaningful manner, current tools are limited, and no USGS science directly addresses this question. As such, the development of a scientifically based decision-support tool that delivers information on development potential, including co-located resources and associated tradeoffs, will be an improvement. By synthesizing relevant science and information, the USGS can provide “actionable intelligence” that stakeholders can confidently rely on for use in their planning and decision making. Because the USGS is the recognized authority on many of the relevant scientific components, we are the logical organization to synthesize this information; we are best equipped to understand both the capabilities and the limitations of these data.

The myriad factors that influence energy resource development patterns, and the complex interplay between them, dictates that a decision-support tool will only be useful if developed through close interaction with stakeholders to ensure that their needs are being addressed, on timeframes that are useful to them, and to effectively communicate both capabilities and limitations of the analyses. Through interaction and communication between scientists and decision makers, we can ensure that stakeholder desires are met and that the decision support product honors scientific limitations.

The objective of this work is to address stakeholder questions regarding development potential for energy resources of all kinds, with initial focus on oil, gas, solar, and wind. The work draws on data and expertise from across much of the USGS, and from outside sources.

The USGS Energy Resources Program (ERP) produces oil and gas resource assessments that include a wealth of geologic and petroleum related data and analysis in addition to the assessment results themselves. This information includes geologic maps and cross sections, petroleum production data, and expert analysis of these data. In addition, the ERP also studies patterns of water use and water production, yielding another set of data for decision-support questions where water-related topics are relevant. Together, this information provides a strong foundation for the petroleum-related parts of the synthesis work.

For solar and wind energy, USGS will collaborate with the National Renewable Energy Laboratory (NREL) to better parameterize their existing energy development tools. NREL has developed two models to forecast energy development. First, Regional Energy Deployment System (ReEDS) forecasts national energy demand and the “mix” of energy generation meeting this demand across a wide array of future scenarios. Second, for solar and wind energy generation, the Renewable Energy Potential (reV) Model, includes wind and solar resource maps, localized costs of energy generation, system performance as estimated capacity factors, and spatial exclusions based on socio-political and land use constraints, to estimate the amount of annual technical potential (MWh/km²) for wind turbines, photovoltaic solar, and concentrated solar production. USGS will contribute refined land-use and land-cover information, and improved information on environmental constraints (for example, sage grouse core areas), to allow more refined predictions of renewable energy potential in specific regions. In the long term, the spatial constraint algorithms in reV could be modified and applied in USGS Assessments Units when considering oil and gas development potential.

Predicting energy resource development potential further requires information on the social-ecological system(s) in which it will be embedded. Factors such as water source and use areas, key wildlife habitat areas, protected areas, transmission infrastructure, and established socio-economic uses such as recreation, aesthetic enjoyment, water-quality regulation, and others are all key to establishing the existing use of a landscape and hence the energy resource development potential. A variety of USGS programs have expertise in social-ecological systems, including Land Change Science, Science Decisions Center, Energy and Wildlife, Water Availability and Use, and Social & Economic Analysis Branch. These groups will work together to characterize the social and ecological dimensions that must be weighed against resource potential to identify the tradeoffs and evaluate development potential. Stakeholder input will be used to establish the weighting schemes.

Synthesizing and integrating the data described above in a meaningful manner represents a largely new scientific direction for the USGS, and it expands the kinds of contributions EarthMAP can make to those who rely on USGS for decision-relevant science. It will require development of new concepts for linking data and new ways to communicate possible future energy development and its consequences to stakeholders. It will also include development, through discussion with decision-makers, of definitions for “resource development potential,” to address their particular concerns and interests. For example, some decision makers might want to know which areas are most favorable for oil, gas and wind power development based on wind, geologic and infrastructure considerations, whereas other decision makers in other areas might be interested to consider areas most favorable for gas and (or) solar development while minimizing impact to particular species of interest and (or) existing uses of public land. We envision a decision-support tool that has many “knobs” that can be used to adjust relative weighting of different map data, a tool that is mainly useful through thorough communication between scientists and decision makers.

This Use Case draws on a wealth of existing USGS and external information, but at the same time represents a new science direction. Determining the variables most important for resource development potential, and the relationships between them, will build from existing capabilities within the USGS and will require contributions from scientists in several USGS programs as described above. Specifically, the development of synthesis approaches for these diverse data, to provide a decision-support tool that is adjustable to the needs of a range of foreseeable decisions, will be an exciting new challenge.

Complementing USGS expertise, there will likely be a need for input on a few additional aspects of the work. For example, colleagues at the National Renewable Energy Laboratory are well on their way to an EarthMAP-like synthesis framework for renewables but need USGS expertise in wildlife ecology and land change science to better inform their geospatial constraints models. Regarding oil and gas development, we will need experts who can provide guidance regarding logistical aspects of oil and gas development, including drilling engineering and transmission/transportation details.

Effective delivery of the resulting decision-support tool, and handling of the underlying data, will most likely be possible using existing web-based tools, possibly with need for some expansion of existing capabilities. It is not anticipated that the tool will require high computational power (NRELs REEDS/reVs runs on their supercomputer), but data synthesis and effective delivery is an important component of this work.

Energy resource development has long been a topic of interest for land management agencies and this is unlikely to change in the foreseeable future. Emphasis on particular energy resource types may vary over time, however, and this is a reason to include not just oil and gas but also solar and wind development potential in this Use Case. The need for greater information on resource development tradeoffs is ongoing both during planning stages of energy resource development and throughout the development process.

The decision-support tool that we describe herein is not limited to any particular geographic area, and its utility is anticipated to be essentially nation-wide (any area with energy resource development potential). By nature, a tradeoff exists with scale, where greater spatial specificity comes with greater uncertainty; that is, when applied at larger scales this decision support tool can be anticipated to provide more meaningful results. Expansion of the developed approaches to include additional energy resource types is feasible, though each resource type will require additional considerations related to the particular geologic, ecological, social, and other factors that control resource development likelihood and associated tradeoffs.

Primary Contact: Seth Haines, shaines@usgs.gov; Darius Semmens, dsemmens@usgs.gov; Jay Diffendorfer, jediffendorfer@usgs.gov; Karen Jenni, kjenni@usgs.gov

Stakeholders require data and tools to characterize and assess effects of past and current resource development activities at a variety of spatial and temporal scales, and to address future effects. Quantitative forecasts and predictions of supply, demand, and effects of development on resources are needed and used by stakeholders to evaluate resource availability, develop land and resource management plans, and assess environmental and economic effects of resource development.

This Use Case will inform decisions such as determination of the environmental impacts of Federal agency actions under the National Environmental Policy Act (NEPA); Federal, Tribal, and State decisions regarding recovery and management of threatened and endangered species under the Endangered Species Act (ESA) and related state laws and regulations; Federal, Tribal, and State management of land, forestry, and energy resources, among others. Relevant decision-makers, stakeholders, and intended information recipients; their interests in and need for improved science information. The pilot Resource Assessment Platform (RAP) would have immediate application to sister agencies of the USGS, and a fully developed version could be of use to stakeholders at many levels including:

Information synthesis necessary to provide actionable resource management intelligence requires seamlessly integrating existing data and models. Tools would be developed to allow users to synthesize and deliver those data of interest to them at speeds appropriate for timely decision making. Pilot-scale implementation of this Use Case in the Colorado River Basin would be related to and supported by current USGS initiatives including implementation of the USGS Water Resources Mission Area’s Next Generation Water Observing Systems (NGWOS) and Integrated Water Availability Assessments (IWAAs) priorities. Several Department of Interior (DOI) agencies have priority initiatives in the Basin, so the pilot would also be designed to support DOI agency activities. Once demonstrated at the pilot scale, the RAP could be extended geographically to other regions or the nation, updated periodically or in real time with updated input data and machine learning, and expanded to support additional resource management decision-making needs.

The RAP will provide stakeholders a single location to support evaluation of various decisions such as prioritizing remediation activities and assessment of future development needs and potential impacts quickly and confidently knowing that the information and tools are the most current available.

The ability of stakeholders to evaluate multiple resources simultaneously will be improved, so that decisions will better reflect integrated conditions and impacts. Availability of a comprehensive, central source of data and tools sufficient for a wide variety of resource assessments will improve timeliness of decision delivery. Opportunities to work directly with decision-makers to define their integrated science needs and understand and meet their needs for science delivery. Scientists will need to work closely with decision-makers to better understand assessments of interest. This ongoing, direct interaction with decision-makers will enhance the usefulness of the platform and ensure it remains relevant as partner needs and priorities evolve. Expected impact of the improved decision or resulting management-action. Because this platform can be modified to focus on variables of interest, stakeholders will better be able to make decisions based on the variables of interest to them. This will have the benefit of reducing the duration and costs of resource assessments, improving timeliness and enhancing the consistency of management decision making.

This project would take advantage of existing USGS scientific and technological capacity in water resources, climate, land cover and land use, geology, mineral and energy resources, species and ecosystems, modeling and parameter estimation expertise, the Advanced Computing Cooperative, 3-DEP, and other capacities. A wide variety of existing scientific, modeling, remote sensing, and computational capacities, both internal and external, could also support the project. This platform would require collaboration between USGS water, ecology, geology, mineral and energy, and social science centers sharing data and expertise. The Colorado River Basin pilot application would be integrated with ongoing USGS NGWOS/IWAAs efforts, and leverage previous work under the National Streamflow Information Program (NSIP), Regional Aquifer-System Analysis Program (RASA), NAWQA and FAS.

For the initial pilot application existing USGS datasets include: National Water Information System (NWIS), Simplified Surface Energy Balance (SSEBOP), EPA’s STORET, USGS National Cooperative Geologic Mapping Program, USGS Geolog Locator, Weather datasets—PRISM and DayMet, USGS National Land Cover Database, LANDFIRE, USDA STATSGO, USDA-Natural Resources Conservation Service (NRCS) SNOTEL (SNOwpack TELemetry), USGS 3D Elevation Program (3DEP), etc. Numerous partner datasets and models also exist and are available for synthesis. Existing and potential partners include: other USGS Science Centers, Federal agencies such as BOR, BLM, NPS, USFS, USFWS, USACE, State land, water, forest, energy and other natural resource management agencies, universities, local and municipal users of resources, and other stakeholders such as nongovernmental organizations.

The data and modeling platforms needed for many resource assessments already exist, both within USGS and with external partners. The challenge would be linking these tools and providing the interactive platform and delivery mechanisms for decision makers. The approach would be a cloud-based interactive dashboard or integrated system of data layers and modeling capabilities. Something like Tableau might be useable or adaptable to meet the objectives of this Use Case, or a new type of platform may need to be developed. Agreements would have to be reached among the agencies to enable their legacy data to be easily pulled into the RAP for use and analysis. To be useful to decision-makers the RAP would need to be reliable, easy to use, with well-documented data lineage and interactive data-delivery mechanisms.

Partnerships with IT experts and programmers in the Advanced Computing Cooperative, the Center for Integrated Data Analytics, and the Oklahoma-Texas WSC would be needed. For the Colorado River Basin DOI pilot, close collaboration with the other DOI agencies will of course be essential. This Use Case will require high-level IT and related expertise to develop, deliver and maintain the RAP, including programming expertise and data scientists. Artificial intelligence and machine learning technology and expertise may be required to create a platform that improves functionality through use.

Once the initial framework is established, longevity and sustainability are maintained by a constant process of incorporating newly available data sources, and expanding the ways in which the data sources can be synthesized and interrogated to address emerging needs. The relevance of this Use Case will expand over time with the addition of newly available datasets and tools, and the organic adaptation of the framework and capabilities to address emerging needs. Once demonstrated at a pilot scale, the geographic and administrative scope could be expanded to cover additional geographic areas and decision-making needs. The usefulness of the RAP as a decision support toolbox will be enhanced by support from persons capable of adapting the framework to support emerging stakeholder needs as they become identified.

Successful development of the RAP will be enhanced by limiting the initial scope to a regional pilot scale. The ultimate vision is to develop an interconnected web of regional tools that allow interrogation across and between regions, ultimately scaling up to handle national level questions. Information that will be used to evaluate the effectiveness of this effort and define metrics of performance and success. Effectiveness of the effort will be evaluated by the ability of the RAP to deliver information in a way that addresses stakeholder needs, and that supports timely decision making on a variety of issues. Other problems, at different scales, different timeframes, and or in different geographies, will benefit from the science and technologies developed for this Use Case. Expanding the scope beyond the regional pilot scale will expand the ability of the RAP to address problems at expanding scales and differing geographies. Incorporation of longer timeseries datasets and AI/machine learning will allow questions to be asked at a point scale, delineate past trends, forecast into the future, and expand RAP functionality.

Primary Contact: Michael Johnson, msjohnson@usgs.gov, Anne Tillery, atillery@usgs.gov

Hurricane Katrina in 2005 and the Deepwater Horizon oil spill in 2010 are two major events that have affected habitats and natural resources on Dauphin Island, Alabama. The latter event provided hundreds of millions of dollars to Alabama for coastal restoration and prompted a collaborative effort between the USGS’s Coastal and Marine Hazards and Resource Program (CMHRP) and Ecosystem and USGS Water Resources Mission Areas, the USACE, and the State of Alabama, funded by the National Fish and Wildlife Foundation, to investigate viable, sustainable restoration measures that reduce degradation and enhance the natural resources of Dauphin Island. The overarching goal of the ALBIRA was to document present conditions and forecast potential conditions under varying sea-level change and storm scenarios for a no-action alternative and for a variety of restoration measures including beach and dune restoration, marsh and back-barrier restoration, and placement of sand in the littoral zone. Scoping sessions with the public and expert elicitation engagement from academia, state and federal partners, and nongovernmental organizations helped communicate climate-related threats to Dauphin Island and identify shared objectives that maximize the sustainability of the system. The integrated science and synthesis components incorporated into the project to address the goals and objectives include: (1) conceptual ecological model development; (2) system-level data acquisitions; (3) integrated hydrodynamic, geomorphic, water quality, and habitat modeling on decadal timescales; (4) evaluation of restoration alternatives using structured decision making; (5) monitoring and adaptive management plan development; and (6) online data catalog and delivery using a web map.

Although this study was externally funded, this effort aligns with the priorities of the USGS Mission Areas and programs discussed above. For example, the study aligned with the Ecosystem Mission Area’s Land Change Science Program’s main goal of developing a comprehensive understanding essential for managers and policy makers on topics such as: (1) rates, causes, and impacts of sea-level rise; (2) changes in distribution of plant and animal communities; and (3) causes and consequences of land-cover change. Additionally, the effort aligned with the USGS CMHRP Level 2 requirements: (CM2.3) “…provide actionable information and reliable guidance on coastal change hazards processes and impacts and, in alignment with the USGS Risk plan, identify and communicate the vulnerability to potential hazards in relation to user needs”; and (CM3.3) “…provide integrated models and forecasts of impacts of episodic and extreme coastal processes, climate change, human disturbance and restoration activities on estuaries, wetlands and reefs.”

The models developed for this project can be extended to predict impacts across different time scales and different management applications. This could include understanding impacts from a contemporary storm through hindcast analysis (for example, understanding the vulnerability of a barrier island and its habitats to coastal change), real-time forecasting (for example, predicting where overwash or breaching may occur, which could inform emergency management decision making such as road access), and post-storm recovery assessments (for example, identifying changes in sand movement and the ability of a barrier island and its habitats to recover to a pre-storm state). In addition to the management actions considered in this project, actions such as creating vegetation features or implementing living shorelines could also be assessed. The science and products developed from this effort are scalable to other barrier islands with local-level calibration, as well as broader areas to provide a more regional picture of coastal impacts. This is currently being accomplished through synergistic USGS projects focused on nearby regions in the northern Gulf of Mexico.

Restoration measures for this effort were developed to address multiple objectives related to the social and ecological concerns of stakeholders on Dauphin Island and surrounding areas. The stakeholder defined objectives were to: (1) maximize ecological function and physical processes (that is, sustainability); (2) minimize social impacts and costs; (3) maximize coastal and marine resources; and (4) minimize time that it would take for a restoration action to provide benefits for the island. In all, 23 restoration and land acquisition measures were developed and assessed relative to their ability to meet these objectives on Dauphin Island. Bayesian Belief Networks—probability-based decision tools—were used to evaluate restoration measures that best met these competing objectives. This objective-centered approach to decision making explicitly incorporated the human dimension to help assess trade-offs among restoration measures. Working with our decision-makers from Alabama, a storymap was developed to communicate the results of the restoration assessment to the citizens and other stakeholders of Dauphin Island. The State of Alabama used the findings of our assessment to target over $150 million worth of restoration measures for implementation.

This project relied on existing USGS expertise in field data collection, predictive coastal morphological modeling, habitat mapping and modeling, and decision analysis to address management decision making and stakeholder needs at Dauphin Island (fig. 5.11). Guided by a conceptual ecosystem model developed for Dauphin Island for this study, field data were collected to gain an understanding of the island’s historical evolution as well as the physical, topographic, bathymetric, geologic, and oceanographic setting to help maximize the likelihood of restoration success. These factors have not only played an important role in understanding how the island has evolved over time but will regulate how it will respond in the future. Field data collected during this study included: (1) bathymetric and geologic surveys; (2) wave and current measurements; (3) water-quality data; (4) sediment distribution information; and (5) habitat data. This information was used to update baseline conditions and provide a primary source of data for model development and validation. Forecasting barrier island evolution provides decision-makers the ability to assess the resiliency of coastal environments to future climate change conditions. The suite of numerical models developed for Dauphin Island provided a quantitative understanding of the physical processes governing the past and present Dauphin Island barrier system, including the nearshore region adjacent to the barrier island complex. Morphologic models were used to describe barrier island evolution accounting for fair-weather conditions, storms, and post-storm recovery. The models were used to assess potential future states of the island considering changes in sea level and storm variability (frequency and intensity of tropical cyclones) over a decadal period. Habitats were predicted using a model that leveraged landscape position linkages (for example, proximity to shoreline, elevation, and relative topography) for barrier island habitats. A spatially explicit habitat suitability model was developed for nearby estuarine waters for oyster and seagrass habitats. Together, the predictive morphological and habitat models were used to assess the previously mentioned management actions developed by stakeholders through structured decision making to gauge their ability to meet multiple objectives for habitats, sustainability, residents, and conservation values.

This effort could be expanded by increasing our understanding of barrier island vegetation recovery and succession, including overwash response and marsh productivity and drowning, in response to sea-level rise and reoccurring storms. Insight from these analyses can be used to better calibrate models used in this study and lead to more thematic detail in habitat model outputs. In this study, we used expert elicitation to discern habitat importance to species of concern. Future work is needed to develop models that can be used to more explicitly understand how barrier island changes may impact important fauna (for example, shorebirds and sea turtles) that utilize these islands. Additional enhancements are needed and discussed more in the next section to increase the efficiency of data exchange between models and dissemination of outputs. Collectively, these enhancements could lead to more robust and efficient models that could be used to make operational decisions related to restoration, such as: (1) quickly running different scenarios; (2) addressing new morphologic change (for example, a newly formed breach); and (3) re-weighting stakeholder values in risk assessment.

The ALBIRA included a multidisciplinary, integrated science team that can support the planning, implementation, monitoring and adaptive management of barrier island restoration alternatives in Alabama. USGS also maintains the data storage, access and delivery of datasets and products to internal and external partners via a project site. For future efforts, there are opportunities to enhance data exchange workflows and model interoperability between modeling teams to increase the feasibility of operational implementation. For example, information exchange, particularly exchange between geomorphic and habitat models, could be enhanced through increased data interoperability and the use of open-source software and (or) developing scripts and tools to process information exchange between various model components. Additionally, the use of High-Performance Computing (HPC) would expedite numerical model simulations and allow for expansion to broader spatial scales and consideration of additional scenarios. As this analysis is expanded out to integrate more real-time analyses, tools could be developed to efficiently integrate and synthesize in situ observations, predictions of total water level and coastal change, and remote sensing data. Cloud services may be helpful for facilitating a more streamlined data exchange and decrease the latency of data product visualization when combined with custom open-source tools. This investment would not only help the science become operational but may also allow for more feedbacks between models (for example, development of post-storm vegetation recovery predictions for more accurate inputs for morphologic models).

Coastal managers and decision-makers will continue to need scientific data regarding the potential impact of future coastal hazards (for example, extreme storms and sea-level rise) to make informed decisions for managing human and natural communities. Within the Gulf of Mexico, over $4.2 billion in DWH funding has been allocated for the restoration of wetland, coastal and nearshore habitats, which our science can inform. This science requires integrated, multidisciplinary studies that incorporate various factors that impact coastal systems at local scales. The models developed for this study can be used to assess the present state and potential future conditions of barrier islands elsewhere. This could include assessments of the effect of future storms, sea-level rise, land use/land cover changes, and restoration alternatives such as hard structures or natural and nature-based features. Finally, the future-focused nature of this effort could be coupled with model outputs from assessments or predictions of the impacts of contemporary extreme storms for a full picture of the dynamic nature of coastal areas.

Primary Contact: Davina Passeri, dpasseri@usgs.gov; Nicholas Enwright, enwrightn@usgs.gov

Stakeholders are central to the Mississippi Alluvial Plain Water Availability project and have an active role in developing each year’s workplan and scientific focus as well as assisting with community outreach efforts. Examples of local, state and federal stakeholders include Arkansas Natural Resources Department, Arkansas Department of Health, Arkansas Department of Environmental Quality, Louisiana Department of Transportation and Development, Louisiana Department of Environmental Quality, Mississippi Department of Environmental Quality, Delta Council, Delta F.A.R.M., Delta Wildlife, Yazoo Mississippi Delta Join Water Management District, Missouri Department of Natural Resources, U.S. Army Corp of Engineers (USACE), U.S. Department of Agriculture, The Nature Conservancy, University of Arkansas, University of Mississippi, Mississippi State University, Louisiana State University, Memphis State University. Through this intentional coproduction of information, the Mississippi Alluvial Plain Water Availability project has produced several examples of actionable science including; (1) collaboration with Hazards to provide data to evaluate the New Madrid seismic zone; (2) predictive modeling to inform a groundwater transfer injection project led by USDA and USACE; (3) development of a regional high frequency water use monitoring network with multiple partners; (4) working with the USACE to assess hydrologic infrastructure and ecological habitat and sediment transport in the Mississippi River; (5) optimization of surface and groundwater monitoring networks; and (6) coupling of hydrologic and economic models to inform stakeholders prioritizing water availability projects for the region.

Integrated science and information synthesis is a necessary component of the MAP project and includes the following:

The MAP project directly addresses known stakeholder priorities through co-production of information (iterative product development). The MAP project then can address Program, Mission Area, bureau, departmental, and administration priorities by aligning and leveraging these priorities to meet Stakeholder needs. Examples include: 1.) the USGS Water Resources Mission Area (WMA) has a priority to predict water use at a daily time scale. This is also a priority to MAP stakeholders and can therefore be leveraged to meet both USGS and Stakeholder needs; and, 2.) DOI has a goal for the USGS to “Apply science to land, water, and species management”; the MAP project is directly addressing and advancing this goal with their Stakeholders, benefiting both DOI and local stakeholders.

Modeling, analysis, and prediction in one of the most stressed hydrologic systems in the US demands a holistic approach that is scalable and efficient (near real-time science for near real-time decisions). For example, MODFLOW 6 allows for regional and local inset models to meet different needs (demands) related to difference scales and areas.

The MAP project has and continues to significantly improve stakeholder decision make via co-production of information and iterative science, allowing for improved investment in alternative water supply options and improved metrics for evaluating water availability in terms of current availability and changes in availability over time and space. Co-production of information with stakeholders, which involves describing problems and questions with stakeholders as the project is being proposed and developed requires frequent and regular engagement with stakeholders to go over preliminary models and output in order to ensure we are doing science with our partners and not to them. Additionally, the MAP project works with stakeholders to develop questions and scenarios that stakeholders want to understand to make decisions to secure future water resources (for example, how much will water availability decrease if we increase water use? Will water availability increase if we switch to alternative water resources? If we keep using water from this location, will we see saltwater encroachment?).

The MAP project is employing several 21st-century science capacities to achieve this goal; including the use of machine learning models for geology, groundwater, water-use, water quality, and streamflow; remotely sensed satellite and airborne geophysical data acquisition to inform and validate hydrologic models, integrated predictive modeling for scenario analyses and optimization including economic models, and decision support tools to help distill modeling and monitoring output to our stakeholders and the public.

Examples of existing foundational science capabilities include:

Advancement of science capabilities include:

Potential partnerships include: 1.) integrating state water-quality datasets (formerly done thru National Water Quality Portal [NWQP]), many states are not able to export data to the NWQP so there is a data lag between collection and consuming data into USGS models; and 2.) partnering with USACE and USDA on water availability issues, bringing all strengths to the table to provide the best information and products to stakeholders in a timely and economically efficient manner.

Science base is currently being used to help with delivery of products and serving datasets both internally (to USGS team members) and externally (to serve data to cooperators); stability of DOI links makes this achievable.

Internal:

External:

The MAP project is designed for long-term use, as it employs an iterative approach to provide the best modeling and monitoring information in near real-time while also continually working to optimize data collection and models. This requires multiple partners, funding sources, and engaged stakeholders. Since the MAP project addresses water availability in a holistic manner including stakeholders at the beginning of the effort, transferability to other areas or applications will be possible and is actively occurring; examples include: 1.) scripted workflows for pulling, processing, and serving datasets and models, which can be updated on the fly as new data are available (groundwater levels, water quality); 2.) Inset model tool for modeling water availability in higher resolution from regional-scale models; 3.) working towards an open-source workflow for airborne electromagnetic data, from acquisition/archiving to presentation and model integration; and 4.) the relationships built between irrigation withdrawal timing and quantity with drivers/triggers has transferability to regions of similar driver/trigger/water use needs.

Information that will be used to evaluate the effectiveness of this effort and define metrics of performance and success include:

Primary Contact: J.R. Rigby, jrigby@usgs.gov

Devastating wildfires and their after-effects are challenging the resources of local and state agencies in the west. These agencies must use their scare resources to protect lives and water resources from post-wildfire debris flows, runoff and sediment pollution from winter storms. The Bureau of Reclamation (BoR), U.S. Army Corps of Engineers (USACE), California Department of Water Resources (CA DWR), local utilities (PG&E), Burned Area Emergency Response (BAER) teams and many county water management agencies have identified multiple needs for short- and long-term predictions of post-fire responses. These needs include:

This Use Case will leverage the strengths of multiple USGS Science Centers and existing funding from Private and Public sources to build new partnerships to solve these post-wildfire challenges. Working together and with local, state, federal, and private partners, USGS scientists in the California Water Science Center (CAWSC), the Geology, Minerals, Energy, and Geophysics Science Center (GMEG), and the Western Ecological Research Center (WERC) are already in an excellent position to provide data and information needed to answer these questions. The USGS team has close working relationships with the partners in this Use Case and fully anticipates a co-development process.

Currently local county agencies work with the BAER team to evaluate landscape response to burned areas and help in determining areas that are at risk of flooding, sediment pollution or debris flows. Local agencies then make decisions on how to protect infrastructure, inform local communities, and respond if a flood or debris flow takes place. Most local and state authorities don’t have the expertise or tools available to monitor, forecast or predict what will happen in their watersheds. One of the few sources of real-time flood prediction is provided by the National Weather Service (NWS) at a very limited number of USGS streamgages (fig. 5.14). Unfortunately, the models used in these predictions do not incorporate changes in the landscape due to wildfire. Burn-severity, for example, is an important factor since highly burned areas can produce much more intense and damaging flooding. Post-wildfire flood predictions are also not readily available for many local agencies and may rely too heavily on outdated hydrological data or science. For example, in California, the BAER team frequently relies on post-wildfire flood predictions from tabulated values derived from a 1949 USDA-Forest Service report. These data should be updated to incorporate the last 70 years of hydrologic data and more recent scientific understanding of landscape response as well as real-time streamgage data.

This type of information is vital to local and state agencies to predict which communities or infrastructure may be at-risk from floods. It also needs to be accurate and timely to avoid exposing the public to unnecessary risk while also avoiding “evacuation fatigue” from sending out too many unnecessary warnings. Real-time information is particularly relevant to emergency responders when a flood is occurring, but other local agencies may need to be more proactive about conveying flood risk, shoring up levees, protecting infrastructure, and finally warning the public to imminent threats. Most of the public investment to date has been in statistically based estimates of runoff responses but the emerging challenge of evolving western wildfire landscape is underscoring the need to shift investments to process-based forecast models.

The USGS currently has the scientific expertise to support this Use Case with experts in runoff modeling, hillslope and stream sediment transport, and streamgaging both here in California and nationally. The foundation of this new, integrated modeling effort will be real-time streamgaging and real-time flood inundation mapping in low-lying, susceptible communities. In addition, real-time rainfall runoff modeling would ingest quantitative rainfall forecast data (NWS already does this at some streamgages) as well as metrological and soil-moisture data. The model would also reflect changes to the landscape due to wildfire, particularly burn severity. The USGS has previously developed several distributed watershed runoff models in areas of California with recent and historical burns. Models for selected study areas would need to be updated to include a wildfire component. In addition, the BAER team uses other watershed models for emergency response such as those developed by the USDA-Agricultural Research Service (ARS) and USACE. After every fire, the parameters of the model would change and need to be updated. The streamflow models (for example: 1- to 7-day predictions) would not only forecast stage/discharge at discrete points, but those points can be used in a flood-inundation forecast map to know where flooding will occur. As the USGS improves the scientific understanding of hillslope response in a post-wildfire landscape, we’ll be able to gradually improve these prediction models.

The USGS is also currently working on developing a low-cost streamgaging network system that can be deployed rapidly in a post-wildfire burn area (or anywhere) and can easily increase the number of gages in an area by an order of magnitude, in addition to incorporate other sensors (soil moisture, rainfall, water quality, etc.). These new environmental sensing data would feed into the real-time component of the post-wildfire EarthMAP as well as the forecasting component. The USGS has already formed partnerships with organizations that manage reservoirs and downstream habitat to improve understanding of long-term watershed responses to fire and has completed several studies specifically focused on their organization needs, including reservoir and downstream water quality, the effects of sedimentation on long-term water supply reliability, and below-dam fish habitat. Improvements to these process-based forecast models of sediment transport associated with post-fire runoff will be immediately useful for these longer-term questions as well.

The key areas for scientific advancement lie in pairing new fire-specific data with existing models to base short- and long-term forecasts on actual measurements, expanding the historical database of post-fire runoff, and improving the delivery of model forecasts.

The USGS already has formed many partnerships that will be leveraged for this post-wildfire EarthMAP Use Case, from technology development, to understanding the real-world management decisions that can be improved upon using these types of data. Both the USGS National Innovation Center (NIC) and CAWSC have partnered with Carnegie Mellon University and NASA to develop new sensor and telemetry tools. We’ve already partnered with Sonoma County to use their area as a test bed for these innovative streamgaging technologies. We’re also regularly communicating with California Department of Water Resources to understand how we can improve situational awareness (related to flooding) in areas hit hard by recent wildfires.

This EarthMAP Use Case will leverage many capabilities that the USGS already has and is currently developing as well new technical capabilities which are within reach. Previous work, particularly at the TXWSC have improved visualizations of real-time streamgage data with real-time overlays of radar and other observational weather data. In addition, the CAWSC is currently working to expand the observation network for streamgaging and flood monitoring which incorporates recent advances in low-power computing and telemetry technologies. Perhaps more importantly, this team of USGS scientists already has strong relationships with the resource managers working on post-fire issues across California and a good understanding of their needs.

The NIC, DOD, and CAWSC have been working with Carnegie Mellon and NASA to deploy a low-cost, LP-WAN (Low Power Wide Area Network) system of non-contact stream-height sensors. These small, low-cost sensors report stream heights at 5-minute or shorter time increments using a spread-spectrum technology called LoRa, optimized for low- power consumption over low distances. Installation of added network uplink sites, around which many low-cost stream sensors can be sited, would provide real-time alerts for post-wildfire hydrologic risks, including flooding (Sonoma Water request). These sensors can be placed downstream of hillslope infiltration monitoring sites where measurements every 1- to 2-months will show changing infiltration thresholds for runoff and recharge, providing a watershed-scale check on hillslope data. Data showing runoff thresholds through time, validated by non-contact stream-height sensors, can be used to improve runoff and water availability forecast models such as the new hydrologic model developed by CAWSC and their partners, as demonstrated in the pilot project in the Russian River.

CAWSC and GMEG could leverage existing support from PG&E, Sonoma Water, DWR, State Water Resources Control Board (SWRCB), and USGS NIC to extend post-wildfire infiltration monitoring networks and low-cost non-contact stream gaging sensors. Data showing infiltration rates and runoff response through time will be used to improve existing hydrologic models in the Russian River, improving post-wildfire flood and water availability awareness and forecasting. Support can also be leveraged from PG&E and DWR/BoR/SFPUC (San Francisco Water Power Sewer)/USACE to improve our ability to forecast sediment supply and subsequent transport that impacts reservoirs that supply drinking water for the San Francisco Bay Area communities. Data showing changes in hillslope sediment supply and stream transport will be used to improve forecasts of sediment threatening reservoirs, and to provide potential hillslope mitigation strategies. Data showing changes in stream channel metrics that matter to spawning and other ecosystem services will be compared to hillslope sediment supply measurements and models. The goal will be to improve forecasts of which channels have the highest susceptibility to post-wildfire impacts, in order to focus cost-effective mitigation efforts.

The need to improve situational awareness and forecast capability for post-wildfire flooding and water supply is an ongoing and cross-regional. Multiple aspects of expected scientific products will be transferable to other parts of the Western U.S. where similar monitoring networks, water availability, and sediment transport models are available. Specifically, the innovative low-cost, dense real-time observational network will be easily portable. The key to the scalability of this Use Case across California will be to clearly characterize how the extra investment in process-based forecasts can help resource managers reduce local risk, stabilize watershed resources, and improved regional water supply reliability by using the new science. Such an assessment must be led by our Public and Private sector partners, whose existing funding and support are keys to longevity and sustainability. The long-term viability of this Use Case will also depend on support from resource managers. Additional measures of success of this Use Case will be determined by whether the BAER teams and other resource management partners use the information products to communicate between different organizations and long-term assessments of hazard and risk reduction.

Primary Contacts: Anke Meuller-Solger, amueller-solger@usgs.gov; Jonathan Stock, jstock@usgs.gov

Drought in the Colorado River Basin (hereafter Basin) has increased in frequency, intensity, and extent in the early 21st century, and is expected to continue intensifying under climate change (fig. 5.15). Drought can severely degrade the ecosystems and natural resources that land managers are charged with sustaining for future generations. Despite the growing threat and effects of these interacting stressors, managers often have limited resources to effectively respond. While many adopt a “wait and see” or reactive management strategies, forward-thinking land and resource management is essential to prepare for impacts and allow for greater flexibility in decision making. Anticipatory management strategies are growing in importance, as abrupt and unexpected transformations of ecosystems have been increasingly observed and are likely to continue across the Basin. TDE scientists have produced models, datasets, and tools useful for decision making under increased drought. However, these products are generally project driven and have not been integrated in one platform that is accessible broadly to stakeholders.

The TDE researchers conduct cross-disciplinary science often in collaboration with Ecosystem and Water Resources Mission Area scientists in other Southwest and Rocky Mountain Region based centers and with academic partners nationally and internationally. They produce products that provide ecologically meaningful metrics of drought, understanding of how drought interacts with other stressors (wildfire, invasive plants and animals, insect outbreaks, etc.), evaluations of how drought and associated stressors impact natural resource conditions (for example, soil erosion and dust, ecosystem productivity and health, phenology, and water use), and assessments of how management actions can exacerbate, mitigate, prioritize, and (or) prepare for changes to natural resource conditions. Integration of existing and anticipated data and tools resulting from these scientific activities in a Drought Data Explorer will make drought-related information more readily accessible to our federal, state, tribal, and private stakeholders across the Basin. Drought and the need for salient and accessible information for management of the numerous ecological, safety, and public health and societal issues drought causes is a top priority from local to national levels of government and has been so identified within the USGS with a focus on drought in the Basin.

A Drought Data Explorer delivering related data and tools will improve stakeholder decision making at different stages of drought intensity. The Drought Data Explorer platform can allow broader integration of drought-relevant science across the region, linking TDE produced products with related data from other centers and Mission Areas. Management decisions typically mirror ecosystem changes and when drought intensity is low, management typically follows a “business as usual” approach, in which management actions are conducted as they have been historically. As drought intensity increases, managers seek out and adopt mitigation approaches, which are aimed at minimizing the severity of these stressors, including planning activities to mitigate risk and conducting vulnerability assessments to target need for proactive management. For example, increasing drought intensity can trigger managers to adopt adaptation approaches in which they attempt to mediate the consequences of drought, such as altering livestock management, reducing tree or shrub density, treating invasive species, limiting soil disturbance, and other preventive measures. When ecosystem transformations are inevitable, land managers may prepare by reallocating resources to degraded areas and conduct ecosystem restoration, potentially including assisted migration of species.

An integrated Drought Data Explorer can be used to identify areas in the Basin that are susceptible to elevated temperatures, decreased rainfall, and low soil moisture. These projections can be then coupled with metrics of vegetation production, actual evapotranspiration, wind erosion vulnerability and dust emission risk, altered fuel loads, and wildfire risk; all developed or being developed by TDE researchers. The Drought Data Explorer will quantify the drought vulnerability of ecosystems under different land uses and management treatments, including projections of the impact of different management action scenarios. Once ecosystems have undergone transformations to degraded states, land managers typically switch their strategies to facilitate the replacement of species and communities that may perform well under the altered climate and wildfire regimes. A Drought Data Explorer could help prioritize areas of land with the potential to respond favorably to restoration and reclamation efforts and develop new methods and techniques to increase restoration success. While many of the preliminary data layers and tools that would contribute to a Drought Data Explorer have been or are being developed by TDE researchers, this information to support decisions are currently being presented to managers on a per-project basis.

The Drought Data Explorer will enhance the capacity of USGS to deliver cutting-edge, interdisciplinary science that serves the Basin’s management community. TDE scientists have extensive experience engaging land and resource managers through one-on-one meetings, conferences, workshops, and field trips. These ongoing forms of stakeholder engagement have generated vast interest and great enthusiasm for developing novel and forward-thinking strategies to manage diverse ecosystems in the face of growing drought risk. The Drought Data Explorer will be developed from start to finish by maintaining these forms of engagement and the open communication TDE scientists have with managers in the Basin.

TDE scientists study how ecological drought manifests in the following specialties: climate and soil moisture; evapotranspiration in natural and degrade ecosystems; biogeochemical cycles and ecosystem productivity; soil erosion and dust emissions; vegetation composition and cover; plant and wildlife species of concern; changes to fuel load and wildfire activity; and invasive plants and insects. Also, extensive research and outreach are being conducted by TDE scientists in collaboration with land and resource managers to develop restoration and rehabilitation methods to address the consequences of drought and associated stressors.

The products developed by TDE researchers are multi-faceted and will be additive when integrated within a Drought Data Explorer. For example, data and information ready to integrate, include applications of the SOILWAT2 ecosystem water balance model to assess temporal and spatial patterns of ecological drought, both in the past and in the future under a suite of climate change scenarios. An extension of this would be parametrization of SOILWAT2 with cutting edge predictive soil maps, recently developed by another TDE team, for the Basin above Lake Mead recently developed. Another example of data ready to incorporate into a Drought Data Explorer are those resulting from a collaboration with EROS and the Western Geographic Science Center. In that collaboration scientists have integrated a vast network of historical permanent vegetation plots across the Basin with Landsat and weather station archives to produce coupled data layers of drought effects on vegetation production. In an upcoming project, TDE researchers will integrate land surface, atmospheric, and hydrologic processes in the Weather Research and Forecasting model using remote sensing products to evaluate controls of these processes over the 57-year Landsat archive. TDE scientists are collaborating with the Geosciences and Environmental Change Science Center who maintain a dust monitoring network and a one-of-a-kind dust-on-snow sample archive which can be used for dust-source provenance and is readily available for model integration with TDE products. Another example of existing and supporting data are within PlotNet, which contains 11,000 upland vegetation and soil cover monitoring datasets from the western US, including the Basin and the results from RestoreNet, a networked experiment co-produced with land managers to improve restoration strategies across the Basin. TDE products being developed in collaboration with the University of Arizona, will provide high-resolution and dynamic change maps of greenness in both riparian and terrestrial Basin ecosystems and annual and growing-season evapotranspiration metrics. Through the work of TDE scientists and their collaborators, there is a large amount of soil moisture, land cover, land use, vegetation change predictions, vegetation water use, and soil data across the Basin that can be and should be integrated.

Partnerships will be strongly leveraged to make the Drought Data Explorer operational, including current partnerships with other USGS Science Centers (for example, EROS, FORT, Arizona WSC, WGSC, WERC, FRESC, GECSC), and with federal (BLM, BIA [Bureau of Indian Affairs], NPS, USFWS, DOE, DoD, FWS, USDA), state (Arizona Game and Fish Department [AZGFD], Arizona Department of Forestry and Fire Management, Utah Division of Natural Resources), tribal (for example, Navajo Nation Division of Natural Resources, Hopi Department of Natural Resources, Ute Mountain Ute tribe, and academic (for example, NAU [Northern Arizona University, UofA [the University of Arizona], ASU [Arizona State University], USU [Utah State University], the University of Utah [UU], BYU [Brigham Young University], NMSU [New Mexico State University]) stakeholders throughout the Basin.

TDE scientists have substantial capability to support this Use Case with their ongoing field, lab, and modeling investigations, their strong relationships with agency and university scientists and management partners, and the research infrastructure they maintain in experiments, datasets, and equipment. While a great deal of this work is ongoing, with additional resources TDE could leverage synergies among, what are now, discrete research projects, and create a much-needed data and information delivery platform. The envisioned Drought Data Explorer will leverage data delivery frameworks that have been developed with individual projects. For example, an interactive, searchable Lower Colorado River Data Explorer is being developed within one project and is envisioned to be expanded to include additional river and upland ecosystems and transboundary, borderlands ecosystems. Other elements which have been developed with TDE science lead include the Smart Energy web platform and the DART tool.

In order for our research and outreach to evolve and become a fully functional and an operational service for scientists, land, water, cultural, and resource managers, and the general public, development of the Drought Data Explorer requires: a) adequate data storage, b) processing resources, c) web and tool developer time, and 3) long-term hosting. The Southwest Biological Science Center has made a commitment to advancing data science within the center and has supported the hiring of science support staff with the skills to develop advanced data delivery. We envision that the Drought Data Explorer can be developed with a modular design so it can accept new data and models as they are developed and is responsive to stakeholder needs for updated information.

TDE scientists have been performing applied, drought-relevant research throughout the Basin for decades. Because of their strong partner relationships and history of success, the viability of this science is high, and the technologies developed are utilized currently and have high probability of generating continued need for use as long as water availability is an issue in the Basin. Adequate staffing is a primary need for this Use Case. The TDE permanent staff or researchers is small (8 researchers) and high productivity is achieved where technician and post-doc support can be maintained and adequate support staff (data and computer scientists) are dedicated to development of advanced data delivery and decision support tools.

Drylands, including those in the Basin, currently cover 40 percent of the Earth’s land surface and are rapidly increasing in size. Drought threatens to increase the area of the nation and globe that must be managed as drylands. The production and delivery of drought-related science products—the data, tools, and models—will be a common goal of the managers of these lands and their resources. A Drought Data Explorer will make existing TDE science more available for utilization and can provide a technical model for integrated data delivery and decision support.

Wildfires will continue to increase in severity and frequency due to drought and a warming climate. Wildfires are a significant threat to all of the Western U.S. Regions and understanding and predicting wildfire risks are priorities for all stakeholders of these areas. Dissolved oxygen (DO) sags due to wildfire ash and debris flows entering aquatic habitats can cause significant die-offs. Understanding how different chemical reactions from burned materials can produce DO sags is a needed first step in developing a predictive tool of stream water-quality change that will help federal, state, and local resource managers preserve protected or endangered species, maintain valuable technological resources, and provide early warning to water treatment facilities.

The proposed Use Case will develop an empirical relation between geology, fire intensity, geochemical modeling, and DO concentrations. Once this relation is quantified, EarthMAP is the appropriate tool to integrate water-quality monitoring, geochemistry, fire science, geology, soils, precipitation, stream flow, and ecology, to predict how future wildfires could impact DO concentrations in downstream aquatic habitats and water resources. If fully developed, this tool could also help document and predict the timing, severity, and duration of fire effects on water quality in downstream networks and predict levels of trace metal concentrations that can be toxic to ecosystems and degrade drinking water quality.

Predicting DO sags is important to the preservation of endemic and endangered aquatic species. Once empirical relations are developed, EarthMAP could provide stakeholders with actionable information on downstream aquatic ecosystems that are in danger of having acute DO sags due to the geology of the burn area coupled with the intensity of the fire. This information could be supplied before a debris flow initiates, providing ample time to translocate species and remove or secure technological resources. Moreover, a predictive tool for stream water-quality change could improve stakeholder decision making by targeting habitats and in-stream operational facilities that are or are not likely to be impacted by changing stream quality.

Water quality in streams and rivers can be affected by runoff from burned areas. Pulses of fire residue can create large sags in dissolved oxygen levels and pH, as well as extreme highs in turbidity and specific conductance (SPC). These pulses have been known to cause die-off of aquatic flora and fauna. Monitoring streams and rivers for DO, pH, turbidity, and SPC is needed to develop an empirical model and predict the timing and magnitude of change caused by surface water runoff from burned areas to downstream ecosystems. SPC can be used to determine the removal of soluble constituents in initial discharge water from burned areas, and changes in DO, pH, and turbidity can identify subsequent or stronger discharge pulses. River networks in burned areas should be continuously monitored for water-quality parameters to help document the severity and duration of effects due to fire.

Understanding the geochemical and biologic pathways that produce sags in DO concentrations is important to predicting how pulses of fire residue will impact river networks. Two pathways have been suggested that likely account for DO sags. The first pathway is the chemical oxygen demand of the fire debris. If sulfides and reduced metals are present in the burn areas or in runoff debris materials, water can react with these materials and decrease DO concentrations in streams and rivers. The second pathway is the biological oxygen demand where organic materials in the burn area or in the surface water runoff remove DO from the water through aerobic decomposition.

Multidisciplinary data from different burn areas are needed for modeling the reaction pathways impacting DO concentrations in aquatic ecosystems downgradient from wildfires. Data collected from soil, ash materials, water, and sediments from burned areas and downgradient surface waters is needed. Concentrations and oxidation states of metals, concentrations of dissolved organic carbon and particulate organic carbon as well as continuous monitoring of water parameters of DO, pH, turbidity, and SPC from each burn area would be used in geochemical modeling efforts. The model results will help to determine the reaction pathways that are driving changes in redox potential, pH, and DO concentrations that would provide information on the cause of DO sag. These model results would then be compared to the mineralogy of surface geology from the burned areas, and fire intensity to determine a relation between geology, soils, fire intensity, and stream water quality, particularly DO concentrations. Once the relationship is developed this information will provide a predictive tool to evaluate how water quality and ecosystem health will be affected in future burn area scenarios.

Wildfire debris has impacted river systems across the Colorado River Basin (CRB), and many of these systems are already being monitored. Little research has used this data to determine the pathways that account for sags in DO concentrations during and after fire debris flows. A review of data collected by USGS and others would be done to leverage already existing data. Existing data from across USGS science centers and other federal, state, and NGO stakeholders should be collated into a database to ascertain if these historic projects have necessary information to inform the predictive tool development for the Use Case. Some examples of repurposing historic data collection activities to determine effects on DO concentrations in downgradient streams include USGS Nevada WSC long-term monitoring of Clear Creek near Carson City, Nevada, and the effects from the Clear Creek fire; Army Corps of Engineers, USGS, and academics at the University of New Mexico study on the Rio Grande and the effects from the 2011 Las Conchas fire in the Valle Caldera, New Mexico; and USGS California WSC and USGS Western Ecological Research Center [WERC] work on the 2009 Fire Station fire in the Angeles National Forest, California.

USGS currently has the capabilities to perform the work for the proposed data collection and to store data in National Water Information System (NWIS). Collecting, analyzing, quality assuring, and storing real-time data, and water, sediment, and soil sample analyses is relatively inexpensive and is being performed by many federal, state, local, and NGO stakeholders. Utilizing USGS information and leveraging other stakeholder data as furnished records is fairly routine. Geochemical modeling capabilities are available with the USGS developed PHREEQC model.

Once enough cases from historic and current data have been evaluated, a relation can be established between different variables to inform the predictive tool that would be developed using EarthMAP capacities. Geospatially linking geologic maps, soil maps, burn areas and wildfire intensity maps, and at-risk aquatic systems and water supply systems should be able to be defined quickly and efficiently with current capabilities. The communication to stakeholders about the findings from individual locations would be rapid. The development of the predictive tool would require leveraging capacities and integrating these capacities in EarthMAP over time.

A main goal of implementing the proposed Use Case is to collect data from burn and downgradient areas rapidly to provide initial datasets for determining stream baseline conditions, the timing and duration of water and sediment inflows from burn areas, the magnitude of change to stream water quality and the geochemical and biologic reactions that degrade water quality. The targeted stream systems would ideally have variable geologic materials of burn sites and fire intensity and could be selected from across the entire CRB. Because of the unprecedented quantities of fires across the CRB landscape, the Use Case could leverage systems that are already being monitored through NGWOS, WMA, or other stakeholder projects, with additional data collection activities added where needed. Evaluation of data and geochemical modeling are fairly routine and establishing the relation between the different variables would be done case by case.

Once relations between data are established, building a predictive tool is well suited for EarthMAP priorities. The predictive tool would leverage interdisciplinary data that already exists from multiple agencies (USGS and other DOI agencies, USDA, DOE, DOD, NGOs, and academic institutions) such as geologic surficial maps, river network information, precipitation maps, soils map, sensitive and endangered ecosystems maps, and fire distribution and intensity maps, along with real-time monitoring of DO, pH, SPC, and turbidity in tributary watersheds, water quality, sediment and soil sampling. Linking these datasets in EarthMAP, along with predictive relations, provides information on how aquatic ecosystem health and water quality would be impacted by DO sags cause by future wildfires.

Bottlenecks in data collection activities for the Use Case would include understanding what data are being collected by different stakeholders at project sites and how USGS can access these data. Communication within the USGS and with other federal, state, local, and NGO agencies can be lacking, resulting in duplicate data collection activities and a wide range of data storage repositories and quality assurance procedures. To leverage resources, it would be ideal to have good communication with all agencies and flexibility in planned data collection activities for each project site. Additionally, having access to furnished records from other agencies may be difficult and internal USGS data approval slow. The Use Case would work with USGS and stakeholder data managers to create semi-automated processes for accepting furnished records into NWIS. Using a semi-automated process could help to move data through the procurement and USGS quality assurance procedures to ensure timely delivery of data.

The long-term viability and sustained relevance of this Use Case is high. When complete, the Use Case would provide a predictive tool for land and resource managers to determine effects of wildfires on surface water DO concentrations, a primary driver of aquatic ecosystem health and a priority constituent for downstream water supply or treatment systems. Predicting the effects of wildfires on DO concentrations utilizes current data collection activities and capacities. Once the relations between geology, fire intensity, and DO is established, the predictive tool should be able to scale up to cover all burn area geographies. To determine effectiveness of the predictive tool, downgradient water quality in streams and water systems could be monitored to determine longer-term change, duration, and return to pre-fire conditions.

Primary Contact: Rebecca Frus, rfrus@usgs.gov

Most of California is relatively dry, and most residents and nearly all of the state’s highly productive agricultural sector depend on water delivery from two remote sources: the Central Valley and the Colorado River. Surface water operations in California’s Central Valley are essential to serving those needs. Because demand exceeds supply except in wetter years, release and diversion curtailments needed to protect fish and wildlife and their habitat impose significant and often expensive limitations to water availability for human purposes, especially agriculture. Water operations limitations for environmental purposes occur at multiple time and space scales, ranging from long-term reservation of water behind dams to support off-channel wetlands and seasonal flow and temperature profiles in the Sacramento River system to narrower real-time adjustment of Delta flows and export rates to prevent undue entrainment of fishes and degradation of aquatic habitat within the Delta. This layered challenge to water operations, and the associated costs of foregone water delivery, create a strong and durable demand for data and analytical products that can help water and fish and wildlife managers assess the cascade of effects of operations decisions from the physical environment to the native species and ecosystems that depend on it. Good information can inform more efficient resource management decisions that conserve species and habitats while providing water for people and farms. Improved science can also elevate high-profile regional planning for a future burdened by both a growing population and regional hydrology that continues to shift due to global climate change.

We propose to develop a data platform that draws together relevant long-term monitoring datasets and monitoring streams representing the physical, chemical, biological, and ecological state of the Sacramento-San Joaquin and associated watersheds, and to build and maintain upon that platform an integrated environmental model suite to support forecasting the response of wetlands ecosystems, aquatic ecosystem services, and fish and wildlife distributions and abundances to water management and policy choices and in the presence of ongoing climate change and land use development.

Because the fundamental goal of this EarthMAP proposal is to draw together a large array of relevant environmental data and develop from it an integrated model suite that can be applied to reliably forecast the ecological effects of water and related resource management decisions, we expect the effort will need to draw on the expertise of at least three USGS Mission Areas: Ecosystems, Water Resources, and Natural Hazards. Development of this Use Case will require collaboration with the Bureau of Reclamation and California Department of Water Resources, which manage the state and federal water projects in Central California, and the U.S. Fish and Wildlife Service, NOAA Fisheries, and the California Department of Fish and Wildlife, which implement of the Endangered Species Act, its California analogue, and have other resource stewardship mission responsibilities.

This Use Case directly addresses a central goal in the recent project descriptions and governing biological opinions pertaining to water operations in California: to support decision making that optimizes water operations while respecting the state’s water rights system and land-use requirements, and not threatening the continued existence or critical habitat of state and federally protected plant and animal species. It is a goal that the involved agencies and others have, for at least the last two decades, spent more than $50 million in science support dollars annually pursuing, most of it to field a large suite of long-term monitoring projects spanning the gamut from hydrodynamics and water quality to fish and bird abundances and distribution. These field programs have provided a magnificently detailed data resource for the Bay-Delta that is not available in most other systems. Some of the physical and chemical monitoring is conducted by the USGS; as well, we are relied upon by all the agencies named in this proposal for advice and assistance with the development of models, visualizations, and other analytical products needed to support various operations and public policy decisions. These partners are likely to be extremely supportive of a thoughtfully developed USGS plan to build a data platform and system of integrated environmental models to forecast water operations effects on ecological services and the habitats and native species living within the system.

We expect that this EarthMAP application would benefit stakeholders and their decision making in several ways.

The core foundations required for this EarthMAP application are already available. Data management is “old hat” for USGS, and both the skilled personnel and some of the relevant data are already housed at the California Water Science Center. Relevant Sacramento-San Joaquin watershed monitoring programs and historical datasets are numerous; most are concentrated at a few locations, including the California Data Exchange (CDEC), the Interagency Ecological Program, and within our California Water Science Center. The water operations-focused integrated environmental model suite we envision in this proposal has already been prototyped as an objective of the USGS CASCaDE (Computational Assessments of Scenarios of Change for the Delta Ecosystem) projects. The application we envision here is perhaps somewhat less daunting than CASCaDE in that the focus is exclusively on water operations effects in the Delta and valley tributaries. The degree of difficulty would depend on details of the forecasting capabilities and other factors that we would have to sort out in collaboration with stakeholder agencies during implementation. It is possible, even likely, that advancement of this project would lead to the identification of new monitoring needs or changes to existing monitoring programs in order to appropriately serve the model suite.

This project offers the promise of integrating and curating the relevant data sources across disciplines and organizations expressly to support the envisioned integrated environmental model suite, and to erect a purpose-built model suite to support the range of quantitative water operations-related questions that are the core interest of this proposal. In all, the Use Case we envision would represent an exciting extension of the integrative, multidisciplinary work USGS has already done to address management challenges in this region to date, and offers a chance to provide clearly better tools to stakeholders charged with making high-profile, high-value decisions pertaining to California water, ecosystems, and species.

We anticipate that implementation of this proposal would require a supra-regional team effort engaging scientists and data managers at the California Water Science Center (CAWSC), the Water Resources Mission Area branch staff in Menlo Park and at Moffett Field, the Pacific Coastal and Marine Science Center, the Western Geographic Science Center, the Western Ecological Research Center, the Western Fisheries Research Center, and possibly others. Because success depends on accurately linking the envisioned capabilities to specific partner/stakeholder requirements, we expect that a strong collaborative steering engagement with Reclamation, the California Department of Water Resources, the Delta Stewardship Council, the State Water Resources Control Board, and the three federal and state fish and wildlife agencies would need to be organized at the outset to ensure that specific first-generation products and realistic timelines could be identified very early on. We do not anticipate any new technology needs at this time, though there may be a need for AI/ML support depending on the implementation objectives. Data platform requirements can be developed in collaboration with stakeholders.

This EarthMAP Use Case is inherently a long-term proposition, and gains value as the commitment is extended. The tension between California water management imperatives has existed for decades and will unquestionably continue for the indefinite future. We believe that careful, effective implementation of the proposed data platform and integrated environmental model suite would create an analytical resource that stakeholders would adopt and internalize in their planning and decision processes, improving water management for the benefit of the public and creating lasting demand for both the products and USGS expertise and technical assistance. It is plausible that progress on the envisioned Use Case in the Bay-Delta could lead to expansion of the Use Case to cover multiple estuaries, though it has high value even as a stand-alone effort. We expect that the most appropriate performance metric for the proposed Use Case is the degree of adoption of its products by Bay-Delta stakeholders, and their reliance upon them in mission-critical resource management decisions.

Primary Contact: Mike Chotkowski, mchotkowski@usgs.gov

Many reservoirs across the Western United States are fed primarily by runoff generated by high elevation alpine snow cover in the spring and summer months. Accurate predictions of snowmelt runoff that will occur during the late spring and summer months allow managers to optimize reservoir operations, water deliveries, hydropower production, and other water management objectives, such as balancing supply and demand.

Improved accuracy for estimates of high elevation stored snow water equivalent and runoff will also facilitate management of aquatic species that require minimum flow volumes or stream temperatures. With higher confidence in the volume of runoff expected to be generated from remaining high elevation snowpack in the spring, reservoir managers would have more leeway to release water earlier in the year in order to meet aquatic species management objectives, such as approximation of a natural flow regime.

Finally, flood forecasters would also benefit significantly from improved estimates of stored snow water equivalent. Over the short term (daily to weekly), flood forecasters would be able to more accurately predict near term hazards by combining near term weather forecasts with gridded snow water equivalent to predict runoff and identify potentially hazardous situations. Over the medium term (monthly), flood forecasters could use gridded estimates of snow water equivalent to drive simulation models in the early spring that would help identify specific weather events (for example, a large rain-on-snow event, a rapid-onset sustained warming event) likely to result in hazardous flooding well ahead of the potential onset of these events. This could allow for earlier warning of potential flood events. Notably, once validated, the Use Case described here could provide these benefits not only for basins feeding reservoirs, but for all basins with high elevation snow cover, including basins lacking the in-situ instrumentation that would otherwise aid in flood prediction. This could enable more specific flood predictions aimed at specific smaller basins.

Making predictions of spring and summer snowmelt runoff is notoriously difficult in basins that include a substantial amount of terrain above the alpine treeline. In these basins, estimates of stored snow water equivalent (SWE) that drive runoff forecasts are often deficient for several reasons. First, automated existing in situ observations of SWE (such as those provided by the NRCS SNOTEL network) are typically only available from lower elevation sites below the alpine treeline and may not reflect snowpack conditions at the higher elevations. Second, modeling and remote sensing used to estimate SWE is generally conducted at relatively coarse (500–1000 m) spatial resolution that is not sufficient to resolve the mosaic of patchy alpine snow cover that generates the majority of the late spring and summer runoff. The combination of these two factors often leads to inaccurate estimates of high elevation SWE. Novel approaches such as those developed for NASA’s Airborne Snow Observatory (ASO) have been demonstrated to be highly effective, but require regular airborne surveys that may be cost prohibitive in many areas. Federal agencies (such as the Bureau of Reclamation) and local water districts responsible for reservoir management, as well as agencies responsible for issuing snowmelt runoff forecasts (such as the Natural Resources Conservation Service), stand to benefit significantly from an approach that can improve the accuracy of snowmelt runoff forecasts without requiring expensive regular airborne surveys. The primary benefit of the approach proposed in this EarthMap Use Case would be increased accuracy in seasonal runoff forecasts for the upcoming spring and summer months of the same year. In addition, the proposed Use Case could also inform longer term decision making by reservoir managers when used to simulate the volume and timing of snowmelt runoff that would be expected over the longer term (decades) under different climate conditions as well as in the case of substantial land cover changes.

The approach proposed here has the potential to improve both the spatial resolution and accuracy of stored SWE estimates for basins with substantial above-treeline terrain. Developing and implementing a standardized modeling and remote sensing approach and resulting data product would be of great benefit to reservoir managers in districts with more limited financial and technical resources. This data product would consist of a regularly updated near-real time gridded estimate of SWE produced by a validated modeling and remote sensing approach that would otherwise be beyond the capability of reservoir managers who did not have access to a staff with the capability to develop and apply a complex, physically-based model that incorporates regular updates from spaceborne remote sensing. This approach would deliver a product that reservoir managers from the Bureau of Reclamation and local water districts have indicated would allow them to make better, more informed reservoir management decisions in basins where spring and summer runoff is generated from patchy, high elevation snow cover.

The USGS currently (2020) has the capacity to implement the proposed remote sensing and physically based modeling approach. The needed datasets, key models, and remote sensing techniques that will form the backbone of the approach are already available from a combination of sources within and outside of the USGS. Successful implementation of the approach will require developing a standardized approach for acquiring the required input datasets and remote sensing imagery as well as incorporating them into the physically-based snow evolution model.

The scientific and technological capacity to implement this Use Case exists primarily within the USGS. Dr. Graham Sexstone (Colorado Water Science Center) has extensive experience with implementation, calibration, and validation of physically-based snow models, such as SnowModel, developed by Dr. Glen Liston (Colorado State University). Dr. David Selkowitz (Nevada Water Science Center) lead the development of the recently released USGS Landsat Level 3 Fractional Snow Covered Area product and is currently working to create a combined Landsat-Sentinel fractional snow covered area product that would provide a much higher temporal resolution than the existing Landsat product. USGS employees and contractors based at the EROS Data Center have extensive experience with the production and distribution of Landsat-derived datasets and are currently developing a Sentinel 2 surface reflectance product that would enable the production of the combined Landsat-Sentinel snow covered area product. Additional datasets required for implementation of the physically based snow model include the National Elevation Dataset and the National Land Cover Database (land cover and forest canopy layers), produced by the USGS, as well as the North American Land Data Assimilation System (NLDAS-2) available from NASA. Daily in situ SWE measurements from the Natural Resources Conservation Service SNOTEL network would also be used for calibration and validation of this approach (fig 5.16).

Future improvements to required remote sensing and modeling datasets have the potential to increase the accuracy of modeled snow water equivalent. In particular, increases in the spatial resolution and accuracy of the National Elevation Dataset, such as those associated with the ongoing and expanding acquisition of LiDAR data across the nation, would improve the accuracy of simulated wind redistribution of alpine snow cover. Higher frequency updates of the National Land Cover Database (land cover and forest canopy layers) could also lead to additional increases in the accuracy of modeled SWE. Finally, daily streamflow observations available from USGS will serve as a crucial dataset for validation of modeled snowmelt runoff. Streamflow observations may also be augmented by the inclusion of reservoir inflow data from agencies such as the Bureau of Reclamation as well as local water districts.

The technical capacity to execute the proposed Use Case is currently available within the USGS. Development of the combined modeling and remote sensing approach would be conducted by Sexstone and Selkowitz, who will soon begin work funded by the Bureau of Reclamation for related work in two pilot basins in Colorado. Scaling the approach to a larger number of basins in the Western United States would require high performance computing (HPC) resources and staff, and near real time implementation of the approach for numerous basins would require the expertise of staff at EROS who have extensive experience producing and distributing regularly updated gridded datasets.

We anticipate this Use Case will be highly relevant over the next several decades. High mountain snowpacks currently serve as a large natural reservoir that fills up over the winter and spring and then delivers water over the period of highest demand in the summer. Projections indicate the timing of snowmelt will be earlier in the spring across much of the Western U.S., resulting in a larger mismatch between peak flows and peak demand later in the summer. Optimization of reservoir operations would thus become even more crucial, with little room for error. At the same time, climate conditions outside the range of conditions covered by the historical record of in situ snow observations may lead to decreased reliability of statistically-based snowmelt runoff prediction approaches, such as those currently used by the Natural Resources Conservation Service. In extraordinarily warm years, significant snow accumulation could even be confined entirely to the highest elevations which lack a historical record of in situ SWE observations. In this situation, statistically-based forecasting approaches would no longer be viable, while the physically-based modeling and remote sensing approach described here would continue to be effective.

The ability to sustain the modeling approach proposed here would be dependent on continuous acquisition of the USGS Landsat and European Space Agency Sentinel 2 remote sensing data. With a modest effort, the approach could be adapted to work with other optical remote sensing instruments that offered similar spectral bands, spatial resolution, and imaging frequency. Additionally, continued near real time production of the NASA NLDAS-2 model would be necessary, although the Use Case approach could be adapted to use an alternative model that produced similar outputs.

Once validated, the approach described here could also be adapted for implementation in high elevation watersheds outside of the United States. Estimates of stored snow water equivalent derived from physically based modeling and remote sensing for basins lacking in situ instrumentation or a record of historical snow cover observations would enable runoff forecasting that would not have been possible otherwise.

The primary metric for evaluation of this Use Case would be validation of modeled snowmelt runoff from stream gauges and reservoir inflow monitoring. The accuracy of the modeled snow water equivalent could also be directly evaluated via comparison with airborne snow survey results for basins where this type of data became available. The ultimate measure of success for this Use Case, however, would be increased snowmelt runoff prediction accuracy in basins where existing forecasts had previously been relatively poor.

Finally, it is worth noting that the science and technology developed for implementation of this Use Case would have broad applicability for a variety of problems where a combination of high spatial resolution remote sensing and physically-based modeling could be used to more accurately monitor a specific hydrological or ecological process. The combination of remote sensing data from Landsat and Sentinel 2 remote sensing data would be particularly applicable for monitoring irrigation and evapotranspiration.

Primary Contact: David Selkowitz, dselkowitz@usgs.gov