Purpose and Scope

The purpose of this report is to describe what potential factors influence benthic algal biomass and community composition across study years and among sites by documenting physical and benthic algal conditions at 20 surface-water sites in the upper White River Basin from 2018 to 2021. Data were collected by the USGS, Trout Unlimited, and CPW. This report (1) presents site-specific data including water temperature, riparian canopy cover, streambed particle size, and algal biomass and community composition; (2) describes the potential for streambed movement during spring runoff using physical channel characteristics and peak streamflow velocities; and (3) explains the results of a linear mixed-effects model used to test hypotheses about the influence of physical and chemical factors in explaining the occurrence of algal blooms across the basin. Hypotheses tested include the following:

The upper White River Basin, as defined in this report, is the area of land drained by the White River from its headwaters in the Flat Tops, a mountain range in Garfield County, Colo., to just upstream from the confluence with Curtis Creek near Meeker, Colo. (fig. 1).

Description of Study Area

The upper White River Basin drains approximately 1,020 square miles of the White River Basin in northwestern Colorado (fig. 1). More information about the study area is available in Day (2023). The study area was divided into five subbasins for this analysis (fig. 1; table 1): (1) upper North Fork White River; (2) lower North Fork White River; (3) South Fork White River; (4) upper main stem White River; (5) lower main stem White River. Land cover within the upper White River Basin is largely forest, 64 percent; herbaceous (including grassland), 14 percent; and shrubland, 9 percent (fig. 1). Cultivated crops (including hay and pasture) comprise 4.5 percent of the basin and are primarily located along the river and concentrated in the lower main stem White River subbasin. Urban development covers 0.5 percent of the land area and is also concentrated in the lower main stem White River subbasin (Multi-Resolution Land Characteristics Consortium, 2021).

Previous Investigations

The upper White River Basin has been the focus of multiple studies of water quality and biology regarding the onset of benthic algal blooms (May and Noble, 2017; HydroSolutions, 2017; Skibo, 2018; GEI Consultants, Inc., 2021; Hodge and Eyre, 2021; Day, 2023). The USGS has collected continuous streamflow data at the White River above Coal Creek streamgage site since water year (WY) 1962. A WY is the 12-month period from October 1 through September 30 and is designated by the year in which it ends. Discrete water-quality data have been collected quarterly at four sites located on two major tributaries and the main stem of the White River across varying timescales (USGS, 2021a). These data, in addition to supplemental water temperature, water-quality, and macroinvertebrate data, have been used to investigate benthic algal occurrence in the upper White River Basin.

In 2016, water-quality, macroinvertebrate, and benthic algal samples were collected from multiple locations in the White River and Coal Creek Basins (May and Noble, 2017). Algal biomass exceeded the Colorado Department of Public Health and Environment (CDPHE) standard of 150 milligrams per square meter (mg/m²; CDPHE, 2017) at four of five sites sampled, and the filamentous green algae C. glomerata was identified as the most visually abundant species at most sites. A nutrient enrichment study was also conducted at five sites and identified nitrogen availability as the limiting factor in algal growth. The authors also suggested that nutrient availability in the main stem White River was sufficient to support nuisance algal blooms along much of the study area. Reductions of both nitrogen and phosphorus were recommended to reduce algae growth in the river.

Two independent analyses of existing streamflow and water-quality data were performed in the study area (HydroSolutions, 2017; Skibo, 2018). The analysis confirmed that adequate nutrient concentrations exist for algal blooms to occur and that a reduction in spring and summer streamflow, particularly in the North Fork White River, likely contributes to the occurrence of algal blooms (Skibo, 2018). Long-term changes in streamflow-normalized concentrations and loads were identified, including decreases in total nitrogen and increases in total phosphorus (HydroSolutions, 2017). Changes in streamflow, including lack of sustained scouring flows, timing of runoff and length of growing season, and occurrence and frequency of low flows and changes in climate were cited as contributors to nuisance algal blooms.

A macroinvertebrate study conducted from 2017 to 2019 characterized macroinvertebrate communities within the upper White River Basin (GEI Consultants, Inc., 2021). Results of the analysis indicated that macroinvertebrate assemblages were generally healthy and well-balanced. The presence of benthic algae such as Cladophora can have negative and positive effects on macroinvertebrates through changing factors like habitat suitability and food availability. The presence of benthic algae may have caused differences in macroinvertebrate community composition compared to sites without algae, though high variability and other unidentified factors could also have influenced macroinvertebrate populations.

Continuous water temperature data collection at White River above Coal Creek began in WY 1978; however, the continuous water temperature record is limited to WY 1978–84 and 2007–22, and monitoring data from 2007–22 are seasonal (May 1 through September 30) (USGS, 2021a). Continuous water temperature collected during parts of 2019 and 2020 at the same monitoring sites used in this study (fig.1; table 1) were previously used to explore spatial and temporal patterns in stream temperature (Hodge and Eyre, 2021).

Data Collection

The USGS, Trout Unlimited, and CPW collected data from 2018 through 2021. This section provides details on the methods of channel cross-section surveys, streambed particle size characterization, streamflow velocity measurements, riparian canopy cover estimation, and benthic alga sample collection and analysis. All channel geometry, particle size, streamflow velocity, canopy cover, and algal taxonomy data are in a USGS data release (Day and others, 2023). Water temperature data are from Hodge and Eyre (2021).

Data Analysis

Streamflow at White River above Coal Creek during the study period, from October 1, 2018, through September 30, 2021, was compared to below-normal, normal, and above-normal streamflow over the period of record (WYs 1962–2021). Streambed movement was estimated to better understand sediment transport and disturbance regimes across sites. A linear mixed-effects model was used to evaluate the relative concurrent roles of different drivers, including nutrient, hydrologic, and physical habitat conditions, on algal biomass across study sites.

Description of Hydrograph

Attributes of the annual hydrograph can strongly influence aquatic biota (Poff and others, 1997). Annual fluctuations in precipitation and air temperatures in the winter and spring play an important role in the timing, duration, and magnitude of snowmelt runoff (Woodhouse and others, 2016), the largest input of water to snowmelt-driven systems like the upper White River Basin. Subsequent low-streamflow periods are influenced by groundwater inputs and anthropogenic uses, such as irrigation and other agricultural purposes. Streamflows during spring runoff are generally cold and turbid and can act as a physical disturbance to the aquatic ecosystem. Streamflows during the summer (July–Sept) are relatively stable, can have lesser streamflow volumes that are warmer, have greater light penetration, and have longer residence times.

The following section describes the annual hydrograph at White River above Coal Creek during 2018–21 in relation to short-term climate and average annual runoff dates and is used to support further investigation of annual variation in water temperature, streambed movement, and algal biomass. An assessment of long-term changes in annual streamflows and how they might influence algal blooms are discussed in Day (2023).

During October 2018 through September 2021 the hydrograph at White River above Coal Creek represented a wide range of conditions compared to the long-term record, 1962–2021 (fig. 2; table 2). In 2018, low snowpack and warm spring air temperatures at this site resulted in an early peak streamflow, 17 days earlier than the average peak streamflow date from 1962 to 2021 of May 28 (USGS, 2021a). Streamflow was around 50 percent of average from April through June 2018 (Natural Resources Conservation Service, 2021) and was in the below-normal range for much of the summer months, July, August, and September. In 2019, high snowpack and cool spring air temperatures resulted in an extended duration of streamflow runoff with peak streamflow occurring on June 21, about 23 days later than the May 28 average peak streamflow date (table 2). Streamflow was 135 percent of average from April through June 2019 (Natural Resources Conservation Service, 2021) and remained above the 90th percentile (much above normal) through July (fig. 2). In 2020, peak streamflow occurred closer to the average peak streamflow date. Streamflow was 63 percent of average from April through June (Natural Resources Conservation Service, 2021) and was in the much-below-normal percentile for most of the summer (fig. 2). In 2021, peak streamflow was close to the average peak streamflow date, and streamflow from April through June was 38 percent of normal (Natural Resources Conservation Service, 2021) and remained much below normal through the summer (fig. 2).

Water Temperature

Water temperature is an important factor that can control the development of algal biomass because it regulates algal nutrient uptake and growth rate (Munn and others, 1989; Bowman and others, 2007). In general, growth rates increase in the temperature range from 13 to 20 degrees Celsius (°C), with lowered growth rates and stress observed in the range from 25 to 30 °C, and strongly negative growth rates at temperatures greater than 30 °C (Whitton, 1970; Wong and others, 1978; Graham and others, 1982; Lester and others, 1988; Dodds, 1991).

The water temperature record at White River above Coal Creek (site no. 18, fig. 1) is relatively short and discontinuous, from 1978 to 1984 and 2007 to 2022, limiting investigation of water temperature trends over time but allowing the comparison of stream temperature among study years. Water temperatures from May through October at White River above Coal Creek were compared across study years (2018–21) and were compared to a temperature threshold of 13 °C, below which Cladophora growth is very slow. Monthly mean stream temperature data collected by CPW during June, July, and August 2020 (Hodge and Eyre, 2021) are summarized and compared across 20 sites to a temperature threshold of 13 °C to facilitate investigation of site-specific factors promoting algal blooms.

Water Temperature at White River Above Coal Creek Near Meeker, Colorado

Differences in water temperatures among the study years mirrored differences in streamflow. Daily mean water temperatures in May were relatively similar between WYs. In 2019, water temperatures remained low through June, reflecting the extended duration of snowmelt runoff (fig. 3). Water temperatures reached and stayed above the temperature threshold favorable to Cladophora growth (13 °C) in early to mid-June in 2018, 2020, and 2021, whereas temperatures remained below the threshold until July 7 in 2019. During all years, temperatures remained above the temperature threshold until early to mid-September. Data are missing from mid-July to mid-August 2021 due to fouling of the temperature sensor, though the temperature records of a site nearby, Yampa River near Maybell, Colo. (USGS site number 09251000), do not show a substantial change in temperatures during this time period (USGS, 2021a).

Water Temperature at 20 Sites

In 2019 and 2020, mean August water temperature generally increased as elevations decreased (fig. 3), except at North Fork White River below Mirror Creek and North Fork White River below Marvine Creek, where temperatures were cooler than at the site upstream (fig. 4A; Hodge and Eyre, 2021). On the two main tributaries to the main stem White River, mean August temperatures were above the temperature threshold favorable to Cladophora growth (13 °C) in 2020 starting at North Fork White River at Rio Blanco County Road 14 and South Fork White River near Buford and continued to the lower main stem sites for both years (fig. 4A). Mean August temperature at about 9,000 ft elevation was also above the temperature threshold at North Fork White River below Trappers Lake, likely due to water warming in Trappers Lake before exiting the spillway. August water temperatures were warmer in 2020 at all sites except North Fork White River below Trappers Lake. Spatial and temporal patterns in water temperature are described in more detail in Hodge and Eyre (2021).

Riparian Canopy Cover

Light reaching the stream channel is limited by riparian canopy cover, and lower levels of light can limit primary production of benthic algae (Lowe and others, 1986; Mosisch, 2001). When riparian canopy cover is limited, stream algae are exposed to more light, typically resulting in increased primary production potential. In addition to increasing algal biomass, increased light can also change the composition of algal assemblages by promoting the growth of filamentous green algae (Steinman and others, 1989; Mosisch, 2001). Mean canopy cover was relatively low across sites, ranging from 0 to 26 percent (fig. 4B). North Fork White River below Mirror Creek, North Fork White River at Buford, and South Fork White River near Budges Resort had the highest mean canopy covers of 22, 26, and 26 percent, respectively. All sites on the main stem White River had <2 percent cover except White River above Highland Ditch Head, which had 9 percent cover.

Streambed Particle Size and Streambed Movement Estimates

The annual spring snowmelt pulse can temporarily reduce benthic algal production and abundance directly or indirectly through scour and, in years with streamflows of sufficient magnitude, streambed movement (Power and Stewart, 1987; Biggs, 1996). Variation in particle sizes may affect algae distribution within the basin, as streambed movement is dependent on streambed particle size and the hydraulic characteristics at each site. Streambed surface particle sizes varied across sites in the upper White River Basin and did not appear to correlate strongly with longitudinal position in the basin (fig. 4C). The D50 and D84 often increased and subsequently decreased between the sites, exhibiting modest variability.

Critical shear stress, or the estimated amount of shear stress required to move a given particle size was estimated for the D50 and D84 at 20 sites, based on streambed particle-size distribution (table 3; fig. 5, grey bar). Where possible, boundary shear stress, or the stress of hydrodynamic forces acting on the streambed, was estimated during spring snowmelt (table 3; fig. 5, orange and black dots). When boundary shear stress exceeded critical shear stress in cross-section calculations, streambed movement of either the D50 or D84 was expected to have occurred (table 3; fig. 5, orange dots). Movement of smaller particle size classes (<D50) was not assessed in this study, though the influence of less intense hydrologic events on algal biomass, including rain events during the summer, is discussed in the “Potential Controls in Algal Biomass Among Years” section of this report. During years with low snowpack, snowmelt streamflow velocities did not exceed the critical shear stress required to move the average streambed particle. Boundary shear stress was sufficient to mobilize the D50 at zero sites in 2018, seven sites in 2019, and two sites in 2020. Boundary shear stress was sufficient to mobilize the D84 at zero sites in 2018, five sites in 2019, and one site in 2020 (table 3; fig. 5). Movement of the D84 can indicate that substantial sediment transport and bed disturbance occurred. It is unlikely that streambed movement occurred at sites in the upper North Fork subbasin (table 3). The movement indicated at a site is based on the geometry measured at one cross-section location and the hydraulics represented by high-water marks recovered in the reach. Boundary shear stress determination is strongly influenced by stream slope, which is obtained by high-water marks surveyed in the field. Variability in the quality and spatial distribution of the recovered high-water marks results in a wide variation in boundary shear stress determinations from site to site and year to year.

Benthic Algal Community Composition and Abundance

Algal communities are commonly described using the abundance of algae in a sample and expressed as number of cells per volume of water, while the biovolume of algae is expressed as volume of cells per volume of water. Biovolume is essentially a measure of how much space algae are taking up in a sample and is used as an estimate of biomass. Thus, relatively small taxa like cyanobacteria may have smaller biovolume but greater abundance when compared to large filamentous green algae. Both abundance and biovolume enable discussion of overall community composition, dominant taxa among sites, and potential species of concern. Community composition was assessed at 20 sites in 2020 and at 5 sites in 2021.

From samples collected in 2020 and 2021, 49 genera of algae were identified (Day and others, 2023). The taxon groups Bacillariophyta (diatoms), Chlorophyta (green algae), Cyanobacteria (blue-green algae), and Cryptophyta and Chrysophyta (golden-brown algae) represented 59, 20, 16, and 5 percent of taxa identified, respectively. About 80 percent of genera identified were rare, contributing less than 1 percent to either total algal abundance or biovolume. Twenty-four genera contributed 10 percent or more to total abundance or biovolume.

In 2020, green algae represented more than 50 percent of algal biovolume at 14 of 20 sites, reflective of the large size of green algae (fig. 7A). Diatoms represented more than 50 percent of algal abundance at 17 out of 20 sites and comprised the majority of biovolume at sites with low biovolume of green algae (fig. 7B). Cyanobacterial biovolume was low, less than 5 percent across sites, but abundance was greater than 25 percent at six sites, with greater abundances more common at sites on the main stem White River. In 2021, the relative biovolume of green algae decreased and diatom biovolume increased at four out of the five sites compared to 2020 (fig. 7C). The relative abundance of cyanobacteria was greater in 2021 than in 2020 across the five sites (fig. 7D).

Green algae and cyanobacteria are likely to dominate when temperatures are warm and when streamflows are at seasonal lows (Stevenson and Rollins, 2017). Four genera of filamentous green algae were identified in the upper White River Basin, including Cladophora, Stigeoclonium, Ulothrix, and Spirogyra (fig. 8A, B). In 2020, filamentous green algae occurred at 14 sites and in all subbasins. The most commonly occurring filamentous green algae in 2020 were Ulothrix and Cladophora, which occurred at seven and six sites, respectively, and seven sites had more than one genus of filamentous green algae identified (fig. 8A). In 2021, filamentous green algae occurred at four of the five sites, with Cladophora and Spirogyra composing the majority of green algal biovolume across sites (fig. 8B). Spirogyra replaced Cladophora at White River above Dry Creek in 2021.

Cyanobacterial blooms frequently occur in eutrophic (nutrient rich, high productivity) lakes and reservoirs but may also occur in oligotrophic (nutrient poor, low productivity) systems (Graham and others, 2008). Cyanobacterial genera identified in the upper White River Basin include Planktothrix, Eucapsis, Dolichospermum, Pseudanabaena, Calothrix, Cuspidothrix, Lyngbya, and Planktolyngbya. Eucapsis was the most common, found at 15 sites in 2020 and 4 sites in 2021, and usually had the greatest relative abundance when present (fig. 8C, D). Planktothrix, Pseudanabaena, and Lyngbya are filamentous cyanobacteria genera that contain toxin-, taste-, and odor-producing strains (Graham and others, 2008). In 2020, Planktothrix was found at North Fork White River at Rio Blanco County Road 14, South Fork White River near Buford, and White River below Big Beaver Creek; Pseudanabaena was found at South Fork White River near Budges Resort. Planktothrix was found at White River above Curtis Creek in 2021. Identification of strains capable of producing these compounds does not necessarily guarantee that toxin production was occurring during the time of sampling. The mat-forming Lyngbya genera was found in 2021 at South Fork White River at Buford and North Fork White River at Buford.

Diatom composition and abundance data are not presented in detail in this report, but the complete dataset is available in a USGS data release (Day and others, 2023). The diatom Didymosphenia geminata, commonly referred to as didymo or “rock snot,” was found at South Fork White River at Buford and White River above Dry Creek in 2020 and 2021. D. geminata is an invasive, bloom-forming species that can affect the diversity, abundance, and productivity of other aquatic organisms (Kumar and others, 2009). Unlike other periphyton growth, which is typically stimulated by the enrichment of nutrients, D. geminata are generally found in oligotrophic streams and rivers.

Benthic Algal Biomass Comparison to Aquatic Standards

Biomass concentrations are compared across sites and years and are also compared to CDPHE interim water-quality criteria for chlorophyll a (figs. 4D, 9) (CDPHE, 2017). Benthic algal biomass ranged from 0.7 to 309 mg/m² during the summers of 2018 through 2021 and exceeded the CDPHE criteria of 150 mg/m² on four occasions, all during 2018 (figs. 4D, 9). The highest median algal biomass concentrations at North Fork White River at Buford and White River above Dry Creek were 51 and 50 mg/m², respectively (fig. 4D). White River above Coal Creek, White River near Meeker, and White River above Curtis Creek are the three most downstream sites in the basin and had median biomass concentrations around 34 mg/m². Algal biomass concentrations in the South Fork subbasin were consistently highest at South Fork White River near Buford across the sampling period (fig. 4D). Algal biomass was consistently low across the sampling period at six sites in the upper White River Basin, including all sites in the upper North Fork subbasin, North Fork White River below Marvine Creek, and South Fork White River near Budges Resort (fig. 4D).

Algal biomass was substantially higher in 2018 than in other study years in most subbasins except the upper North Fork, where concentrations were consistently low (figs. 4D, 9). An important limitation associated with targeting one point in time (for example, peak algal conditions) across a large spatial scale is that the timing of peak conditions can vary among sites. For example, at some sites algae may be at peak biomass, while algae at other sites have begun to senesce or die off. Attempts to minimize this error were implemented through sampling all sites as quickly as possible once sampling started. Additionally, this study assessed the effect of boundary shear stress during peak streamflow on algal biomass, though small rain events or other hydrologic disturbances that occur after peak runoff (but prior to sampling) can also reduce algal biomass. Because of these limitations, local volunteers and USGS personnel monitored algal occurrence at established locations using repeat photography and observations to confirm general findings of algal occurrence in the basin. In 2020, a rain event occurred prior to algal sampling, which doubled the streamflow at White River above Coal Creek (fig. 2). It is possible the increase in streamflow velocity or turbidity associated with the rain event decreased algal biomass, though observations from volunteers and USGS personnel support the overall findings of low algal biomass in 2020. In 2021, streamflows following snowmelt runoff exhibited greater variation than in other study years in response to summer rain events, with three flow disturbances occurring during this time (fig. 2). It is possible that these streamflow fluctuations could have detached larger algal filaments or increased turbidity, leading to overall decreases in biomass accrual or growth rates at some sites. Despite relatively low algal biomass values in 2021, observations by volunteers and USGS personnel suggested that the 2021 algal biomass values were likely an underestimate compared to actual conditions.

Potential Controls in Algal Biomass Among Sites

A linear mixed-effects model was created to investigate hypotheses (see “Purpose and Scope”) about how measured variables may explain variation in algal biomass across sites. Assessing the influence of temperature and nutrient concentrations was a main objective of the study and these factors were hypothesized to increase algal biomass. As a result, the modeling assessment was limited to data collected during 2019–20 when temperature and nutrient data were available, though findings from the model are used to discuss algal occurrence across all years.

Potential Controls in Algal Biomass Across Years

Reports of algal biomass from 2016 and 2017 (May and Noble, 2017; Kelley, 2017; fig. 9) and streamflow and water temperature conditions from 2015 to 2017 are incorporated into this assessment to better identify causes of variation in algal biomass across years. Observations of benthic algal blooms in the upper White River Basin have been reported since 2012 (Hodge and Eyre, 2021), though biomass samples were not collected until 2016 by CPW. In July of 2016, algal biomass exceeded the CDPHE chlorophyll a standard (150 mg/m²) at five out of six sites sampled, including one site each on the North and South Fork White River (May and Noble, 2017). Elevated algal biomass was also reported by local land managers and volunteers in 2017 (Kelley, 2017). In 2018, as part of this study, bloom conditions or near bloom conditions were measured at multiple sites, including sites on the North and South Forks and main stem White River. No exceedances of the CDPHE chlorophyll a standard were measured in subsequent years through the summer of 2021, though greater algal biomass was observed in 2021 than what was measured during sampling (represented by green algal symbol in fig. 10). The broad spatial extent of bloom conditions in 2016–18 and lack of bloom conditions in 2019–20 indicates that the factors contributing to the occurrence of algal blooms likely operate at a basin scale.

Relatively large, late, and long-lasting peak streamflows, such as those measured in 2019 (fig. 10; table 5), may limit algal blooms during the WY in which they occur and into subsequent years, as evidenced by extremely low algal biomass during the summers of 2019 and 2020 (fig. 9). High streamflows during early spring and summer of 2019 resulted in water temperatures below those favorable to C. glomerata until mid-July (fig. 3) as well as movement of the D50 and D84 at many sites across the basin (fig. 5). In 2020 water temperatures were within the same range as those measured during the previous bloom years of 2016–18 (fig. 11; table 5), indicating that the effects of streambed movement during 2019 was likely a factor limiting algal biomass into 2020. The cause may include multiple processes. Large flood events can immediately reduce algal biomass by physically scouring algae off substrata and can delay recolonization if channel morphology and stability change (Power and Stewart, 1987). Flushing flows that redistribute finer sediments rich in nutrients or changing hyporheic flow paths may also be important. Thus, although streambed movement was not a perfect predictor of algal occurrence at a site scale, the widespread occurrence of streambed movement across the basin during 2019 is likely related to the observation of reduced algae in subsequent years.

By incorporating previous accounts of years with nuisance levels of algal biomass in the basin, it is also apparent that early or low-magnitude peak streamflow conditions during the year of interest (or preceding year) are not prerequisites for algal bloom occurrence. In 2015, 2016, and 2017, peak streamflow dates and maximum streamflow values were relatively similar to the values averaged over the period of record (fig. 10; table 5). Likewise, streamflows during spring and summer were all in the normal range compared to the period of record, with the exception of a short decrease in streamflow in 2017 before the peak.