Scientific Investigations Report 2006–5060
U.S. GEOLOGICAL SURVEY
Scientific Investigations Report 2006–5060
The construction, calibration, and testing of the USGS Hagg Lake model was documented by Sullivan and Rounds (2005). In short, the two-dimensional, laterally averaged model CE-QUAL-W2 version 3.12 (Cole and Wells, 2002) was applied to Hagg Lake for the years 2000 through 2003. The lake model grid was constructed by combining a bathymetric coverage of the current lake with a topographic coverage of the surrounding area, so that both the current lake and an enlarged lake could be modeled. Lake hydrodynamics, water temperature, orthophosphate, total phosphorus, ammonia, algae (2 groups), chlorophyll a, zooplankton, and dissolved oxygen were modeled, calibrated, and tested with measured lake data. Other modeled constituents included dissolved and particulate organic matter, nitrate, dissolved solids, and suspended sediment; these constituents were included in the model because of their importance to the cycles of other constituents, but were not calibrated due to a lack of data. The model accurately captured the most important seasonal and spatial influences on lake water quality.
Most model scenarios described in this report were based on the Hagg Lake model using conditions from 2002, a typical hydrologic year. Some scenario simulations also were run for 2001, a drought year. Model scenarios were run by modifying the dam height, adjusting inflows and outflows, and adding new lake outlet structures. Such changes were superimposed on the previously calibrated model for 2001 or 2002.
Although the USGS Hagg Lake model was constructed and calibrated for existing conditions, it should be able to make useful predictions of the hydrodynamic, thermal, and water quality changes resulting from a dam raise. CE-QUAL-W2 has been applied to hundreds of reservoir systems around the world (Cole and Wells, 2002). Those applications have demonstrated that when the model is provided with accurate bathymetric data, a balanced water budget, and good meteorological data, it will accurately simulate the heat budget and temperature dynamics of the system, including the timing of the onset of stratification and lake turnover in autumn. The physics of heat transfer processes are well known and have been translated into accurate mathematical algorithms in CE-QUAL-W2. Past performance and recent USGS applications of this model demonstrate its success in simulating water temperature in reservoir and river systems (Bales and others, 2001; Green, 2001; Rounds and Wood, 2001; Sullivan and Rounds, 2004; Sullivan and Rounds, 2005). As in the calibrated model, temperature predictions are expected to be accurate to within 1ºC or less as a root mean square error.
Similarly, as long as the water quality algorithms in CE-QUAL-W2 capture the most important processes affecting water quality in the lake, and as long as the expanded lake’s water quality is controlled by those same processes, the model’s water quality predictions should be useful and relatively accurate. The most important processes affecting dissolved oxygen in the existing system are water temperature and sediment oxygen demand, with minor influences from the algal community (Sullivan and Rounds, 2005). Ammonia concentrations are controlled primarily by the presence, duration, and extent of hypolimnetic anoxia. Phosphorus concentrations are affected primarily by inflows, algal uptake, and hypolimnetic anoxia. None of these influences are likely to change greatly in an expanded lake, although inputs and outputs differ between scenarios.
The greatest uncertainty in the model’s predictions for an expanded lake likely will revolve around the lake’s algal communities. The first issue is that only limited algae and chlorophyll a data were available to calibrate the Hagg Lake model. Although sampling of the lake occurred regularly and species distributions were determined, the lake was sampled at only one location and at one depth near the surface, and no samples of the outflow were taken; thus the testing of the model’s ability to simulate measured algal dynamics was somewhat limited. For instance, some algae grow best near the thermocline with the lower oxygen and higher carbon dioxide and nutrient concentrations (Horne and Goldman, 1994). Secondly, the model’s representation of algae is a necessary simplification of the real system. Although more than 80 species of algae were identified in Hagg Lake for the years modeled, the calibrated model separated that community into only two algal groups, one of blue-green algae and one representing all other algae (Sullivan and Rounds, 2005). To simulate more groups would require more data on growth rates, settling rates, and other group-specific parameters that were not available for Hagg Lake. The model was calibrated specifically to try to capture the timing and extent of the lake’s annual late-summer bloom of blue-green algae (typically Anabaena planctonica). The modeling team focused on that bloom because blue-green algae can produce toxins and often are associated with taste and odor issues in drinking-water supplies. Finally, the Hagg Lake models were run on an annual timescale, and were not designed to examine possible longer term algal species shifts following a system perturbation such as a dam raise.
Despite these limitations, changes in the algal community in an expanded lake would likely be due to influences that the model already captures with good accuracy: water temperature, light conditions, and nutrient concentrations. Model predictions of the algal community’s response to a dam raise must be viewed as an informed estimate that can be used to direct future planning and research.
For more information about USGS activities in Oregon, visit the USGS Oregon Water Science Center home page.
U.S.
Department of the Interior | U.S. Geological
Survey
Persistent URL: https://pubs.water.usgs.gov/sir20065060
Page Contact Information: Publications
Team
Page Last Modified:
Thursday, 01-Dec-2016 19:05:01 EST
