Surface-water quality and flow Modeling Interest Group

Modeling Discharge, Heat, and Water Quality in the Tualatin River, Oregon, with CE-QUAL-W2

USGS, Water Resources Division
10615 SE Cherry Blossom Drive
Portland, OR 97216
Internet: sarounds@usgs.gov, tmwood@usgs.gov, ddlynch@usgs.gov
Phone: (503) 251-3280, 251-3255, 251-3265
Fax: (503) 251-3470

Rounds, S.A., Wood, T.M., and Lynch, D.D., 1999, Modeling discharge, temperature, and water quality in the Tualatin River, Oregon: U.S. Geological Survey Water-Supply Paper 2465-B, 121 p.

A laterally averaged, two-dimensional model was used to simulate discharge, temperature, and water quality in the Tualatin River (map, 8K GIF) during the summers (May-October) of 1991, 1992, and 1993. During low-flow periods, the lower main stem of the Tualatin River (river miles 38.4 to 3.4) is characteristic of a long, slow-moving lake. Water-quality problems encountered during the summer include intermittent violations of the State of Oregon minimum dissolved oxygen and maximum pH standards, exceedances of the action level for nuisance phytoplankton growth, and impairment of several of its designated beneficial uses (aesthetics, aquatic life, and water-contact recreation). This river was modeled with a modified version of CE-QUAL-W2, a U.S. Army Corps of Engineers reservoir model that is appropriate for use on the lower main stem of the Tualatin River. Eighteen water-quality constituents were simulated:

• chloride
• suspended solids
• dissolved solids
• labile dissolved organic matter
• algae
• detritus
• soluble orthophosphate
• ammonia
• nitrate
• dissolved oxygen
• bottom sediment
• total inorganic carbon
• carbonate alkalinity
• pH
• carbon dioxide
• bicarbonate
• carbonate
• zooplankton

The model was calibrated for 18 months of data: May through October in each of 1991, 1992, and 1993. Only six calibration parameters were used for the water-quality routines in the model; of these, the model was most sensitive to the maximum algal growth rate and the zooplankton mortality rate. Values for most of the parameters required by the model were either independently measured or taken from the available literature. The maximum algal growth rate was varied over the summer in accordance with observed patterns in measured primary productivity rates. The zooplankton mortality rate was kept constant throughout each summer but was calibrated to different values for each year due to differences in observed zooplankton population levels. The calibration process resulted in a model that performed very well; it captured the dynamics of the most-important water-quality processes in the lower main stem Tualatin River during each of three hydrologically distinct summers. Accuracy in day-to-day fluctuations was sacrificed somewhat in order to more accurately simulate the overall cycle of algal growth for these 18 months while varying the calibration parameters only as absolutely necessary. This level of accuracy was sufficient to simulate the interactions among nitrogen, phosphorus, phytoplankton, zooplankton, and dissolved oxygen, but was insufficient to accurately simulate pH during every algal bloom. The ability to extrapolate beyond the calibrated conditions of the model, however, was determined to be more important than accurately simulating the pH during the shorter time scales of individual algal blooms; therefore, this calibration philosophy was retained, and the further calibration of pH was not pursued.

Using the model as a diagnostic tool, a number of general conclusions were made during the calibration process:

• Water quality in the lower main stem Tualatin River is dominated by three physical constraints and conditions - residence time, air temperature, and solar insolation. Given ample nutrients in conjunction with the long travel time, warm climate, and sunny days frequently encountered during the summer low-flow period, phytoplankton blooms of sufficient size to have an important influence on water quality will develop.

• Carbonaceous biological oxygen demand (CBOD) and sediment oxygen demand (SOD) are the most important oxygen consumption processes. Instream nitrification was negligible most of the time due to the low concentrations of ammonia found in the river during these summers. Reaeration is a slow process in the modeled reach; the slow reaeration rates are important in determining instream dissolved oxygen concentrations.

• The pH in the lower main stem cannot be modeled well unless the algal dynamics are simulated accurately during each algal bloom. That level of short-term accuracy was not required to model other constituents well and was somewhat incompatible with the goal of modeling trends and water-quality changes on longer time scales.

• During most of the modeled time period, algal growth was limited only by light conditions. Only during large algal blooms and near the surface of the river was phosphorus found to limit algal growth. Data to substantiate the existence of such a transitory phosphorus limitation to algal growth in the Tualatin River, however, was not available. The model indicates that phosphorus can limit the peak size of algal blooms and therefore can be used to limit the number and frequency of violations of the State of Oregon maximum pH standard.

• With respect to the State of Oregon minimum dissolved oxygen standard, the phytoplankton were found to be important both in their presence and their absence. Violations of the standard in midsummer were usually associated with the crash of a large algal bloom; such violations were normally of short duration although the minimum dissolved oxygen concentration associated with a large crash may be less than 4 mg/L. Violations of the standard during September and October, on the other hand, were normally associated with low populations of phytoplankton. The CBOD and SOD consumed more oxygen during those periods than the small populations of phytoplankton were able to produce via photosynthesis. This period was often characterized by dissolved oxygen concentrations near or below the standard for extended periods of time.

• The model confirmed that a large nonpoint source of phosphorus from ground water and small, ungaged tributaries is present within the lower main stem reach of the Tualatin River. The phosphorus and water budgets cannot be balanced without it.

• Few of the scenarios tested for this report have significant effects upon dissolved oxygen conditions in the main stem.

• During September and October, the most significant improvements in dissolved oxygen (as much as 1 mg/L) were obtained only through a large amount of flow augmentation (minimum flow of 200 ft3/s at river mile 38.4), or through a lesser amount of flow augmentation (minimum flow of 150 ft3/s at river mile 38.4) combined with a reduction in the loads of CBOD from the boundaries.

• For the period May through August, several scenarios showed some ability to limit algal growth during large blooms. When these scenarios failed to reduce the impact of the background oxygen demands (SOD, CBOD), however, dissolved oxygen concentrations between algal blooms still showed a tendency to decrease to near-problem levels.

• Phosphorus reduction scenarios showed that if the total phosphorus TMDL (total maximum daily load) is achieved at the boundaries to the main stem Tualatin River and the wastewater treatment plants are efficiently removing phosphorus from their effluent and meeting their wasteload allocations, then the main stem river will be in compliance with the TMDL. Even if the TMDL is achieved, however, the predicted effect on dissolved oxygen concentrations is unclear. If particulate and organic phosphorus is removed rather than soluble orthophosphate, then dissolved oxygen conditions will improve, especially in October, primarily because CBOD will be removed. If soluble phosphorus is removed instead, then dissolved oxygen conditions may actually worsen because of reduced photosynthetic production of oxygen without the loss of CBOD at the boundaries.

• The most promising scenarios, in terms of providing the most improvement in dissolved oxygen conditions, most likely will include both a decrease in residence time via flow augmentation and a decrease in the background oxygen demands (CBOD and SOD).

Back to the SMIG Features Page

Home | Mailing List | Features | Conferences | Classes | Reading | Model Archives | Feedback