Surface-water quality and flow Modeling Interest Group
Fig. 1
This fact sheet summarizes some of the capabilities that were added to HSPF to simulate complex reservoir operations in the upper Truckee River Basin in eastern California. Examples of such operations include releases based on water-storage and flood-control criteria; releases to meet agricultural, municipal and industrial, and hydropower demands; and exchanges of water categories1 between reservoirs. In this summary, simplified or isolated examples are used to illustrate specific reservoir operations that can be represented by HSPF. Actual day-to-day system management or simulation is more complex. It requires consideration of present and forecast flow conditions and of reservoir stages and volumes at various locations in the river basin as well as compliance with numerous legal agreements, legal decrees, and demands.
HSPF was chosen to simulate Truckee River operations primarily because it can (1) simulate continuously over time, including periods of storm runoff and low flows, (2) simulate at a variety of time steps, including daily and hourly, (3) simulate the hydraulics of complex natural and manmade drainage networks, (4) produce simulation results for many locations along the river and its tributaries, and (5) compute a detailed water budget that accounts for inflows and diversions as well as different categories of water in the river and associated reservoirs.
HSPF uses conditional logic to simulate reservoir operations (Tom Jobes, Aqua Terra Consultants, written commun., 1995). Accordingly, if certain conditions are met, then certain actions are taken. Conditions that are evaluated during simulations include the time of year; reservoir stage, reservoir storage, or volume of a given water category in a reservoir; streamflow magnitude, temperature, or water quality at a given location; and fulfillment of water demands. Thus, for example, release from the Lake Tahoe category "pooled water" could be programmed to occur if the date is during the period from April 1 through October 31 of a given year, if the elevation of Lake Tahoe is within the range 6,223.0 to 6,225.5 feet, and if the demand for additional water at the Farad gaging station (fig. 1) has not been met.
The following examples show results from preliminary HSPF simulations of Truckee River operations based on selected existing legal decrees and regulations. These examples are intended to illustrate only how HSPF can simulate operations rather than to convey citable or quantitative model results.
Daily reservoir elevations simulated by HSPF on the basis of Prosser Creek Reservoir flood-control criteria are shown in figure 2. The period from August 7 to November 1 includes the precautionary drawdown period, during which the reservoir elevation is lowered to 5,703.7 feet to allow storage room for winter and spring inflows. This elevation is maintained from November 1 to April 10 by releasing all reservoir inflows. From April 10 to May 20, water storage is based on rules governing maximum daily storage. From May 21 to August 6, the reservoir typically is operated more to meet downstream water demands than to maintain flood-control space.
Fig. 2
A simplified graph of daily water storage simulated for Donner Lake and Prosser Creek Reservoir by HSPF on the basis of legally decreed storage priorities is shown in figure 3. Donner Lake has an earlier storage priority than that for Prosser Creek Reservoir and thus can begin storing water sooner. From April 3 to 14, water storage remains constant at the winter-capacity storage: 2,500 acre-feet at Donner Lake and 9,800 acre-feet at Prosser Creek Reservoir. On April 15, Donner Lake storage begins to increase while Prosser Creek Reservoir storage remains constant. Because of a later storage priority, Prosser Creek Reservoir storage does not begin until April 20.
Fig. 3
An example of an HSPF simulation that illustrates how water stored in Lake Tahoe and in Boca and Prosser Creek Reservoirs can be released to attain Floriston rates for the period August through October is shown in figure 4. Simulated reservoir releases from pooled water compensate for and react to the variability of natural inflow to the Truckee River to approximate Floriston rates at the Farad gaging station (fig. 4A). The simulated flow at the Farad gaging station consists of the reservoir releases (fig. 4B) plus the natural, unregulated inflow to the Truckee River, routed downstream to the Farad gaging station.
Fig. 4
Natural, unregulated flow into the Truckee River can be augmented by four categories of pooled water stored within the two reservoirs and Lake Tahoe to maintain Floriston rates. These categories are Lake Tahoe pooled water (hereafter abbreviated "Tahoe_pooled"), Lake Tahoe exchange water stored in Prosser Creek Reservoir ("Prosser_exchange"), and two categories of Boca Reservoir water called adverse-to-canal ("Boca_adverse") and nonadverse-to-canal ("Boca_nonadverse"). Under a complicated set of provisions within the Truckee River Agreement of 1935 and other legal decrees, orders of release to maintain Floriston rates are assigned to these water categories for different conditions and time periods of a given year (fig. 4). If the rates cannot be achieved by natural, unregulated flow, Tahoe_pooled is the first choice for release. However, hydraulic properties of the outlet may limit release rates from Lake Tahoe. Therefore, in this example, releases from Boca and Prosser Creek Reservoirs must augment Tahoe_pooled releases to maintain Floriston rates. Water from the category Prosser_exchange may be released to augment releases from Lake Tahoe. Although not formally addressed in legal decrees or regulations, the U.S. District Court Water Master attempts to maximize recreational use of Prosser Creek Reservoir from April 1 through Labor Day. Therefore, Prosser_exchange releases are made, if possible, only from nonrecreational storage in the reservoir to attain Floriston rates in the HSPF simulation. Nonrecreational storage in the reservoir is that volume of water above 19,000 acre-feet that is less important for recreational activities (such as fishing, boating, and swimming). HSPF simulates this condition for the recreation season by the release of nonrecreational Prosser_exchange, which has been assigned the second highest order of release to maintain Floriston rates.
The third water category in line for release to maintain Floriston rates is Boca_adverse water. Boca_adverse water is the first 25,000 acre-feet of pooled water stored in Boca Reservoir. The category name comes from the fact that it has a higher storage priority than water diverted to the Carson River Basin via the Truckee Canal. Therefore, as storage of nonrecreational Prosser_exchange water becomes depleted (mid-August in this example), Boca_adverse water is released.
In the model, Labor Day triggers a change in the choice for release from Boca_adverse water to recreational Prosser_exchange water (fig. 4). Boca_adverse water releases are reduced to zero while recreational Prosser_exchange water releases are increased to help achieve Floriston rates. As storage of recreational Prosser_exchange water becomes depleted, Boca_adverse water is released again. Finally, as Boca_adverse water becomes depleted, pooled water in excess of the first 25,000 acre-feet, that is, Boca_nonadverse water, is released.
The Tahoe-Prosser Exchange Agreement of 1959 specifies the operation of Lake Tahoe and Prosser Creek Reservoir in order to meet multiple uses. This agreement requires releases from Lake Tahoe to maintain a minimum instream flow during periods when water would otherwise be stored and accumulated for later release. In exchange, the agreement requires that an equivalent volume of water be stored concurrently in Prosser Creek Reservoir in order to compensate for the release from Lake Tahoe. The water stored in Prosser Creek Reservoir through this exchange then is used as though it were Lake Tahoe storage and accounted for as a distinct water category called Prosser_exchange water.
How HSPF simulates the linked operation of Lake Tahoe and Prosser Creek Reservoir is shown in figure 5. The agreement requires a minimum instream flow below Lake Tahoe of 50 cubic feet per second between October 1 and March 31 and a flow of 70 cubic feet per second for the remainder of the year. In this simplified example, HSPF simulated the release of water from Lake Tahoe to maintain Floriston rates until May 24 and thus met the instream-flow requirement specified in the agreement. Therefore, the storage of Prosser_exchange water remained constant through May 24. After May 24, most of the water released from Lake Tahoe was no longer required for Floriston rates, so Lake Tahoe outflows could have been reduced to zero on certain days. Instead, the release from Lake Tahoe was maintained at 70 cubic feet per second. Storage of Prosser_exchange water increased after May 24 at rates corresponding to that fraction of Lake Tahoe releases made solely to meet the instream-flow requirement.
Fig. 5
To model this exchange of water, HSPF determines the minimum-flow requirement for a given date, the present rate of release from Lake Tahoe, and the availability of releasable storage in Lake Tahoe and of storable water in Prosser Creek Reservoir. If the conditional logic specified for the model determines that an exchange should and can be made, then Lake Tahoe releases for minimum instream flows are exchanged for Prosser Creek Reservoir water and designated as Prosser_exchange water. The model must track the accumulation and later release of this Lake Tahoe exchange water (stored in Prosser Creek Reservoir) through time.
Bohman, L.R., Berris, S.N., and Hess, G.W., 1995, Interactive computer program to simulate and analyze streamflow, Truckee and Carson River Basins, Nevada and California: U.S. Geological Survey Fact Sheet FS-165-95, 4 p.
-Steven N. Berris, Glen W. Hess, and Kenn D. Cartier
Fact Sheet FS-082-96
February 1996
Back to the SMIG Features Page
Home | Mailing List | Features | Conferences | Classes | Reading | Model Archives | Feedback