High-Precision Reservoir Release Scheduling Model with Real-Time Tributary Forecasts

Project ID: 1839
Principal Investigator: Shane Coors
Research Topic: Ecosystem Needs
Funded Fiscal Years: 2006
Keywords: None

Research Question

Highly specific and operationally challenging criteria due to ecological responsibilities are being placed upon Reclamation managers and operations staff throughout all regions. To achieve these operational criteria, a prototype model has been developed that:

* Is driven by real-time hydrologic and climactic data (not historic averages)

* Incorporates innovative high-precision hydraulics

* Provides an interface for operators to design short-term (5-day) hourly release schedules

The model has been developed and tested in multiple basins and has proven to be very useful in real operations. However, of the four components that make up the model, one-- the tributary forecast generation component-- is still in its infancy.

This Science and Technology (S&T) Program research project will automate the assigning of the parameters required to generate a four-day headwater forecast. An accurate tributary forecast is required in order to set releases to precisely meet ecologically driven operational goals.

Need and Benefit

Throughout Reclamation, one of the foremost challenges in operating our facilities is to minimize any adverse impacts they have to the ecosystems in which they operate. As continuing research allows biologists to better identify ecological problems and more precisely specify operational objectives to mitigate these negative impacts, Reclamation's capability to meet these operational objectives must also advance. The operational criteria being placed upon Reclamation facilities due to ecological considerations are increasingly precise and demanding. Developing innovative and more precise modeling techniques that operate on a shorter time scale will allow Reclamation to meet these ecological criteria fully and precisely, while avoiding negative impacts to other water delivery responsibilities.

A good example is on the Green River below Flaming Gorge Reservoir, which is home to three endangered fish species and has provided the impetus for initial development of the prototype model. Significant research and cooperation between Reclamation and others resulted in a comprehensive recovery plan. One of the primary operational objectives set forth in the recovery plan is to achieve prescribed peak flows and durations in a critical section of the river that is over 100 miles downstream from the dam and 30 miles downstream from the Yampa confluence. To optimally achieve these flow targets, the peak releases from the reservoir must be timed to precisely coincide with the natural spring peak flow from the Yampa River. Successfully implementing this operation requires the capacity to "see" the Yampa peak coming roughly three days ahead of time, time release changes from the dam to reach the confluence (18-24 hour travel) at precisely the right time, and set and adjust the release magnitudes such that the targets are met but not exceeded.

Prior to the initial development of the model discussed herein, Reclamation did not possess the technology to operate with the precision required to efficiently implement this recovery program. Within the water resources community, there did not seem to be a modeling system that operates on an hourly time scale and is driven by real-time hydrologic and climactic data.

Benefits to Reclamation from possessing the ability to operate with this precision come in the form of appreciable water savings, more consistent achievement of ecologically driven operational objectives, and minimized bypasses of power turbines. In short, as we operate more precisely, water is conserved. In the San Juan basin, operations of Navajo Reservoir in 2001 (which intended to match the spring peak release with the peak outflow of the Animas River, a downstream tributary) produced a peak flow below the confluence of 8, 000 cubic feet per second (cfs) for one day and flows greater than 7, 000 cfs for 10 days while releasing about 95, 000 acre-feet (AF) over 10 days. Had the operations been done with the precision real-time modeling system advocated herein, peak flows of 8, 000 cfs could have been attained for 10 days (which is one of the flow recommendations from the San Juan River Basin Recovery Implementation Program [SJRIP]) with only 81,000 AF of Navajo releases. Thus an additional 13,000 AF was released beyond what was necessary to fully meet the objective, and the flow recommendation was still not achieved. Further, to achieve this objective (8,000 cfs for 10 days) given the actual release timing of 2001, it would have required 125,000 AF of total releases. This means an additional 44,000 AF would have required to achieve this target. This over-release occurred because Reclamation did not have the capacity to precisely match Navajo releases with the natural Animas peak.

Because operational constraints are becoming increasingly precise and demanding throughout all regions of Reclamation, benefits from fully developing this technology will be realized Reclamation-wide.

Contributing Partners

Contact the Principal Investigator for information about partners.

Research Products

Please contact research@usbr.gov about research products related to this project.

Return to Research Projects

Last Updated: 4/4/17