Water Supply Forecasting
Water supply and water management in the West
Most of the western United States has a dry climate, where agriculture depends on irrigation. The water supply for irrigation largely comes from rivers and creeks, whose streamflow originates from the springtime melting of winter snow.
A complex network of dams and canals has been built since the beginning of the 20th century to store and distribute this water for irrigation and other uses (such as hydropower). Forecasting of expected streamflow is an important aid in planning water storage and release as well as agricultural planting decisions.
Water supply forecasting overview
A water supply forecast is a prediction of streamflow volume that will flow past a point on a stream during a specified season, typically in the spring and summer. The Natural Resources Conservation Service (NRCS) issues water supply forecasts for hundreds of points in the western United States near the first of the month between January and June each year and at other times as requested.
The basis of water supply forecasting lies in the fact that most of the streamflow in western North America originates as winter snow. This snowpack serves as a natural reservoir, storing water during the winter and releasing it during the spring and summer snowmelt season. The delay between when the snow accumulates and when it melts is what makes it possible for hydrologists to use measurements of winter snowpack to make predictions of spring and summer streamflow.
In some areas, however, snowmelt is not as dominant, and therefore forecasting is more difficult. For example, on the west side of the Cascade Mountains, the east slope of the Rocky Mountains, and in parts of the Southwest, winter, spring, and/or summer rainfall can supply a significant amount of the streamflow volume. Because the source of this runoff occurs in the future at the time forecasts are issued, and it cannot easily be predicted, forecast uncertainty is higher in these areas than in dominantly snowmelt basins.
Water supply forecasts can, therefore, vary in accuracy depending on the strength of the snowmelt component. Nevertheless, useful predictions can be made in most areas of the West.
How water supply forecasts are produced
The NRCS uses statistical models to produce water supply forecasts. These are equations that express a fitted mathematical relationship (usually linear) between several predictor variables and the target seasonal streamflow volume. Predictor variables are primarily snow water equivalent at selected measurement sites but can also include precipitation, antecedent streamflow, and a few other miscellaneous quantities.
The primary data source for predictor variables is the NRCS Snow Telemetry (SNOTEL) system and manual snow courses; other data come from various Federal, state, and Canadian provincial agencies. Data for the target seasonal streamflow volume come from the U.S. Geological Survey and other Federal and state agencies. In addition, the target streamflow volumes are usually adjusted for the effects of human water management, such as changes in reservoir storage and irrigation canal diversions, to correspond, as closely as possible, to natural flow conditions.
Statistical forecasting models are developed using spreadsheets and custom software. Predictor variables are carefully selected to balance the multiple goals of forecast accuracy, month-to-month consistency, spatial representativeness, and physical understandability.
During forecast operations, hydrologists retrieve the necessary data and run the forecast models. They then review the results in consultation with the state NRCS Water Supply Specialists. If needed, adjustments are made, and then the forecasts are published.
Interpreting water supply forecasts
All forecasts have uncertainty. For water supply forecasts, the sources of uncertainty include unknown future weather, model simplifications, and data limitations (error or inadequate spatial coverage).
To express this uncertainty, the forecast is presented not as a single value but as a range of values, each with a specific probability of occurrence.The wider the spread among these values, the more uncertain the forecast. As the season progresses, forecasts generally become more accurate.
It is important for forecast users to understand this uncertainty and take it into account when making decisions so as to manage the level of risk they are willing to accept in receiving more or less water than planned for.
Five forecast values are published for each forecast point and each forecast period:
Note: There is still a 20% chance that the actual streamflow volume will fall either below the 90% exceedance forecast or above the 10% exceedance forecast.