Close your eyes. Picture a mountain forest, with mossy pines and aspen dappling shade around large granite boulders. Beneath your sandals runs a cool alpine stream. As that stream flows down out of the forest, it joins others and grows, picking up melting snow and fresh rainfall along the way. Eventually, the stream grows into a river, winding slowly through open plains before flowing into a reservoir. A local city maintains the reservoir to keep itself safe from flooding and to supply drinking water for its residents, selling some water to nearby farmers. When the city opens the dam gates to release water, it runs through hydroelectric turbines that generate electricity for the region. Downstream of the reservoir, kayakers paddle in the waters and fishers wade nearby before the river passes under concrete bridges in the city and eventually flows out into the ocean.
Through this journey, the stream’s water is crucial for a diverse range of people, plants, and animals. For each of them, having too little or too much water could be catastrophic. The people who operate and oversee the world’s water infrastructure – from public agency dam operators to utilities, tribes, conservation organizations, and regulators – are charged with keeping the amount of water flowing through rivers at safe and beneficial levels.
Forecasting streamflow, then, is a vital tool because it helps water operators understand how much water will flow into their reservoir or river in the future. Knowing what’s coming is the first step in determining what to do.
There are other ways, of course, to approximate this information. Measuring snowpack in a mountain basin in winter, for instance, can help estimate the amount of water that will melt out into the river once spring comes. Weather forecasts can predict if there’s going to be a storm in the area and how intense it might be. But each of these are, on their own, incomplete and can be misleading.
In 2021, for instance, California’s Lake Oroville nearly had to shut down its hydropower generation due to low flows that were compounded by an overly optimistic snowpack readout. The state’s deputy director for the State Water Project explained that Oroville “received only 20% of the runoff that we would have expected from the snowpack we had." Scientists believe that high temperatures, dry soils, and “old rules of thumb becoming less reliable” helped account for this disparity.
Meanwhile, rain forecasts are similarly limited. Six inches of rain can produce significantly different streamflow depending on whether it lands in a meadow or parking lot and whether the soil has recently been saturated by other storms.
Streamflow forecasts, in comparison, account for each of these factors and how they relate to each other in producing the number that ultimately matters for operators: streamflow.
HydroForecast brings together the world’s best weather forecasts, satellite imagery that quantifies snow and vegetation, datasets on land use and soil characteristics, and local measurements of snow and real-time flows to predict upcoming streamflow rates and volumes with unparalleled accuracy. As the ‘old rules of thumb’ continue to become less indicative of future flows, models that incorporate the most robust information and make sense of how each variable affects streamflow will become even more crucial.
If you’d like to learn more about how improved streamflow forecasts could help your organization better manage water resources, reach out to us at team@hydroforecast.com.
1 As Colorado warms, dry soil sucks up more water. That's bad news for rivers and farmers.