CEATI Monthly Read: Dworshak Dam, Idaho

By Laura Read on February 15, 2021

Getting to know HydroForecast through the CEATI forecasting competition

The Streamflow Forecast Rodeo is a competition hosted through a partnership between the Bureau of Reclamation and the Centre for Energy Advancement through Technological Innovation (CEATI)'s Hydropower Operations and Planning Interest Group. The Rodeo website summarizes: "[This] challenge seeks to improve the skill of short-term streamflow forecasts (10 days) via a year-long competition." Check out the map below to see the 19 hard-to-forecast sites selected for the competition.

overview ceati sites
Rodeo site map, highlighting the location of Dworshak

Each month we’ll shine a spotlight on a different site, review HydroForecast’s overall performance, and take a look at interesting events.

North Fork Clearwater River, Idaho

Join us on an adventure to North Central Idaho where elk roam the relatively uninhabited forests and lands, and native trout species meander in the tributary streams fed by the Bitterroot mountains. Waters from Clearwater, Elk River, and other tributaries collect in Dworshak reservoir, and eventually flow into the mighty Snake River. 

Basin boundary denoting the Dworshak reservoir and the tributary rivers.
Basin boundary denoting the Dworshak reservoir and the tributary rivers.

Quick basin facts

  • Total drainage area into Dworshak reservoir is 6,300 km2 (2,440 mi2).
  • Total generation capacity of Dworshak dam is 400 megawatts: enough to power roughly 300,000 homes.
  • The North Fork Clearwater watershed provides critical riverine habitat for salmon and steelhead species and forests for elk and is the home of the world’s largest steelhead trout hatchery, the Dworshak National Fish Hatchery.
  • At a height of 717 feet, the dam is the highest straight-axis gravity dam in North America.
  • Lewis and Clark camped in the area, where they rested from their trip over the Bitterroot Range and built canoes for their trip on to the Pacific Ocean.

Dworshak Dam
Dworshak Dam, Photo credit: bluefish.org

Applying HydroForecast

HydroForecast provides state-of-the-art, accurate streamflow forecasts using a hybrid approach that combines physical science with artificial intelligence. HydroForecast offers a range of advantages over existing forecasting techniques, and we've joined the CEATI competition in order to exhibit, live, these strengths. Under the hood, every forecast is created by an ensemble of neural networks that are provided different members of meteorological forecast ensembles. HydroForecast is rapid to deploy in a new basin and resilient to basin and climatic changes.

Forecast summary, October to February

The most interesting and highest inflow period of the year at Dworshak is the spring freshet (snowmelt). That period of the live competition is yet to come so we’ve evaluated our model on previous years to understand its performance by conducting what we refer to as a reforecast. We conduct a reforecast by providing the model with data inputs from remote-sensing, meteorology and streamflow observations from prior years as if it was seeing it in real-time, run the model, then compare its performance with the actual observations during that period.

Below, we highlight the active spring melt period from April to July, and show the model’s performance at the 24-hr and 48-hr lead times, respectively. These plots show the realization of our ensemble model, where we have connected all the forecasts at a certain lead time (e.g. 24-hour) over a longer time period. The plots show the observations (black), reforecasted predictions (green) and the model’s predicted confidence bounds (dark blue - 50% confidence interval, light blue - 90% confidence interval).

dworshak 24 48 hr
Inflows to Dworshak Dam at 24-hour (top) and 48-hour (bottom) lead times

These plots show how HydroForecast is able to capture each rise and fall highly accurately. They also show the noise (spikes) in the observations. These spikes are common, since reservoir inflows are generally calculated (not measured) from known reservoir elevation and outflow measurements.

Nash-Sutcliffe Efficiency (NSE) values
Nash-Sutcliffe Efficiency (NSE) values

Comparison to alternative forecasts

Over the life of the CEATI competition thus far (10/1/2020 to 2/1/2021), the "by lead time" statistics indicate Upstream Tech is outperforming the other models at Dworshak, no matter how you slice and dice the 0-10 day window. In the 0-5 day lead-times for example, we have a normalized Root Mean Squared Error of 0.21, which is 60% lower than the next closest competitor. We have the only positive NSE value in the competition, and are tracking an overall bias of only 2.52%.

Highlighting a recent event

During the snow accumulation period in the winter, flows can run an order of magnitude lower than during the active spring melt. These low flows in the winter can be challenging to model: as temperatures hover around freezing, the model needs to know whether to translate any forecasted precipitation into snow (to accumulate) or into rain (to runoff). The plots below illustrate the 10/20/2020 - 11/20/2020 period, where flows are at about 1/10th their spring volumes. We show the observations in black and two model versions: v2 (green) -- a first iteration of our hourly model and v6 (orange) - our current production model as of writing. Notice the three small spikes (under the grey bars) and the differences in how the model responds. We have tuned v6 to capture peaks more accurately at these 24-hr and 48-hr lead times.

Inflows to Dworshak Dam at 24-hour (top) and 48-hour (bottom) lead times from 10/20/2020 to 11/20/2020
Inflows to Dworshak Dam at 24-hour (top) and 48-hour (bottom) lead times from 10/20/2020 to 11/20/2020

The following panel of plots show the precipitation and temperature over the same period. Check out the oscillation in temperatures around the freezing mark (horizontal line). The three spikes highlighted above in the streamflow correspond with precipitation events shown below. A couple of things we notice:

  • Temperature: Notice that the precipitation event around 10/24 did not cause as much runoff as those on 11/6 and 11/18. The reason? The temperatures were below freezing, so most of the precipitation is accumulating as snow. The two Nov. spikes had about the same amount of precip ~2.5mm, but fell right on the borderline of 0 degrees C. Our production v6 was able to capture this runoff quite well.
  • Multiple precipitation inputs: the plots below show NOAA's GFS precipitation in green and the European Center for Medium-Range Weather Forecast (ECMWF) in purple. Disagreement between these two products helps inform uncertainty in our forecasts, whereas when the precipitation is in agreement, the model is more confident in its predictions. Including multiple precipitation products ensures that our model can learn the systematic biases in each product and how to represent those in the streamflow predictions.

Basin-scale precipitation rates (mm/hr) (top) and temperature (degrees C) (bottom) from the NOAA GFS (green) and ECMWF (purple) models
Basin-scale precipitation rates (mm/hr) (top) and temperature (degrees C) (bottom) from the NOAA GFS (green) and ECMWF (purple) models


We’ll be watching this basin as snow continues to accumulate and then melt through July. At the time of this post, we’re pleased with how the model is tracking the precipitation and snow patterns. Our near perfect NSEs during validation and low RMSEs and percent bias during this initial forecast period are a positive indication that HydroForecast is capturing the hydrologic dynamics of the Dworshak basin.

Stay tuned!