Developing Tools for Improved Water Supply Forecasting in the Rio Grande Headwaters

Principal Investigator

David Clow

Start Date

2016-08-03

End Date

2020-09-30

Release Date

2016

Snow is a critical water resource in the western United States (U.S.). Seasonal snowpacks serve as large natural reservoirs in the Rocky Mountain west that store water through the winter and release it during spring and summer months when demand is greatest; melting snowpacks often supply water to meet demands hundreds of miles away from the source (Mote 2006). In the Rio Grande Basin (RGB), snowmelt from seasonal snowpacks in the headwater area supplies 50 – 75% of annual basin streamflow (Rango 2006); therefore, water supply forecasts provided prior to the snowmelt season are of upmost importance for water management decision making. Errors in these statistically-based water supply forecasts are ubiquitous, and forecasting skill has declined over time in certain areas of the country such as Colorado and New Mexico (Pagano et al. 2004). Process changes related to changing climate, land cover, and/or dust deposition, or operational changes in data collection and forecasting may be responsible for this worsening of forecast skill (Pagano et al. 2004). Understanding the conditions that have led to large errors in existing forecasts is an important step towards making science-based recommendations for improving future water supply forecasts in the RGB. This study proposes to evaluate the performance of various operational products for forecasting spring and summer streamflow in the Rio Grande Headwaters basin at the Rio Grande Near Del Norte gage from 2000 through 2015. The products that will be evaluated include the NRCS water supply forecasts, NWS River Forecast Center forecasts, and outputs from the NOHRSC Snow Data Assimilation System (SNODAS). In this study, we will quantify errors in each product in recent years, and will investigate conditions and processes associated with large forecasting errors. Additionally, in collaboration with the USGS National Water Census Program and the NRCS Colorado Snow Survey Program, we propose to test coupling of a physically-based snowpack evolution model (SnowModel) with a physically-based hydrologic model (Precipitation Runoff Modeling System, or PRMS). We will compare streamflow runoff estimates derived from this physically-based modeling approach to those from existing, empirical operational forecasting techniques for the 2016 and 2017 water years. This research will provide recommendations on (1) what currently operational forecasting method provides the best performance over a range of conditions in the Rio Grande Headwaters, and (2) how operational forecasts in the RGB could be improved using new technologies and methods.

ScienceBase WMS

• National and Regional Climate Adaptation Science Centers
• South Central CASC