Seasonal Runoff Forecasts Based on the Climate Forecast System Version 2
Abstract
Seasonal runoff forecasts are needed for many hydroclimatological applications, such as drought outlook, agricultural planning, seasonal hydrologic prediction, and multi-purpose reservoir management. Recently, NOAA National Centers for Environmental Prediction (NCEP) has transitioned to their second generation of the Climate Forecast System (CFSv2) in operation. CFSv2 is a coupled ocean-atmosphere-land model with advanced physics, increased resolution, refined initialization, and improved land surface model, and provides forecasts up to nine months in advance. Information on the accuracy and skill of the CFSv2 forecasts is sought for the daily operation of many applications. In this study, we conduct an assessment of the prediction skill of seasonal runoff forecasts from CFSv2 using its retrospective forecasts from 1982 to 2009. Forecast skill of spatially aggregated cumulative runoff (CR) from direct CFSv2 forecasts and those obtained from the Variable Infiltration Capacity (VIC) model driven by daily precipitation, temperature, and wind forecasts from CFSv2 (i.e., hydroclimate forecasts) are compared with forecasts based on the ensemble streamflow prediction (ESP) technique. All forecasts are verified against historical VIC simulations with input forcing of precipitation and temperature derived from a set of 2131 high-quality index stations selected from the National Climatic Data Center's (NCDC's) Cooperative Observer stations across the contiguous United States. The monthly CR is spatially aggregated to 48 sub-regions created by merging the 221 U.S. Geological Survey (USGS) hydrologic sub-regions in order to evaluate regional characteristics. Preliminary results suggest that forecast skill of CR is seasonally and regionally dependent. Direct runoff forecasts from CFSv2 have the lowest skill on average, indicating limited use for hydrological drought prediction. Month-1 CR prediction from hydroclimate forecasts is superior than that from the other two forecast methods, with the lowest root-mean-square errors averaged over the 48 sub-regions for all seasons. The contributions of the improved precipitation and temperature forecasts from CFSv2 and better representation of small-scale surface characteristics from the VIC model to seasonal hydrologic prediction are investigated.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2012
- Bibcode:
- 2012AGUFM.H41A1148C
- Keywords:
-
- 1812 HYDROLOGY / Drought;
- 1816 HYDROLOGY / Estimation and forecasting;
- 1833 HYDROLOGY / Hydroclimatology