Statistical Applications of Physical Hydrologic Models and Satellite Snow Cover Observations to Seasonal Water Supply Forecasts
Abstract
Despite research demonstrating the value of physically based hydrologic models and satellite observations of snow covered area (SCA) for water supply forecasts, these tools remain underutilized by operational agencies. Part of the reason for this disparity lies in the contrast between experimental forecasting techniques, which tend to employ methods such as ensemble streamflow prediction, and operational forecasting systems that rely on simpler methods like statistical regression to link surface observations of snow water equivalent (SWE) to seasonal runoff volumes. We explore methods that bridge this gap via a hybrid approach which uses model-simulated snow states and raw satellite data as predictors in regression models that are adapted to the operational environment. Our study areas are the Sacramento and San Joaquin River basins, managed by California's Department of Water Resources (DWR), and responsible for approximately 50% of the state's runoff. In our first approach, regression models are trained on SWE and SCA output from the Variable Infiltration Capacity (VIC) macroscale hydrology model, which is forced by daily precipitation and temperature observations from 1915 to present. Our second approach uses output from a VIC model that has been updated with SCA data from the Moderate Resolution Imaging Spectroradiometer (MODIS), whose 500 m spatial resolution and daily temporal resolution provide near-ideal conditions for hydrologic applications. Our third approach bases regression models directly on raw MODIS SCA data. For these last two approaches, we describe efforts to extend the relatively short record of MODIS data (2000 to the present) with data from the 1 km Advanced Very High Resolution Radiometer, and the 1 km SCA product from the National Operational Hydrologic Remote Sensing Center. In each case, we compare the skill of retrospective forecasts based on the hybrid regression models to those based on DWR's existing regression models. We give particular attention to performance of the alternative methods early and late in the forecast season when observing stations used in the DWR regressions often are snow free, but seasonal forecasts are nonetheless useful.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2008
- Bibcode:
- 2008AGUFM.H41B0871R
- Keywords:
-
- 1816 Estimation and forecasting;
- 1847 Modeling;
- 1855 Remote sensing (1640);
- 1863 Snow and ice (0736;
- 0738;
- 0776;
- 1827);
- 1884 Water supply