Recorded presentationJ4.6
Verification of NAEFS land-surface forecasts
Wanru Wu, NOAA/NWS/NCEP/CPC, Camp Springs, MD; and K. Mo
To provide experimental objective drought monitoring and outlook over the United States and Mexico in support of the National Integrated Drought Information System (NIDIS), this study explores the possibility to apply the North American Ensemble Forecast System (NAEFS) for short-range drought outlook updating daily. The NAEFS combines state-of-the-art weather forecast tools developed at the United States National Weather Service and at the Meteorological Service of Canada to provide numerical weather prediction products in both countries for a forecast period of 1-14 days (see details at http://www.emc.noaa.gov/gmb/ens/NAEFS.html). Means and spreads for week1 (1-7 days) and week2 (8-14 days) forecasts were calculated from 60 ensemble members daily with 1x1 degree spatial resolutions and 6-hour temporal intervals. The land-surface variables such as precipitation, soil moisture, evaporation and surface air temperature were verified against the North American Regional Reanalysis (NARR) for the summer months. The forecast errors were large but mostly systematic, especially for soil moisture and evaporation. The error correction was not sensitive to the training period as long as the training period is longer than about 2 weeks. The surface variables were not initialized in the NAEFS forecasts; the fields were taken from the previous forecast cycle. The errors in the initial conditions were examined. The results showed that a large portion of the systematic errors was inherited from the initial conditions. For the most regions of the United States, the forecast skills of surface temperature and precipitation within 14 days largely depend on the model's ability to predict the atmospheric circulation. The initial errors in soil moisture and evaporation do not affect the overall forecast skills of precipitation and temperature. However, soil moisture initialization will improve forecasts over the areas where the land-atmosphere coupling is strong such as the Central United States and Northwest Mexico. 30-day training period for bias correction has been applied to the experimental operational drought forecasts at http://www.cpc.ncep.noaa.gov/products/Drought. Recorded presentation
Joint Session 4, Land-Atmosphere Interactions, Part IV
Wednesday, 23 January 2008, 8:30 AM-10:00 AM, 224
Previous paper