A NASA-funded study uses a computer model to understand an observed link between winter and spring snowfall in the Western U.S. and El Nino Southern Oscillation. Almost 75 to 85 percent of water resources in the Western U.S come from snow that accumulates in the winter and early spring and melts as runoff in spring and summer.

Understanding this connection and using it to predict future snowfall rates would greatly help both citizens and policy makers.

One of the missions of NASA’s Earth Science Enterprise (ESE), which funded this research, is to better understand how the Earth system is changing. Within this framework, NASA is committed to studying variability in the water cycle, how well we can predict future changes in the earth system and the consequences of change in the Earth system for human civilization.

Lead authors Jiming Jin and Norman Miller of the U.S. Department of Energyís Lawrence Berkeley National Laboratory, Berkeley, Calif., in collaboration with Soroosh Sorooshian and Xiaojang Gao at the University of Arizona, Tucson, find that higher and lower tropical Pacific sea surface temperatures (SSTs) that characterize El Nino and La Nina change atmospheric wind patterns in the mid-latitudes in winter and spring, shift the way moist air gets transported in the atmosphere, and directly affect Western U.S. precipitation and snow accumulation.

El Nino / Southern Oscillation (ENSO) marks a see-saw shift in surface air pressure between Darwin, Australia and the South Pacific Island of Tahiti. When the pressure is high at Darwin it is low at Tahiti and vice versa. El Nino, and its sister event La Nina, are the extreme phases of this southern oscillation, with El Nino referring to a warming of the eastern tropical Pacific, and La Nina a cooling.

By better understanding the connections between these processes, scientists can update their computer climate models to improve their ability to forecast future snowfall and water availability in the west.

“If the computer climate models can accurately describe the processes that connect ENSO and snowpack in the Western U.S., then the model can be used to predict the impact of ENSO on snowfall in those areas,” said Jin. “In addition, the model can give us more detailed information than observations, which can lead to a further understanding of those observed processes.”

The researchers entered over 45 years of data from 1949 to 1995 into their computer climate model. They included observed global sea surface temperatures, wind data, the amount of water contained in snowpack for the beginning of the first four months of each year from over 300 western U.S. field sites, and precipitation and surface air temperature observations.

They also used NASA-derived Normalized Difference Vegetation Index (NDVI) data for improved model predictions. NDVI measures the amount of solar energy reflected and absorbed by vegetation. NDVI was created by Compton Tucker of NASA Goddard, using data from the National Oceanic and Atmospheric Administrationís (NOAA) Geostationary Environmental Orbiting Satellite (GOES) Advanced Very High Resolution Radiometer (AVHRR) instrument. The data used in this research also comes from the Moderate Imaging Spectroradiometer (MODIS) instrument aboard NASAís Terra satellite.

During the weak ENSO episodes, Washington, Oregon, Idaho, and Montana experienced decreased precipitation during weak El Ninos, and increased precipitation during the opposite La Nina phase. During the strong El Nino episodes, stronger winter and spring precipitation was found south of Sacramento, including parts of California, Nevada, Utah, Colorado and all of Arizona and New Mexico. However, during strong La Nina events, the researchers did not find any changes to precipitation patterns in the western U.S.

The model matched well with actual observations except when it came to weak ENSO episodes, including both El Nino and La Nina. During those events, the mid-latitude atmosphere in the model reacted too strongly to the shifts in tropical Pacific SSTs, and moist air masses from that region moved incorrectly. Still, it shows that different intensities of ENSO episodes have differing affects on western U.S. snowfall. The researchers hope to fine-tune the model’s responses in the future.

This research may yield a forecast tool that greatly benefits citizens and water resource managers in the Western U.S. Jin and Miller are currently developing new snow assimilation techniques that show improved forecast skill, which they hope will make water allocation decisions more accurate and cost efficient.

These findings were presented at the 83rd Annual Meeting of the American Meteorological Society in Long Beach, Calif.

This research was funded by NASA’s ESE Interdisciplinary Science and Applications Programs. The Applications Division applies the results of the nation’s investment in ESE to issues of national concern, such as environmental quality, resource management, community growth, and disaster management to support policy makers at the state and local levels.

For more information, please see:
http://www.gsfc.nasa.gov/topstory/2003/0210snowpack.html

Normalized Difference Vegetation Index (NDVI)
http://daac.gsfc.nasa.gov/CAMPAIGN_DOCS/LAND_BIO/ndvi.html

NASAís Terra Satellite
http://terra.nasa.gov

The Advanced Very High Resolution Radiometer
http://www.ngdc.noaa.gov/seg/globsys/avhrr.shtml