Linking Snowfall and Snow Accumulation to Generate Spatial Maps of SWE and Snow Depth
Quantifying snow amount (including water equivalent and depth) on the ground is important for water resources. In some regions, snowmelt causes the largest annual flows in streams and rivers. Snow also affects the energy and water cycle between land and atmosphere by changing the surface physical properties, such as the reflectance of solar radiation, thermal conductivity, infrared radiation emissivity, and roughness. This effect remains significant even after the snowmelt (through soil moisture), and this memory provides a potential source of predictability at weekly to seasonal time scales. Despite its importance, the snow measurements still contain large uncertainties. In situ snow measurements are highly accurate but not representative over mountainous areas. Satellite remote sensing gives much better spatial coverage but with much more uncertainty. Another approach is to link snow estimates to precipitation and other atmospheric information through the data assimilation in a land model or land-atmosphere coupled model, but there is still considerable uncertainty in these estimates.Here Broxton et al. develop an innovative approach to link point measurements and areal estimates of snow amount based on the above "old idea" of connecting snow amount with precipitation. Compared with the substantial horizontal variability of snow amount itself over mountainous areas, snow amount normalized by accumulated snowfall are much more spatially consistent. Therefore, the normalized snow amount, rather than the snow amount itself, is spatially interpolated in this study. Using in situ measurements from two national networks in the U.S., this method is found to be much more accurate than four commonly used interpolation methodologies that consider the dependence of snow on horizontal distance and elevation but do not use the snowfall information. Furthermore, fewer measurement sites are needed for the interpolation. This method is then used to produce 2.5 arcminute (~3 km) daily maps of snow water equivalent over the western U.S. based on gridded data of precipitation and temperature (needed to estimate ablation) as well as in-situ point snow measurements. It also ensures that the snow water equivalent remains consistent with the precipitation data and ablation estimate.These maps are shown to be comparable to another gridded product that is based on the assimilation of much more snow data in a more complicated energy balance snow model. These gridded snow data are expected to have a variety of applications, such as snow monitoring, snow initialization in operational weather and climate prediction models, and the evaluation of snow amount from weather and climate models, reanalysis, and land data assimilation systems.
- Article Category
- Research Articles
Linking snowfall and snow accumulation to generate spatial maps of SWE and snow depth
- First Published:
- | Vol:
- | DOI:
Eos.org: Earth & Space Science News
Submit a manuscript
AGU Career Center
John Orcutt, ESS Editor-in-Chief
Download the App
New Android App Available!
iOS App for iPad or iPhone