Satellite-derived parameters continue to fill in the traditionally data void areas: the oceans and polar regions. For example, geostationary satellites provide useful wind information in the low and mid-latitudes and they have made a positive impact in global numerical models. But geostationary satellites are of little use at high latitudes due to poor spatial resolution. We will present preliminary results to estimate winds in the polar regions by tracking clouds and water vapor in MODerate Resolution Imaging Spectroradiometer (MODIS) data. We are using a modified version of the automated procedure that was developed at UW-CIMSS (University of Wisconsin-Madison Cooperative Institute for Meteorological Satellite Studies) and is currently used operationally by NESDIS (National Environmental Satellite, Data, and Information Service) for tracking clouds and water vapor targets from geostationary satellite image loops. The modifications for use with MODIS data take advantage of MODIS' high spatial resolution (1 km) and its advanced capabilities for surface/cloud discrimination and cloud height determination.

The orbital characteristics that affect wind estimation from polar-orbiting satellites will also be discussed. Equatorward of 60 degrees latitude the temporal sampling of the Terra and future Aqua satellites is too sparse to obtain meaningful wind estimates. However, poleward of about 75 degrees the coverage is such that useful wind information can be obtained throughout the course of a day. Estimates of wind speed, direction, and altitude will be shown for an extended case study in both polar regions. We are working with several modeling groups and we will report on the status of assimilating these winds in their global models.