Freezing and thawing processes of soils have a significant impact on energy and water transfer between the atmosphere and the land surface, surface hydrology, and agriculture because of substantial differece in thermal and hydrological properties between frozen and thawed soils. The timing, duration, areal extent, and depth of surface soil freezing and thawing are important parameters for regional climatic and hydrologic studies, and changes in these parameters are important climatic indicators and integrators. The conventional point measurements of surface soil freezing and thawing provide information for process studies in local areas. It is impractical to conduct large- or regional-scale investigations of surface soil freezing and thawing by using the point measurement method. In this study, onset of surface soil freezing in permafrost regions and frozen soil extent in seasonally frozen soil regions were investigated using SSM/I data. The criteria used to detect surface onset of freeze and frozen extent includes (i) a negative spectral gradient between Tb(37V) and Tb(19V), (ii) a Tb(37V) cutoff. The algorithm was initially validated against field measurement in northern Alaska for permafrost regions and
in northern Plains of the United States for seasonally frozen soil regions. The preliminary results indicate that frozen soil extent as determined by the algorithm is generally in good agreement with the zero deg. Celius isotherm of air temperature. The boundary of frozen soil extent moves southward as the cold air invades from north in northern Plains during the early winter. The results also show that frozen soil extent was greater in the morning (about 6:00 a.m. local time) than in the afternoon (about 6:00 p.m. local time). Because the current algorithm cannot be applied where snow cover is present, snow cover extent is also included for comparison. We will discuss the progress and problems concerning the application of passive microwave remote sensing to the study of frozen soils