Derivation of volumetric liquid water content from radar and optical images for permafrost soil mixed with snow cover

The overarching goal of this research is to develop an integrated algorithm that enables remote sensing of LWC from radar backscattering coefficient data for permafrost landscape where the surface is often a mixed terrain of soil and snow cover. We propose to develop an integrated algorithm that derives volumetric liquid water content (LWC) in soil and snow separately and then integrate them to produce a LWC image from a combination of a radar image and optical images of permafrost land cover of soil and snow. The proposed studies involve the application of an algorithm tested for temperate soil to permafrost soil; followed by the development of a similar algorithm for derivation of LWC in snow. For an input of radar image (such RADARSAT ScanSAR, ALOS PALSAR), pixels will be classified first into soil and snow based on coregistered MODIS daily snow cover images, resulting in two images: soil image and snow image. Sub-algorithms for derivation of LWC in soil and snow are applied separately to the soil image and snow image that are then merged to produce an image of volumetric LWC for permafrost. Derived soil moisture will then be compared with ground measurement as a measure for the validation of the algorithm. Our research will make possible mapping LWC in permafrost region, an important parameter in permafrost plant ecosystem, permafrost thaw and freeze morning.

Contact Info