Global Estimates of L-band Vegetation Optical Depth and Soil Permittivity over Snow-covered Boreal Forests and Permafrost using SMAP Satellite Data

This article expands the tau-omega model to properly simulate L-band microwave emission of the soil-snow-vegetation continuum through a closed-form solution of Maxwell's equations, considering the intervening dry snow layer as a loss-less medium. The feasibility and uncertainty of retrieving vegetation optical depth (VOD) and ground permittivity, given the noisy L-band brightness temperatures with 1 K (1-sigma), are demonstrated through controlled numerical experiments. For moderately dense vegetation canopy and a range of 100--400 $kg.m^{-3}$ snow density, the standard deviation of the retrieval errors is 0.1 and 3.5 for VOD and ground permittivity respectively. Using L-band observations from the Soil Moisture Active Passive (SMAP) satellite, a new data set of global estimates of VOD and ground permittivity are presented over the Arctic boreal forests and permafrost areas during winter months. In the absence of dense ground-based observations of ground permittivity and VOD, the retrievals are causally validated using dependent variables including above-ground biomass, tree height, and net ecosystem exchange. Time-series analyses promise that the new data set can expand our understanding of the land-atmosphere interactions and exchange of carbon fluxes over the Arctic landscape.