Observations and Modeling of a Pine Forest Floor at L-Band

Recent studies of passive L-band observations over forests have shown that the average canopy transmissivity in temperate coniferous and deciduous forests is on the order of 0.4-0.5. Although the canopy would therefore be expected to transmit a reasonable amount of ground emission, the total emission observed above the canopy shows very little variation with varying soil moisture content. Moist litter present on the forest floor is known to obscure the soil emission. Therefore, more knowledge on the L-band radiative properties of litter and understory layers is needed to better understand the emission of the whole forest system. In order to contribute toward this issue, a field experiment was conducted in a pine forest in southwest France. Radiometric observations were done on the canopy and on different configurations of the forest floor, following sequential stripping of each forest floor layer. In combination with a long-term data set of above-canopy observations, this resulted in emissivity values of bare soil, soil-litter, soil-litter-grass, and soil-litter-grass-canopy configurations for a range of soil and litter moisture values. Calculations involved the use of the Wilheit and L-MEB models. The sensitivity to soil moisture was found to be substantially suppressed by the presence of a grass understory and litter. This corroborates the low correlation between soil moisture and L-band brightness temperature measured above the canopy. Several results of recent modeling and laboratory studies are also confirmed by this paper, which is, to our knowledge, the first to use <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">in</i> <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">situ</i> experimental data in this context.