Search

The joint National Aeronautics and Space Administration and Japanese Aerospace Exploration Agency (JAXA) Global Precipitation Measurement (GPM) mission will provide considerably more observations over complex and dynamically changing land backgrounds. A physically based precipitation retrieval using GPM's satellite constellation of passive microwave (PMW) observations has to accommodate the spatially and temporally varying radiometric signature of the land surface to constrain the set of candidate rainfall solutions. The challenge for retrieval algorithms is to identify and isolate precipitation profiles whose simulated observations agree with the satellite observations and are also representative of the surface conditions. Microwave emissivity modeling results are presented from a physically based land algorithm that retrieves soil moisture, vegetation water content, and surface temperature, along with the emissivity using polarized 10, 18, and 37 GHz channel measurements from the WindSat sensor onboard the Coriolis satellite, and results from the Tropical Rainfall Measuring Mission (TRMM) Microwave Imager (TMI). The emissivity mean, coefficient of variation, covariance, and correlation slope are examined for the range of clear-scene surface properties observed by WindSat and TRMM between 2003-2012 and 2002-2011, respectively, under a range of seasons, time of day, rain events, etc. These joint data provide a means to examine the extent to which the surface geophysical properties control the microwave land surface emissivity covariability, to better utilize these lower frequency observations in overland PMW-based precipitation retrievals. [ABSTRACT FROM PUBLISHER]

Knowledge about the surface soil water content is essential because it controls the surface water dynamics and land-atmosphere interaction. In high mountain areas in particular, soil surface water content controls infiltration and flood events. Although satellite-derived surface soil moisture data from passive microwave sensors are readily available for most regions globally, mountainous areas are often excluded from these data (or at least flagged as biased) due to the strong topographic influence on the retrieved signal. Even though a substantial volume of literature is available dealing with topographic effects on spaceborne brightness temperature, no systematic analysis has been reported. Therefore, we present a comprehensive analysis of topographic effects on brightness temperature at C-band using a two-step approach. First, a well-controlled field experiment is carried out using a mobile truck-mounted C-band radiometer to analyze the impact of geometric and adjacent effects on the radiometer signal. Additionally, a comprehensive radiative transfer model is developed accounting for both effects and tested on the ground-based data. Second, recorded Advanced Microwave Scanning Radiometer-Earth Observing System (AMSR-E) data over the Tibetan Plateau were used to analyze the error due to the impact of topography using the developed model. The results of the field experiment clearly show that the geometric effect of a single hill has a much larger impact on brightness temperature compared to the adjacent effect of multiple hills, whereby, due to the geometric effect, the bias is up to +20 K for horizontal and -13 K for vertical polarization. For the adjacent effect, the bias is less than 3 K for both polarizations. Additionally, the developed radio transfer model was able to reproduce both effects with high accuracy. For the AMSR-E data, the model shows that the brightness temperature recorded is biased in the same way as the ground-based measurements and that uncertainties induced by the wide existence of atypical mountain regions in the Tibetan Plateau will have a great impact on the retrieving error (maximum 30%). The largest impact on the retrieval error, on the other hand, is calculated for the soil moisture with a maximum relative error of 44%. The negligible impact can be attributed to false parameterization of the soil texture, soil surface temperature, and sky temperature. Finally, the overall absolute error in the estimated water content is quantified on average with 4%, whereby single pixels indicate a maximum absolute error of up to 16%. In conclusion, we show that recorded spaceborne brightness temperatures are highly biased by topographic effects in mountainous regions using a comprehensive radiative transfer model. Additionally, we suggest using this model to invert the effective surface emissivity of mountain areas for standard processing of higher level data products such as surface soil water content. [ABSTRACT FROM AUTHOR]