Modeling of Multi-Frequency Microwave Backscatter and Emission of Land Surface by a Community Land Active Passive Microwave Radiative Transfer Modeling Platform

Utilizing combined active and passive microwave signals at different frequencies provides complementary information of soil and vegetation physical states for ecosystem monitoring. We present here a prototype of a unified multi-frequency backscatter and emission simulator called the Community Land Active Passive Microwave Radiative Transfer Modeling Platform (CLAP). By assessing CLAP’s performance using in situ and satellite multi-frequency measurements of grassland backscatter and emission at the Maqu site, the impacts of different grass properties (i.e., structure, water, and temperature dynamics) and soil conditions (i.e., varied moisture and temperature profiles) are investigated. Results indicate that CLAP with cylinder parameterization for vegetation representation simulates grassland backscatter at X- and C-bands better than the disc parameterization does (e.g., root mean square errors [RMSEs] of 2.1 vs. 3.7 dB) during summer. Dynamic vegetation water content (VWC) partially explains diurnal variations of observed signals at low frequencies, while changes in vegetation temperature predominantly affect high-frequency signals. The study also reveals that grassland optical depth is frequency-dependent, with diurnal variation linked to VWC irrespective of frequency. CLAP using the cylinder parameterization and either the in situ measurements or the process model outputs simulates well the observed C-band backscatter at vertical transmit/vertical receive polarization (e.g., RMSE of 1.9 dB) during winter, while it cannot simulate signal dynamics at other bands during this period. Achieving consistent matches between CLAP modeled and observed signals across all frequencies, particularly for both passive and active signals, remains challenging. The limitations discussed in the study shed light on future directions to enhance CLAP’s modeling capability.