Modeling and Verification of Microwave Scattering Characteristics of Typical Global Tropical Rainforests

Abstract

To ensure the microwave scatterometer is accurately calibrated, natural targets with stability, homogeneity, and isotropy are selected as references. The broad and continuous spatial distribution of the tropical rainforest, along with its relatively consistent vegetation cover, makes it an ideal choice.A tropical rainforest optimal stable area selection algorithm, combining mean, standard deviation, and relative standard deviation, is proposed using measurements from the advanced scatterometer (ASCAT) onboard the second Meteorological Operational satellite (MetOp-B) from 2019 to 2021. It is used to identify stable areas within the Amazon, Congo, and Southeast Asian rainforests. Results show that the Amazon rainforest has a larger stable area compared to the Congo and Southeast Asian rainforests, indicating more consistent backscatter across space. However, the Southeast Asian rainforest exhibits scattered stable areas and unstable backscatter properties.To accurately model the intrinsic characteristics of targets within stable areas, influences of seasonal variations, incidence angles and azimuth angles are comprehensively considered. The scatterometer, as an independently measured remote sensing instrument, is not affected by seasonal variations on the earth and experiences minimal temperature-related fluctuations. Therefore, seasonal characteristics of backscatter coefficients in rainforests can be modeled to reduce their impact. Different incidence and azimuth angles can cause variations in the backscatter coefficient. To address this issue, responses to these aspects are also modeled. It is observed that daytime data, with lower model errors, shows greater stability in the stable areas of the Amazon and Congo rainforests. Therefore, daytime data from these areas should be selected to assess instrument stability.A stability verification of ASCAT measurements from the stable areas of the Amazon and Congo rainforest on MetOp-C, covering the period from 1 July 2019 to 31 October 2023, is carried out based on model coefficients derived from the continuous three-year data of ASCAT on MetOp-B. The calibration stability verification quantifies the magnitude of variations in ASCAT measurements over different periods. Through analysis, it's found that measurements from the ASCAT on MetOp-C shows regular fluctuations of about 0.05 dB, indicating relatively stable characteristics.

Keywords

• scatterometer
• tropical rainforest
• optimal stable area selection
• feature modeling
• verification