研究目的
To estimate soil moisture from polarimetric radar data using a linear regression model derived from Aquarius data, minimizing the need for ancillary data.
研究成果
The linear regression model provides reasonable global soil moisture estimates with minimal ancillary data. The pixel-based soil moisture captures small-scale patterns but shows errors in high Northern latitudes, which can be improved by constraining coefficients with spatially neighboring pixels.
研究不足
The approach assumes that radar backscatter over vegetated scenes changes mainly due to variations in soil moisture, which may not account for all variables affecting backscatter. The method also shows erroneous estimates in high Northern latitudes.
1：Experimental Design and Method Selection:
A linear model was used to describe the relationship between radar backscatter and soil moisture, eliminating the need for complex electromagnetic models.
2：Sample Selection and Data Sources:
Nearly 4 years of L-band Aquarius radar and radiometer derived soil moisture data were used to derive global coefficients.
3：List of Experimental Equipment and Materials:
Aquarius L-band radar and radiometer data, SMAP radar data.
4：Experimental Procedures and Operational Workflow:
Weekly polarization dependent coefficients were determined from Aquarius data and applied to SMAP radar data to estimate soil moisture.
5：Data Analysis Methods:
The soil moisture estimates were compared with the SMAP Level 2 radiometer-only soil moisture product for evaluation.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
