[A-CG44] Terrestrial monitoring using geostationary satellites

[ACG44-P06] A new vegetation monitoring method using geostationary satellite data based on a kernel-driven BRDF model

*Yueru Wen¹, Kazuhito Ichii¹ (1.Center for Environmental Remote Sensing, Chiba University)

Keywords:Vegetation monitoring, Geostationary satellite data, Kernel-driven BRDF model, Mid-high latitude regions

Vegetation, which plays a crucial role in shaping and sustaining the environment, exhibits distinct reflection and scattering characteristics for incident light from various directions. These characteristics are primarily dictated by its spectral and structural properties. Traditional vegetation monitoring methods based on remote sensing data, such as the Normalized Difference Vegetation Index (NDVI), Enhanced Vegetation Index (EVI), and Structural Scattering Index (SSI), have been extensively applied in Low Earth Orbit (LEO) satellites. However, significant differences exist between LEO and Geostationary Orbit (GEO) satellites, encompassing aspects such as sensor designs, atmospheric influence, observation modes, and angles. For GEO, higher satellite and solar zenith angles increase atmospheric effects, which will reduce sensitivity of traditional method, especially in mid-high latitude regions. In addition, even when atmospheric correction and Bidirectional Reflectance Distribution Function (BRDF) correction are applied, it is impossible to completely eliminate the impact of the atmosphere when implementing these corrections on geostationary satellites. In this study, a new method for monitoring vegetation changes using geostationary satellites data was proposed. Multiple linear regression was conducted to compute the BRDF kernel parameters pixel by pixel using the reflectance after atmospheric correction during the day and the BRDF kernels model. Then, a new vegetation index was computed by BRDF kernel parameters. The result was evaluated using the MODIS/Terra vegetation indices product MOD13A1. Notably, in mid-high latitude regions, this method demonstrates superior performance compared to other vegetation index. It has been shown that this new approach offers valuable insights for discerning vegetation canopy structures and monitoring seasonal changes over extended time series.