Assessing the sensitivity of multi-frequency passive microwave vegetationoptical depth to vegetation properties

Vegetation attenuates the microwave emission from theland surface. The strength of this attenuation is quantified in models interms of the parameter vegetation optical depth (VOD) and is influenced bythe vegetation mass, structure, water content, and observation wavelength.Earth observation satellite sensors operating in the microwave frequenciesare used for global VOD retrievals, enabling the monitoring of vegetation atlarge scales. VOD has been used to determine above-ground biomass, monitorphenology, or estimate vegetation water status. VOD can be also used forconstraining land surface models or modelling wildfires at large scales.Several VOD products exist, differing by frequency/wavelength, sensor, andretrieval algorithm. Numerous studies present correlations or empiricalfunctions between different VOD datasets and vegetation variables such asthe normalized difference vegetation index, leaf area index, gross primaryproduction, biomass, vegetation height, or vegetation water content. However,an assessment of the joint impact of land cover, vegetation biomass, leafarea, and moisture status on the VOD signal is challenging and has not yetbeen done. This study aims to interpret the VOD signal as a multi-variate function ofseveral descriptive vegetation variables. The results will help to selectVOD at the most suitable wavelength for specific applications and can guidethe development of appropriate observation operators to integrate VOD withlarge-scale land surface models. Here we use VOD from the Land ParameterRetrieval Model (LPRM) in the Ku, X, and C bands from the harmonized Vegetation Optical Depth Climate Archive (VODCA)dataset and L-band VOD derived from Soil Moisture and Ocean Salinity (SMOS) and Soil Moisture Active Passive (SMAP) sensors. The leaf area index,live-fuel moisture content, above-ground biomass, and land cover are able toexplain up to 93 % and 95 % of the variance (Nash-Sutcliffe modelefficiency coefficient) in 8-daily and monthly VOD within a multi-variablerandom forest regression. Thereby, the regression reproduces spatialpatterns of L-band VOD and spatial and temporal patterns of Ku-, X-, andC-band VOD. Analyses of accumulated local effects demonstrate that Ku-, X-,and C-band VOD are mostly sensitive to the leaf area index, and L-band VOD is most sensitive toabove-ground biomass. However, for all VODs the global relationships withvegetation properties are non-monotonic and complex and differ with landcover type. This indicates that the use of simple global regressions toestimate single vegetation properties (e.g. above-ground biomass) from VODis over-simplistic.

Automated summary

Vegetation attenuates microwave emission from the land surface. This attenuation is quantified by vegetation optical depth (VOD) and can be used to estimate above-ground biomass, monitor phenology, and model wildfires at large scales. Different VOD products exist, but the relationships with vegetation variables are non-monotonic and complex, indicating the need for more sophisticated estimation methods.