Assessing daytime vegetation water content estimates derived through the land parameter retrieval model from AMSR2 X-band observations

Vegetation annotates the radiation emitted by the Earth’s surface. The degree of annotation can be quantified by space-born passive microwave radiometers and is commonly known as Vegetation Optical Depth (VOD). VOD is directly connectedto and influenced by several factors, such as the water content, vegetation density, the wavelength used for observation andthe land cover type. One widley used algoritm is the Land Parameter Retrieval Model (LPRM). LPRM is a model that re-trieves both soil moisture and VOD simultaneously from V and H polarized microwave observations. Nighttime LPRM VODhas been extensively validated and used in many applications (e.g. Moesinger et al. [1], Zotta et al. [2]). LPRM assumesnightime equilibrium of canopy and soil temperature [3]. This does hold for daytime observations which we aim at doinghere.Here, we introduce an approach that separates the canopy and soil temperature using reanalysis data. We use it toretrieve AMSR2 VOD at X-band. First, existing VOD retrievals, retrieved for daytime and nighttime observations under thethermal equilibrium assumption, are compared to each other and to independent vegetation parameters such as the LeafArea Index (LAI) and the Fraction of Absorbed Photosynthetic Active Radiation (fAPAR). We also took Land cover classes(CCI Land Cover) into account to see in which biomes daytime VOD and nighttime VOD already agree with each other andanalysed why. In a second step, we plug in the soil and vegetation temperature from reanalysis separately into the LPRM tosee how it affects daytime VOD. We evaluate where and by how much it improves, especially in biomes where nighttime anddaytime retrievals are assumed to differ significantly. Furthermore, we will transfer the approach to Ku-band observations.First results indicate that our approach works best in dense vegetation (e.g. 60-Tree cover, broadleaved, deciduous,closed to open (>15%)), except for tropical rainforest. This class shows the largest discrepancy between daytime andnighttime retrievals due to an underestimation of daytime VOD caused by strong transpiration and large day-night temper-ature contrast.References[1] L. Moesinger, W. Dorigo, R. de Jeu, et al. The global long-term microwave Vegetation Optical Depth Climate Archive(VODCA). Earth System Science Data 12, 177–196 (2020).[2] R.-M. Zotta, L. Moesinger, R. van der Schalie, et al. VODCA v2: multi-sensor, multi-frequency vegetation optical depthdata for long-term canopy dynamics and biomass monitoring. Earth System Science Data 16, 4573–4617 (2024).[3] Manfred Owe, Richard de Jeu, and Thomas Holmes. Multisensor historical climatology of satellite-derived globalland surface moisture. Journal of Geophysical Research: Earth Surface 113 (2008). eprint: https : / / agupubs .onlinelibrary.wiley.com/doi/pdf/10.1029/2007JF000769.