Hy-PiPE

Runtime: 11/2022 - 10/2025
Team: Dr. Daniel Doktor (PI), Dr. Mohammad Hajeb

Earth observation systems provide standardised data that can comprehensively map dynamic vegetation processes. Multi-spectral sensors deliver high-resolution data (in space and time), making them suitable for detailed thematic land-cover characterisation. For capturing biophysical and biochemical parameters at the plant or even leaf level, the spectral resolution of a satellite sensor plays a critical role. The hyperspectral satellite of the Environmental Mapping and Analysis Program (EnMAP) offers new possibilities for monitoring spatio-temporal changes in leaf and canopy parameters, with a spatial resolution of 30 meters and more than 200 spectral channels.

The derivation of plant condition variables from remote sensing data typically relies on the inversion of radiative transfer models (e.g., PROSAIL) (Berger et al., 2018; Verrelst et al., 2021). Process-based agroecosystem models (AEM; e.g., Jones et al., 2003), which simulate the plant-soil-atmosphere system, can use these condition data derived from remote sensing to improve yield estimates. One goal of this project is to leverage the technical advantages of the EnMAP satellite for the derivation of plant condition variables (traits). Based on regional case studies, concepts will be developed, and processes established using radiative transfer modeling, which will eventually be applied on a national scale over the course of the project.

Furthermore, the compilation of various spatially and thematically high-resolution datasets is planned for pixel-specific parameterisation of radiative transfer and agroecosystem models. In addition to using detailed nationwide maps of the spatial distribution of agricultural crops, a key focus is on developing a spatial hyperspectral soil dataset incorporating EnMAP data. Based on this, the ultimate goal is to establish a routine for assimilating EnMAP-based canopy condition data into agroecosystem models to enable large-scale yield predictions with high spatial resolution.