Description: The Environmental Mapping and Analysis Program (EnMAP) is a spaceborne German hyperspectral satellite mission that aims at monitoring and characterizing the Earth’s environment on a global scale. After several years delay due to a major design issue to meet the mission requirements, the mission is now back on track and planned for launch in 2021. This paper presents an update of the mission status with recent activities and developments from the space and the ground segment. Furthermore, a draft plan for the independent validation of EnMAP radiance and reflectance products was developed and will be introduced, along with highlights of the science preparatory activities in 2019 including airborne campaigns, algorithm consolidations, and HYPERedu education initiative.

Description: Forecasting crop yields is becoming increasingly important under the current context in which food security needs to be ensured despite the challenges brought by climate change, an expanding world population accompanied by rising incomes, increasing soil erosion, and decreasing water resources. Temperature, radiation, water availability and other environmental conditions influence crop growth, development, and final grain yield in a complex non-linear manner. Machine learning (ML) techniques, and deep learning (DL) methods in particular, can account for such non-linear relations between yield and its covariates. However, they typically lack transparency and interpretability, since the way the predictions are derived is not directly evident. Yet, in the context of yield forecasting, understanding which are the underlying factors behind both a predicted loss or gain is of great use. Here, we explore how to benefit from the increased predictive performance of DL methods while maintaining the ability to interpret how the models achieve their results. To do so, we applied a deep neural network to multivariate time series of vegetation and meteorological data to estimate the wheat yield in the Indian Wheat Belt. Then, we visualized and analyzed the features and yield drivers learned by the model with the use of regression activation maps. The DL model outperformed other tested models (ridge regression and random forests) and facilitated the interpretation of variables and processes that lead to yield variability. The learned features were mostly related to the length of the growing season, temperature, and light conditions during the growing season. For example, our results showed that high yields in 2012 were associated with low temperatures accompanied by sunny conditions during the growing period. The proposed methodology can be used for other crops and regions in order to facilitate application of DL models in agriculture.

Description: Measurements of reflected solar radiation by imaging spectrometers can quantify water in different states (solid, liquid, gas) thanks to the discriminative absorption shapes. We developed a retrieval method to quantify the amount of water in each of the three states from spaceborne imaging spectroscopy data, such as those from the German EnMAP mission. The retrieval couples atmospheric radiative transfer simulations from the MODTRAN5 radiative transfer code to a surface reflectance model based on the Beer-Lambert law. The model is inverted on a per-pixel basis using a maximum likelihood estimation formalism. Based on a unique coupling of the canopy reflectance model HySimCaR and the EnMAP end-to-end simulation tool EeteS, we performed a sensitivity analysis by comparing the retrieved values with the simulation input leading to an R2 of 0.991 for water vapor and 0.965 for liquid water. Furthermore, we applied the algorithm to airborne AVIRIS-C data to demonstrate the ability to map snow/ice extent as well as to a CHRIS-PROBA dataset for which concurrent field measurements of canopy water content were available. The comparison between the retrievals and the ground measurements showed an overall R2 of 0.80 for multiple crop types and a remarkable clustering in the regression analysis indicating a dependency of the retrieved water content from the physical structure of the vegetation. In addition, the algorithm is able to produce smoother and more physically-plausible water vapor maps than the ones from the band ratio approaches used for multispectral data, since biases due to background reflectance are reduced. The demonstrated potential of imaging spectroscopy to provide accurate quantitative measures of water from space will be further exploited using upcoming spaceborne imaging spectroscopy missions like PRISMA or EnMAP.