Sustainable Agriculture and Water Resources
Agriculture production systems are facing unprecedented challenges from increasing demand for food for a growing population, but an intensified agriculture bares also risks for environmental pollution and unsustainable use of water resources. In addition, climatic change and soil degradation affect food production in the long term. The ÔÇÿZero HungerÔÇÖ goal was highly stressed as one of the major Sustainable Development Goals by the United Nations, which obviously calls for sustainable agricultural practices and wise management.Therefore, monitoring and prediction of agricultural systems has been a strong motivation for scientists and remains challenging where gaps in our understanding and capability of monitoring systems are limited. Earth Observation (EO) is already used to estimate land surface variables, but the step to a full process understanding of agricultural systems has not yet been taken. Adequate mitigation strategies during droughts, nitrogen pollution events, and pests cannot be adequately employed without this understanding. Integrated agricultural system studies at local, regional and national levels are needed to achieve synergies and adequately address trade-offs among the food-water-energy nexus during land and climate change.Objective The aim of this project is to monitor essential variables in agriculture based on various in situ and remote observations to investigate agricultural processes and to carry out a full agro-ecosystem health diagnosis by data assimilation. This synoptical perspective allows us to conserve, protect, and improve the efficiency of the use of natural resources to facilitate sustainable agricultural development. A near-real-time prototype will enable informing stakeholders about timely management actions. With this proposal, we plan to prepare for agricultural applications of future missions, such as BIOMASS, EnMAP.Research contents and methodsDuring Dragon 5, we will develop a multi-variable EO strategy for agricultural areas. Essential variables such as crop type, LAI, biomass, soil moisture, evapotranspiration, and soil carbon content are monitored by remote sensing methods to characterize the current state of agro-ecosystems. Machine learning, process-based radiative transfer, hybrid inversion, and Bayesian scaling approaches will be applied to timely retrieve the variables in focus. To bring together the different variable types, multi-source ensemble Kalman filter methods will be applied to improve the performance in crop growth and hydrological simulation, where the links between these two aspects will be investigated especially for extreme weather conditions. This will pave the way for simulating near-future states for the implementation of a sound and sustainable soil, land, water, nutrient and pest management, and appropriate use of fertilizers.DeliverablesWe will provide essential agro-ecosystem variable products in the hydrological (soil moisture, evapotranspiration) and plant physical (crop type, LAI, biomass, chlorophyll, and nitrogen content, fire occurrence) domain. It will make use of European, Chinese and TPM EO data in different wavelengths, if it is in visual, infrared, thermal or microwave region of the electromagnetic spectrum. We deliver a strategy how an early warning system can be implemented for crop health diagnosis and investigate the effects of different environmental threads. The transferability of the approaches will be tested at European and Asian sites to assure global applicability. A vital deliverable is also the education of young scientists in the field of remote sensing of agriculture to multiply their EO expertise in their further career at different stakeholders.FundingThe Dragon 5 project will be supported by the National Natural Science Foundation of China (NSFC) (Grant No. 41971305, 41701371, 41807001, 41701236), and the European Commission Horizon 2020 Program ERA-PLANET/GEOEssential (Grant Agreement no. 689443).
