UT researchers Dr. Yijian Zeng, Prof. Dr. Bob Su, Dr. Christiaan van der Tol, Prof. Dr. Raul Zurita-Milla and Dr. Michael Ying Yang (all from ITC Faculty) have been awarded a grant from Netherlands eScience Center’s ASDI 2020 call. In their research project, they will combine two computer models to understand how water-carbon dynamics of ecosystems vary with climate extremes, such as droughts and heatwaves.
Droughts and heatwaves impact how water, energy and carbon move through ecosystems. Climate extremes like these jeopardize the ability of ecosystems to remove carbon dioxide from the atmosphere to mitigate or even reverse climate change. New models can trace these exchanges. SCOPE – one of these models – simulates the photosynthetic activity of vegetation and links it to satellite data. Another model is STEMMUS, which traces the soil moisture availability for plants. “Models like SCOPE and STEMMUS work great, but are limited to one specific aspect of the problems we are trying to solve”, explains Zeng, the principal investigator. In this new project called EcoExtreML, the researchers will combine both models to release their full potential.
This project will open up new opportunities for Earth-Observation for retrieving higher-level products like the moisture levels in deeper soil (root-zone-soil-moisture) and the amount of carbon that is stored underground (carbon sequestration). However, combining STEMMUS and SCOPE globally is no easy task. Both models alone, already require high computational power. Together it becomes almost impossible to compute. This is where the ASDI call comes in. The call is meant for projects that require advanced eScience technologies like Artificial Intelligence. “We will combine process modelling and machine learning to be able to overcome the huge amount of computational load by both models when applied globally”, says Zeng, “The Netherlands eScience centre and the Center of Expertise in Big Geodata Science (CRIB) at ITC Faculty will empower this project with the state-of-the-art eSciences.”
The combined model, STEMMUS-SCOPE, will be able to predict what vegetation will do in response to an extreme climate event. With this information, it becomes possible to evaluate how to increase carbon sink in an area. “When vegetation is healthy, it will uptake CO2 from the atmosphere. With our model, we will be able to tell quantitatively what is happening in the soil-plant-atmosphere continuum, in terms of water, carbon and energy exchanges through vegetation”, says Zeng. Furthermore, using the data assimilation algorithm, STEMMUS-SCOPE will be able to overcome gaps in satellite data, caused by, for example, cloudy conditions or a long satellite revisit-time of up to 27 days. “This project will provide continuous data in space and time for assessing ecosystem productivity at local, regional, and global scales, and serving society’s growing need for crop production, food security, and other ecosystem services for a sustainable quality of life.”
The full name of the project is ‘Accelerating Process Understanding for Ecosystem Functioning under Extreme Climates with Physics-Aware Machine Learning (EcoExtreML)’. As one of the eight winning projects from the ASDI 2020 call from the Netherlands eScience Center, EcoExtreML receives a grant for a 3-year postdoc position and 2.5-year in-kind support of a research software engineer from the Netherlands eScience Center. Dr. Yijian Zeng – assistant professor at the Department of Water Resources (WRS; ITC Faculty) – is the principal investigator. A vacancy for a researcher on this project has recently been opened.