"Land-climate interactions mediated through soil moisture and vegetation play a critical role in the climate system, in particular for the occurrence of extreme events such as droughts and heatwaves. They are, however, poorly constrained in current Earth System Models (ESMs), leading to large uncertainties in climate projections. These uncertainties affect the quality and accuracy of projections of temperature, water availability, and carbon concentrations, as well as that of projected impacts on agriculture, ecosystems, and health.
In the past years, in-situ and remote sensing-based datasets of soil moisture, evapotranspiration, and energy and carbon fluxes have become increasingly available, providing untapped potential for reducing associated uncertainties in current climate models. The DROUGHT-HEAT project aims at innovatively exploiting these new information sources in order to 1) derive observations-based diagnostics to quantify and isolate the role of land-climate interactions in past extreme events (""Diagnostic Atlas""), 2) evaluate and improve current ESMs and constrain climate-change projections using the derived diagnostics, and 3) apply the newly gained knowledge to frontier developments in the attribution of climate extremes to land processes and their mitigation through ""land geoengineering"".
The DROUGHT-HEAT project integrates the newest land observational datasets with the latest stream of ESMs. Novel methodologies will be applied to extract functional relationships from the data, and identify key gaps in the ESMs' representation of underlying processes. These will build on physically-based relationships, machine learning tools, and model calibration. In addition, they will encompass the mapping and merging of derived diagnostics in space and time to reduce ""blank spaces"" in the datasets. The project is unprecedented in its breadth and scope and will allow a major breakthrough in our understanding of the processes leading to heatwaves and droughts."
Field of science
- /natural sciences/computer and information sciences/artificial intelligence/machine learning
Call for proposal
See other projects for this call