Unprecedented detail: global drought scale

News Vice – Some parts of the world are suffering from extreme heat and persistent drought, while others are being flooded. Overall, continental water levels vary greatly over time, and global sea levels also change significantly. By combining hydrological model Water Gap and GRACE satellite data, a team of geologists from the University of Bonn has come up with a new data set that shows how the distribution of total water over Earth’s landmasses has changed over the past 20 years. More accurate than ever. Their findings are now published in the „Journal of Geodesy”.

„The new method allows us to test model calculations on the future effects of climate change, particularly how rising temperatures and changes in precipitation patterns will affect the water balance in different regions of the world,” says Professor Dr-Ing. Jürgen Gusse from the Institute of Geodesy and Geoinformation at the University of Bonn. The process involves comparing climate models, which compare a period of time in the past, with the results of actual measurements, and Guse and his team are planning more such studies in the coming months.

The improved resolution achieved by the team shows that droughts are more common around the world than the GRACE satellite data would suggest individually. „We see that even extensive droughts, such as the massive drought that hit the entire Amazon in 2010, spread over much wider areas than the satellite data would automatically indicate,” Gusse says. „This means the satellites are not picking up many localized droughts.”

A team of researchers from the University of Bonn, working with colleagues from Goethe University Frankfurt and the Polish capital Warsaw, has now combined satellite measurements with high-resolution weather data for the first time. „What’s special about this method is that it enables us to improve the resolution of water distribution maps generated from about 300 kilometers to 50 kilometers,” explains Kusey, a member of the Modeling and Sustainable Futures transdisciplinary research areas. Regional Climate Change Collaborative Research Center at the University of Bonn. To do so, the researchers used „WaterGAP”.
A mathematical technique borrowed from hydrological modeling and weather forecasting developed at Goethe University Frankfurt.

A mass of water that causes changes in the gravitational field

Between 2002 and 2017, the GRACE (Gravity Recovery and Climate Experiment) twin satellites measured changes in Earth’s gravity. Its successor project “GRACE-FO” was launched in 2018, and it was this data that the University of Bonn researchers used. Because Earth’s gravity depends on mass changes, this allows us to draw conclusions about the water cycle near its surface. Changes in groundwater and surface aquifers and the melting of glaciers are affected by gravity.

„A unique advantage of GRACE gauges is that they cover all types of reservoirs, including underground water reserves hidden deep below the Earth’s surface and tens of thousands of artificial lakes and wetlands,” says Gusei’s colleague Helena Gartener. The drawback, he says, is that the spatial resolution of the gravity data is relatively imprecise to about 300 to 350 kilometers as a result of the measurement principle. This means that reliable reports can only be made for an area of ​​100,000 square kilometers. To give some idea of ​​scale, this minimal area is „only” 70,000 or so square kilometers larger than Bavaria, Germany’s largest federal state.

In contrast, global hydrological models allow a resolution of 50 kilometers or less. These use meteorological measurements of precipitation, temperature, and radiation, as well as maps of land use and soil composition, and data on how industry, agriculture, and other consumers use water. Hydrologic models simulate evaporation and changes in water levels in soil and groundwater aquifers, lakes, rivers, and reservoirs. „However, the disadvantages of these models are that they can only reflect reality to a certain extent and meteorological measurements often have systematic errors,” Gusse says, if for example no data on groundwater extraction is available.

For the first time, researchers have now combined measurements from the GRACE and GRACE-FO satellites with the WaterGAP hydrological model, which integrates high-resolution meteorological data. This helps in improving the resolution of the water distribution maps thus generated
50 kilometers. To do so, the researchers used a mathematical technique called data assimilation, which is commonly found in weather forecasting. However, scientists do not simply take the results of hydrological modeling and satellite data and calculate average values.
As Gusse explains: „The calculations of the hydrological model are adjusted so that you get closer to the satellite data, while the hydrological model modifies the physics.”

1,000 measuring stations for testing purposes

The researchers used about 1,000 gauging stations to test the quality of continental water distribution maps produced by combining satellite data with a hydrological model. „Of course, you will always see some regional differences,” admits Helena Gertner. However, across the board, he says, the integrated data fit measurements better than calculations based on GRACE satellite data or hydrological modeling.

Organizations involved and funding secured:

Besides the University of Bonn, the Institute of Physical Geography at Goethe University Frankfurt, the Sengenberg Leibniz Biodiversity and Climate Research Center Frankfurt and the Faculty of Civil Engineering and Geography at the Military Technical University in Warsaw, Poland, were also involved in the study. It was financially supported by the Federal Ministry of Education and Research (BMBF) and the German Research Foundation (DFG).

Publication: Helena Gartener, Jürgen Kusse, Kerstin Schulz, Petra Toll, Anna Klose: Global Land Water Storage Dataset Release 2 (GLWS2.0) Integrating GRACE and GRACE-FO data into a global hydrological model, “Journal of Geodesy, DOI: 10.1007 /s00190-023-01763-9; Present:
https://link.springer.com/article/10.1007/s00190-023-01763-9