Artificial Intelligence Learns Continental Hydrology

The complex distribution of continental water masses in South America has been determined with a new Deep-Learning-Method using satellite data.

The complex distribution of continental water masses in South America has been determined with a new Deep-Learning-Method using satellite data.

Changes to water masses which are stored on the continents can be detected with the help of satellites. The data sets on the Earth’s gravitational field which are required for this, stem from the GRACE and GRACE-FO satellite missions. As these data sets only include the typical large-scale mass anomalies, no conclusions about small scale structures, such as the actual distribution of water masses in rivers and river branches, are possible. Using the South American continent as an example, the Earth system modellers at the German Research Centre for Geosciences GFZ, have developed a new Deep-Learning-Method, which quantifies small as well as large-scale changes to the water storage with the help of satellite data. This new method cleverly combines Deep-Learning, hydrological models and Earth observations from gravimetry and altimetry.

So far it is not precisely known, how much water a continent really stores. The continental water masses are also constantly changing, thus affecting the Earth's rotation and acting as a link in the water cycle between atmosphere and ocean. Amazon tributaries in Peru, for example, carry huge amounts of water in some years, but only a fraction of it in others. In addition to the water masses of rivers and other bodies of fresh water, considerable amounts of water are also found in soil, snow and underground reservoirs, which are difficult to quantify directly.

Now the research team around primary author Christopher Irrgang developed a new method in order to draw conclusions on the stored water quantities of the South American continent from the coarsely-resolved satellite data. “For the so called down-scaling, we are using a convolutional neural network, in short CNN, in connection with a newly developed training method”, Irrgang says. “CNNs are particularly well suited for processing spatial Earth observations, because they can reliably extract recurrent patterns such as lines, edges or more complex shapes and characteristics.

In order to learn the connection between continental water storage and the respective satellite observations, the CNN was trained with simulation data of a numerical hydrological model over the period from 2003 until 2018. Additionally, data from the satellite altimetry in the Amazon region was used for validation. What is extraordinary, is that this CNN continuously self-corrects and self-validates in order to make the most accurate statements possible about the distribution of the water storage. “This CNN therefore combines the advantages of numerical modelling with high-precision Earth observation” according to Irrgang.

The researchers’ study shows that the new Deep-Learning-Method is particularly reliable for the tropical regions north of the -20° latitude on the South American continent, where rain forests, vast surface waters and also large groundwater basins are located. Same as for the groundwater-rich, western part of South America’s southern tip. The down-scaling works less well in dry and desert regions. This can be explained by the comparably low variability of the already low water storage there, which therefore only have a marginal effect on the training of the neural network. However, for the Amazon region, the researchers were able to show that the forecast of the validated CNN was more accurate than the numerical model used.

In future, large-scale as well as regional analysis and forecasts of the global continental water storage will be urgently needed. Further development of numerical models and the combination with innovative Deep-Learning-Methods will take up a more important role in this, in order to gain comprehensive insight into continental hydrology. Aside from purely geophysical investigations, there are many other possible applications, such as studying the impact of climate change on continental hydrology, the identification of stress factors for ecosystems such as droughts or floods, and the development of water management strategies for agricultural and urban regions.

Funding:

This study was funded by the Helmholtz-Association as well as The Initiative and Networking Fund of the Helmholtz-Association through the Advanced Earth Modelling Capacity project (ESM).

Original study: Irrgang, C., Saynisch-Wagner, J., Dill, R., Boergens, E., & Thomas, M. (2020). Self-validating deep learning for recovering terrestrial water storage from gravity and altimetry measurements. Geophysical Research Letters, 47, e2020GL089258.

doi.org/10.1029/2020GL089258

 

Scientific contact:

Dr. Christopher Irrgang
Scientist in Section Earth System Modelling
Helmholtz Centre Potsdam
GFZ German Research Centre for Geosciences
Telegrafenberg
14473 Potsdam
Phone: +49-331-288-2847
e-mail: irrgang@gfz-potsdam.de

Media contact:

Josef Zens
Head of Public and Media Relations
Helmholtz Centre Potsdam
GFZ German Research Centre for Geosciences
Telegrafenberg
14473 Potsdam
Phone: +49 331 288-1040
e-mail: Josef.Zens@gfz-potsdam.de

Additional News

Ehrung von Prof. Onno Oncken mit einem wissenschaftlichen Kolloquium

DEUQUA Logo mit Mammut und Friedenstaube

DEUQUA 2022 Tagung am GFZ

PAW Logo

Postdoc Appreciation Week Germany

Building, photo taken in winter, Isaac Newton Institute

Simons Scholarship for Dr Monika Korte

Die Verteilung der seismischen Stationen auf einer Karte der Region.

How deeply does Eifel volcanism sleep?

Geomagnetic Field. Space with stars, Earth with animation around

GFZ film among the finalists of the Earth Futures Festival 2022

Earth's radiation belt: High-energy particles modelled around the Earth. The particles are ring-shaped

A new population of particles in the Earth’s radiation belts

[Translate to English:] Die teilnehmenden GFZ Mitarbeiter als Gruppenfoto

2nd proWissen run in Potsdam with successful participation by GFZ employees

The group on the first day of work.

New faces at the GFZ - start of the training year 2022/2023

Forest vs no forest on two sides of a road

Agriculture drives more than 90% of tropical deforestation

Group photo: people on the roof terrace of a house

2nd International Symposium of International Association of Geodesy’s Commission 4…

[Translate to English:] Foto eines Bergs mit darüber gelegter Skizze des geologischen Profils.

How thick should clay be as a host rock for a repository?

White dots of different thicknesses in a hexagonal pattern on a black ground.

Synthesis of hexagonal SiGe semiconductor using high pressure and temperature

Landslide on a slope directly adjacent to a settlement with small houses.

Landslides increasingly threaten the world's urban poor cities

Different coloured liquids mix in an aquarium. A child watches.

Catching up after Corona: "GEOtogether" brings joy for pupils in collaborative…

A woman and a man stand on a stage holding a picture with a coloured map of Türkiye..

Four decades of joint Turkish-German earthquake research

Egon Althaus sitting around a table with colleagues outside on a project

We mourn the death of Egon Althaus (1933-2022)

A dry dam near Capetown, South Africa.

The challenge of unprecedented floods and droughts in risk management

Drawing of a fictional historical submersible.

Eleven short research stays with GFZ participation funded

Drilling platform on Lake Junin with several people on it

Tropical glaciers followed the rhythm of the ice sheet expansion in the northern…

Schematic representation of the VECTOR project: A large arrow with different levels - from the earth's surface to underground.

Improving the exploration efficiency in Europe

Hoby Razafindrakoto

Project from Dr. Razafindrakoto to create a seismological lab in Madagascar wins ARISE…

3D digital Earth at night

Open-Earth-Monitor getting started

The dam of the Steinbach Dam in the Eifel region, cut by flooding but not destroyed.

Flood risk management after the Eifel flood in July 2021

Group picture of the Cermak7 Conference in front of the Museum Barberini in Potsdam

International heat flow conference and workshop in Potsdam

Castor platform in the ocean. The sea is still.

Filling geological gas reservoirs: Causal research in the most important event of induced…

People sit on chairs in a circle in a room.

GFZ PhD Days

Jeffrey Perez in front of the Logo of the meeting

Two GFZ Researchers participate in the 71st Lindau Nobel Laureate Meeting

Young woman standing with certificate in hand in a hall in front of the lettering EAGE

Best Paper Award for Evgeniia Martuganova

Four persons are holding a large golden key, standing in front of a house

Housing for visiting scientists

back to top of main content