Using machine learning to better understand climate change

Image
pr_zaspel.jpg
Professor Peter Zaspel developed a machine learning method in order to be able to reconstruct past climatic conditions. (Source: Jacobs University) ,

 

July 6, 2021
 
The more reliable data is available on past climate periods, the more accurate climate change models can be. Scientists at Jacobs University Bremen and the Alfred Wegener Institute (AWI) aim to use new machine learning methods to uncover climate information hidden in sediment cores. To this end, a doctoral position was created at the Helmholtz School for Marine Data Science.

Sediment cores are climate archives. Much like tree growth rings, they tell of changing environmental conditions, such as temperature fluctuations, precipitation, or water acidification. "However, they often reveal information about climate only indirectly. In addition, the data often does not add up because it was collected using different methods," described Peter Zaspel when explaining the problem. The professor of computer science is responsible for the research project at Jacobs University. He cooperates with physicist Professor Thomas Laepple from the AWI. Also involved in the project are the biogeochemistry professor, Jelle Bijma from AWI and Jacobs University, and Professor Vikram Unnithan, geoscientist at Jacobs University.

In order to be able to reconstruct past climate conditions, a machine learning method developed by Professor Zaspel is being used in the project. It is a multi-fidelity method that has already been successfully tested in another field. "It generates the best possible information from data that is only comparable to a limited extent," said Professor Zaspel. This new reconstruction technique aims to provide better insights into past climate trends.


Questions can be addressed to:
Peter Zaspel
Professor of Computer Science
Phone: +49 421 200-3051
Email: p.zaspel [at] jacobs-university.de

 

About Jacobs University Bremen:
Studying in an international community. Obtaining a qualification to work on responsible tasks in a digitized and globalized society. Learning, researching and teaching across academic disciplines and countries. Strengthening people and markets with innovative solutions and advanced training programs. This is what Jacobs University Bremen stands for. Established as a private, English-medium campus university in Germany in 2001, it is continuously achieving top results in national and international university rankings. Its more than 1,500 students come from more than 110 countries with around 80% having relocated to Germany for their studies. Jacobs University’s research projects are funded by the German Research Foundation or the EU Research and Innovation program as well as by globally leading companies.

Contact:
Melisa Berktas| Corporate Communications & Public Relations
m.berktas [at] jacobs-university.de | Tel.: +49 421 200-4135
 

About Helmholtz School for Marine Data Science:
The “Helmholtz School for Marine Data Science” (MarDATA) is a graduate school, financed by the Helmholtz Association. It aims to define and educate a new type of “marine data scientists” by introducing and embedding researchers from computer sciences and mathematics into ocean sciences, covering a broad range from supercomputing and modeling, (bio)informatics, robotics, to statistics and big data methodologies. Researchers from the German leading institutes for marine research, the GEOMAR Helmholtz Centre for Ocean Research Kiel and the Alfred-Wegener-Institute (AWI) Helmholtz Centre for Polar and Marine Research, will jointly educate and supervise doctoral (PhD) candidates together with information & data science specialists from their partner universities in Kiel (Kiel University) and Bremen (University of Bremen and Jacobs University). The broad education in joint block courses, international summer schools, and colloquia goes beyond a single discipline towards genuine scientific insight into and a more systematic treatment of marine data.

For more information: http://www.mardata.de/| Twitter

Contact:
Dr. Enno Prigge | Scientific Coordinator MarDATA
eprigge [at] geomar.de