The Frontiers of Deep Learning for Earth System Modelling 

Author(s):  
David Hall

<p>This talk gives an overview of cutting-edge artificial intelligence applications and techniques for the earth-system sciences. We survey the most important recent contributions in areas including extreme weather, physics emulation, nowcasting, medium-range forecasting, uncertainty quantification, bias-correction, generative adversarial networks, data in-painting, network-HPC coupling, physics-informed neural nets, and geoengineering, amongst others. Then, we describe recent AI breakthroughs that have the potential to be of greatest benefit to the geosciences. We also discuss major open challenges in AI for science and their potential solutions. This talk is a living document, in that it is updated frequently, in order to accurately relect this rapidly changing field.</p>

2012 ◽  
Vol 5 (3) ◽  
pp. 1987-2006 ◽  
Author(s):  
S. Valcke ◽  
V. Balaji ◽  
A. Craig ◽  
C. DeLuca ◽  
R. Dunlap ◽  
...  

Abstract. This paper presents a review of the software currently used in climate modelling in general and in CMIP5 in particular to couple the numerical codes representing the different components of the Earth system. The coupling technologies presented show common features, such as the ability to communicate and regrid data, but also offer different functions and implementations. Design characteristics of the different approaches are discussed as well as future challenges arising from the increasing complexity of scientific problems and computing platforms.


2020 ◽  
Author(s):  
Luisa Cristini ◽  

<p>With climate change and the conjoint challenges of food availability, clean water and geo-energy resources, our society is facing major challenges in the near future. These challenges are hard to address, because projections of Earth system change involve uncertainties that require quantification. Therefore, the Earth system science community tries to develop tools that provide decision-makers with the information required to effectively manage these issues.</p><p>The Advanced Earth System Modelling Capacity project (ESM) aims to enable such tools, investigating problems by looking at interactions between different Earth system components and improve their representation in numerical models. The project was funded by the German Helmholtz Association in April 2017 and involves eight research centers across Germany. The ultimate goal of the project is to represent the Earth system and how it changes with a world-leading modelling infrastructure that will support the process of developing solutions for the grand challenges we are facing.</p><p>The five different work packages of the project are working on topics such as enhancing the representation of Earth system model compartments, develop a flexible framework for coupling of Earth system model components, advance the Earth system data assimilation capacity, diagnose Earth system models, further develop cutting-edge frontier simulations, cross-scale modelling, and contribute to the shaping of a national strategy for Earth system modelling. The project also engages in training activities to educate and transfer knowledge to the next generation of scientists.</p><p>Since its initiation the project contributed with important results to several key model systems and platforms. In this presentation, we will highlight some current results and discuss advances in our Earth system modelling community and the way forward.</p>


2012 ◽  
Vol 5 (6) ◽  
pp. 1589-1596 ◽  
Author(s):  
S. Valcke ◽  
V. Balaji ◽  
A. Craig ◽  
C. DeLuca ◽  
R. Dunlap ◽  
...  

Abstract. This paper presents a review of the software currently used in climate modelling in general and in CMIP5 in particular to couple the numerical codes representing the different components of the Earth System. The coupling technologies presented show common features, such as the ability to communicate and regrid data, and also offer different functions and implementations. Design characteristics of the different approaches are discussed as well as future challenges arising from the increasing complexity of scientific problems and computing platforms.


2021 ◽  
Vol 3 (8) ◽  
pp. 667-674
Author(s):  
Christopher Irrgang ◽  
Niklas Boers ◽  
Maike Sonnewald ◽  
Elizabeth A. Barnes ◽  
Christopher Kadow ◽  
...  

2020 ◽  
Author(s):  
Luisa Cristini ◽  
Robert Sausen ◽  
Mariano Mertens ◽  
Nadine Wieters ◽  
Sara Pasqualetto

<p>The Earth System Modelling Capacity (ESM) project is a 3-year effort funded by the Helmholtz Association started in April 2017 and involving eight research centres across Germany. The project has a strong knowledge transfer component aiming to provide decision-makers with relevant tools in order to face grand challenges in the near future and to support early career scientists (PhD students and early career postdoctoral researchers) from ESM project partner centres as well as the national and international community in developing and strengthening their knowledge on Earth system modelling, as one of the primary efforts to establish a legacy for the project.</p><p>From 9<sup>th</sup> to 19<sup>th</sup> of September 2019, the ESM summer school was held in Bad Aibling (Germany) with 50 participating students from 26 institutes placed all over the world. A core objective of the school was to train and educate early-career scientists from a wide range of discipline and with a diverse international and gender background to apply cutting edge science in the study of the Earth system and at the same time to engage in a stimulating exercise of knowledge transfer for the project.</p><p>During the 10-day summer school, students had four lectures daily about topics related to the Earth system and its components, from atmospheric dynamics to terrestrial modelling, from the modelling of waves and oceans to that of ice sheets and glaciers. The school included practical exercises and hands-on sessions that involved coding and building mini-cluster computers, building on the advanced technical knowledge of ESM partners and scientists. The lectures were held by thirty researchers from the ESM Project’s partner institutes and beyond. Two poster sessions were also organized, where students had the chance to present their work to their peers and to the senior scientists, exchange experiences, share results and receive feedbacks from fellow students and lecturers.</p><p>In this presentation, we will present the concept and key features of the summer school, content and organisation, and also offer the students’ feedback collected after the school in an effort to showcase an example of how summer schools remain a powerful mean to value diversities and create an inclusive environment in (Earth system) science.</p>


Author(s):  
Sophie Valcke ◽  
René Redler ◽  
Reinhard Budich

2013 ◽  
Author(s):  
Wolfgang Hiller ◽  
Reinhard Budich ◽  
René Redler

Sign in / Sign up

Export Citation Format

Share Document