Understanding climate
for the benefit of society

Climate scientist Jerry Tjiputra at Uni Research. (Photo: Andreas R. Graven)

Project on climate models and Big Data gets Toppforsk funding

Climate scientist Jerry Tjiputra at Uni Research aims to develop a new method to disentangle and predict future climate change in enormous amounts of data.

By  Andreas Graven, Uni Research 

Tjiputra and colleagues now have the opportunity to achieve this goal, as a new project he leads receive prestigious FRIPRO Toppforsk funding from the Research Council of Norway (RCN)

The FRIPRO Toppforsk funding was announced today, January 31st.

This initiative from RCN shall contribute to the development of internationally leading researchers and research environments.

– We are of course very happy to take the leading role in the international research environment, to achieve a better understanding of why some models behave very differently - and eventually: to produce the best projections of regional and global environmental change such as warming or precipitation patterns, says Tjiputra.

He is also affiliated with Bjerknes Centre for Climate Research.

The aim for the project is to develop a novel tool using machine learning, statistic and big data to making it possible to identify the best performing models from which the researchers could get the most reliable projections.

The evaluation panel for FRIPRO Toppforsk stated, among other things, that the project will “improve optimal future climate projections from large model ensembles."

“This is genuinely exciting work that could make a huge impact on the field of climate modelling", the panel added.

– Climate scientists around the world develop an increasing number of these highly complex models, which produce tremendous amount of data everyday. Yet, given our limited resources, only a small fraction of them are analyzed properly for targeted studies or synthesis documents such as the Intergovernmental Panel for Climate Change assessment reports. It is a bottleneck problem, Tjiputra says.

– The new 4-yr project funded by RCN (23mNOK) aims to develop an innovative tool based on the state-of-the-art machine learning and big data technologies that will help us efficiently analyze these huge amount of data, Tjiputra adds.

Center for Big Data Analysis (CBDA) at Uni Research Computing, and Geofysisk institutt (UiB), are the project partners.

The research will provide politicians and other policy makers with a better groundwork, when working on society adaptions to future climate change.

24 research communities get funding in this second Toppforsk round from the NRC

– With this funding, one and a half billion NOK has been invested to develop more world leading research environments in Norway, over a period of two years. The government has a long-term aim on building leading research environments in our country. This is one of the priorities in our long-term plan for research and development. Therefore, we added a growth to FRIPRO in the 2018 government budget to obtain this. I look forward to the results of the Fellesløft in a few years, minister of research Iselin Nybø says.

Project:

Constraining the Large Uncertainties in Earth System Model Projections with a Big Data Approach Project leader: Jerry Tjiputra, Uni Research Climate, and Bjerknes Centre for Climate Research. Partners: Center for Big Data Analysis (CBDA), Uni Research Computing and Geophysical Institute (UiB).

 

Organic

Overturning circulation and its implications for global carbon cycle in coupled models (ORGANIC)

ORGANIC applies for funding to study the physical and biogeochemical interactions in the climate system using state-of-the-art model system.

Organic logo

 

The Team - at Kick off January 2015

ORGANIC kick-off. Photo: Gudrun Sylte
From the ORGANIC kick-off January 30. 2015. In front: PI researcher Jerry Tjiputra, Uni Research. From the left, prof. Ulysses Ninneman, researcher Yongchi Gao (NERSC), Yanchun He (NERSC), prof. Christoph Heinze (UiB). Photo: Gudrun Sylte

The ORGANIC project will enhance our knowledge in climate variability simulatedby the NorESM model and identify uncertainty that comes with its future projections. The focus is to elucidate the linkage between large scale overturning circulation with the biogeochemical cycling in the ocean.

This link is necessary since hydrography tracers such as temperature andsalinity do not give us a comprehensive overview on the overturning circulation.

On the other hand, biogeochemical tracers such as nutrient and CFCs are closely tied to the ocean circulation and can be used as indicators for patterns and ventilation rates of the ocean.

Due to the non linear interactions between climate and ocean carbon cycle, it is vital for an Earth system model to accurately simulate the relevant former and latter processes individually as well as interactively in order for it to produce a sound future climate projections.

 

The outcome of ORGANIC will be of highly relevant for both global and regional climate studies, particularly in regions where the ocean ventilation will be perturbed by anthropogenic forcing.

 

The work is divided in three work packages:

 

WP1: Sensitivity of thermohaline circulation patterns and rates

WP2: Equilibrium states of ocean biogeochemistry

WP3: Marine sediment dynamics

 

The proposed interdisciplinary work will involve scientists from natural, mathematical and computational scientists. The study utilizes the nationally developed Norwegian Earth system model and observational sets from contemporary and paleo periods.

The methods that will be developed throughout the project, the Matrix Free Newton Krylov, will provide novel and efficient approach to increase our understanding in the sophisticated interactions between the physical and biogeochemical processes in the climate system.