Advanced Prediction in Polar regions and beyond: modelling, observing system design and LInkages associated with a Changing Arctic climate
APPLICATE is an €8 million project financed by the EU HORIZON 2020 Research and Innovation programme that involves 16 partners from nine countries (Belgium, France, Germany, Iceland, Norway, Russia, Spain, Sweden and the United Kingdom) and will be carried out over a period of four years. The multinational and multidisciplinary consortium will work to enhance weather and climate prediction capabilities not only in the Arctic, but also in Europe, Asia, and North America. A focus on the Arctic is important for improved predictions of weather and climate in the mid-latitudes because the changes taking place in the Arctic due to climate change — the retreat of sea ice, warming seas and a warming atmosphere — have the potential to influence weather and climate in the mid-latitudes.
Luisa Cristini | Claudia Hinrichs | Thomas Jung | Sara Pasqualetto | Tido Semmler
Follow APPLICATE on Twitter @applicate_eu.
"Processes and impacts of climate change in the North Atlantic Ocean and the Canadian Arctic" aims to educate PhD students in an interdisciplinary environment that combines the strength in marine geosciences and environmental physics in Bremen with complementary skills and expertise in sea-ice and ice-sheet modelling in a consortium of eight Canadian partner universities. The scientific team with the PhD students in its centre aims to advance the understanding of the variability of the Arctic Ocean and the cryosphere on time scales of decades to millennia and to use these results to robustly assess the impact of projected future climate changes on the Arctic.
Deniz Aydin | Yuqing Liu | Martin Losch
The eFlows4HPC project brings together 16 academic and industrial organizations from across Europe. The AWI work together with leading European HPC (High Performance Computing) experts on making climate simulations with a new generation climate model easier to use and more efficient.
The aim of the project is to deliver an innovative, HPC based workflow platform. It will consist of the software stack and an additional set of services that enable the integration of HPC simulations with big data analytics and machine learning in scientific and industrial applications. AWI participate in the project with a coupled model, that consists of next generation unstructured mesh ocean model FESOM2 (fesom.de) as its ocean and OpenIFS (https://www.ecmwf.int/en/research/projects/openifs) as atmospheric component. The coupled model will be integrated into the workflow platform in the context of a climate modelling use case.
EPICA
"Eddy Properties and Impacts in the Changing Arctic" (EPICA) is a BMBF funded project that will use high-resolution modeling and observations to understand the properties and impacts of (sub)mesoscale eddies in the Arctic Ocean, especially in the MOSAiC period in comparison with the past. Interactions of the ocean with the atmosphere and sea ice, as well as mixing and transport of water masses in the ocean, are governed by numerous small-scale turbulent processes. In particular, mesoscale and submesoscale eddies are important features that shape the ocean hydrography and influence the mixed layer dynamics. However, their properties and precise roles in the Arctic Ocean are still poorly understood, with serious implications for our understanding of Arctic Ocean dynamics and for the realism of eddy parameterizations in climate models.
The proposed project will combine the high-resolution modelling capabilities offered by the multi-resolution Finite volumE Sea ice-Ocean Model (FESOM) with the unique year-round data collected during the MOSAiC campaign. We will synthesize the observational data with model results to study processes forming eddies, understand the importance of eddies for the ocean, sea ice and air-sea exchange, estimate ocean internal variability relative to forced variability, and derive eddy diffusivity which can help to improve eddy parameterizations, thus the fidelity of climate models. The resolution of the global model will be increased to 1 km in the whole Arctic Ocean and further refined to subkilometer scales at the MOSAiC sites. The decades-long eddy-resolving model results for the Arctic Ocean and the well-evaluated and documented km-scale model configuration will be delivered to the broad community, which can be used to advance the understanding, prediction and projection of Arctic climate. Within this project we collaborate with the Physical Oceanography Section, which is responsible for analyzing related MOSAiC observations.
Qiang Wang | Benjamin Rabe | Ivan Kuznetsov | Dmitry Sein | Sergey Danilov | Thomas Jung | Nikolay Koldunov | Patrick Scholz | Dmitry Sidorenko | Claudia Wekerle
Advanced Earth System Modelling Capacity: A contribution to solving Grand Challenges by developing and applying innovative Earth System Modelling capacity
ESM is a €10 million project started in April 2017 funded by the Helmholtz Association over a period of three years. The project comprises eight Helmholtz Research Centres and aims to improve the representation of the components of the Earth system and their coupling, as well as to perform a series of selected numerical experiments to address Grand Challenges (Frontier Simulations). A long-term strategy for the development of an Earth System Modelling capacity is also an objective of the project.
Dirk Barbi | Luisa Cristini | Helge Goessling | Thomas Jung | Suvarchal Kumar Cheedela | Lars Nerger | Sara Pasqualetto | Thomas Rackow | Tido Semmler
Follow ESM on Twitter @project_esm.
ESM-TOOLS is a software product developed and maintained at AWI Bremerhaven as part of the Helmholtz project ESM with the aim to unify model infrastructure, giving a common framework for downloading, compiling, running and organizing coupled or standalone models. ESM-TOOLS is composed of three tools: (1) esm-master: Makefile-based tool to download, configure and compile the models (2) esm-environment: Machine-dependant settings for compiling and running of models, all collected in one place. (3) esm-runscripts: collection of functions enabling to use short and concise runscripts, practically identical independent of machine or even model. The underlying functions organize the whole experiment, including copying of data, modifying of namelists, sanity checks, iterative coupling, and much more.
Miguel Andrés-Martínez | Dirk Barbi | Luisa Cristini | Deniz Ural
Follow ESM-TOOLS on Twitter @ToolsEsm.
FRontiers in Arctic marine Monitoring
Our ability to understand the complex interactions of biological, chemical, physical, and geological processes in the ocean and on land is still limited by the lack of integrative and interdisciplinary observation infrastructures. The main purpose of the open-ocean infrastructure FRAM is permanent presence at sea, from surface to depth, for the provision of near real-time data on Earth system dynamics, climate variability and ecosystem change. The Climate Dynamics section supports the FRAM infrastructure program with high-resolution ocean and sea-ice simulations.
Vibe Schourup-Kristensen | Claudia Wekerle
The Helmholtz-Initiative Climate Adaptation and Mitigation: two Sides of the same Coin (HI-CAM) is a two-year initiative for incentivizing cross-cutting research on climate mitigation and adaptation at Helmholtz centres and beyond. This projects brings together the expertise of 50 PIs from 14 Helmholtz centres and is supported with 12 Mio €.
AWI Climate dynamics scientists are integrated into Cluster 2: Adaptation to Extremes. The impact of global warming for various sectors, their interconnections and feedbacks are analyzed with the aim to identify appropriate adaptation measures in a “future projection approach”. To do so, novel storyline scenarios for recent extreme events, such as the heat wave in summer 2019, are explored. In these storylines, the exceptional events are reconstructed in different background conditions (pre-industrial, present, 2 and 4K warmer climates) using the AWI-CM model by imposing the observed development of the large-scale atmospheric circulation (e.g. the jet stream and other large-scale structures) by employing a scale-dependent spectral nudging approach. Thereby the highly uncertain dynamical changes are factored out, and thermodynamic effects modifying extreme events can be investigated in isolation with a high signal-to-noise ratio. Focussing on scenarios for recent extreme events that are fresh to people’s memory is expected to make climate change more tangible and provides a new event-based approach toward adaptation.
In this project, two main hypotheses will be tested: (1) Machine learning will lead to much better parametrizations in climate models and (2) machine learning methods can help to overcome computational bottlenecks in high-resolution model runs on extreme-scale highperformance computers. Firstly, it is planned to develop new parameterizations for representing ocean eddies in the Finite Volume Sea Ice-Ocean Model (FESOM2.0), which has been developed at AWI. Secondly, it is planned to go one step further by using ML methods for replacing certain parts of Earth system models that present computational bottlenecks for high-resolution configurations. Furthermore, it is planned to examine how satellite observations can be integrated in the ML model. Here, we aim in particular at obtaining very high-resolution models, where the underlying mathematical model leads to an ill-posed inverse problem. This requires adapting regularized deep learning approaches based on generative adversarial network architectures. Improve the realism of climate models and hence their ability to project future climate change.
Sonal Rami | Thomas Jung | Stephan Juricke | Nikolay Koldunov
The Pilot Lab Exascale Earth System Modelling (PL-ExaESM) explores specific concepts to enable exascale readiness of Earth system models and associated work flows in Earth system science.
Earth system models simulate the Earth’s climate including physical and biogeochemical processes and are important tools for assessing the magnitude and impacts of future climate change. Such models have to run in ever finer resolutions in order to capture extreme events, which are often responsible for the majority of weather-related and other environmental impacts. With our simulations local scale phenomena can be reproduced and extreme events will be simulated much more accurately than at present. To use needed supercomputers efficiently, new programming concepts must be developed and implemented into Earth system models. Furthermore, these models generate huge amounts of data, and storage technology is also evolving. This implies that new ways for handling Earth system model output and new modelling workflows have to be developed. To enable such large simulations on such gigantic machines necessitates a close collaboration between Earth system modellers and computer scientists. The PilotLab ExaESM offers a platform where these two communities can meet and interact to build the next-generation Earth system models.
In the PL-ExaESM, scientists from 9 Helmholtz institutions work together to address 5 problems of exascale Earth system modelling such as scalability or system design.
Jan Hegewald | Thomas Jung
The Weddell Sea in the Atlantic sector of the Southern Ocean is one of the most dynamic air/ice/ocean interaction areas and the Larsen Ice Shelf adjacent to the Antarctic Peninsula has undergone dramatic changes in recent decades which need to be understood. Additionally, the Antarctic continental shelves and the exchange of shelf water with the open ocean play a key role for global ocean circulation. The aim of REDOCCA is to study the impact of changes in the atmospheric conditions on the ocean circulation in the Weddell Sea for the mid and the end of the 21st century. Several simulations of AWI-CM, produced for CMIP6, regional high-resolution simulations of COSMO-CLM (run by colleagues at the University of Trier), and FESOM stand-alone simulations will be analysed with a focus on polynya representation and the formation of shelf water. Climate developments of the last 160 years as well as the 21st century are investigated. The focus will be on the representation of local wind systems, shelf water production, sea ice production and melt, and ice shelf basal melt rates adjacent to the Weddell Sea, especially the Larsen Ice Shelf.
Tido Semmler | Vanessa Teske | Ralph Timmermann
The Sea Ice Drift Forecast Experiment (SIDFEx) is a community effort to collect and analyse Arctic sea-ice drift forecasts, with the primary goals to advance ice drift forecast capabilities and to diagnose model deficiencies. Amongst other things, SIDFEx delivered real-time drift forecasts for the MOSAiC Central Observatory and the Distributed Network, by streamlining and merging individual forecasts from multiple international forecast centres and research groups. However, since SIDFEx is largely a community effort without dedicated funding, resources are lacking to exploit the unprecedented opportunity the SIDFEx data in combination with the multitude of MOSAiC observations are to improve drift forecasts and to diagnose errors of the underlying models. SIDFExplore is meant to carry SIDFEx to a completely new level so that significant progress towards better Arctic forecasts can be made. Specifically, SIDFExplore will:
- evaluate the skill of individual drift forecasts in terms of metrics such as distance, speed, angle, and deformation, and diagnose how the skill depends on parameters such as atmospheric state, ice and snow thickness, and geographical region;
- elaborate on these findings to guide forecast system improvements, in close collaboration with the external partners that contribute forecasts to SIDFEx;
- optimise, develop, and test new methods, including based on machine learning, to perform “on-the-fly” calibration and to generate multi-model consensus drift forecasts;
- ensure additional long-term impacts of SIDFEx by publishing papers, data, and software, by planning of a possible continuation of the SIDFEx system beyond MOSAiC, and by exploring ways to transfer know-how to the Search-and-Rescue sector (including open-water).
With this scientific agenda and the wide participation of leading international weather and climate modelling centres, SIDFExplore is ideally positioned to exploit MOSAiC observations in order to improve climate models, one of the most central and original motivations for MOSAiC as put forward by the Atmospheric Working Group of the International Arctic Science Committee (IASC).
Helge Goessling | Lorenzo Zampieri
Ocean mass varibility on time scales of months to decades is still insufficiently understood. On these time scales, large-scale bottom pressure anomalies are associated both with wind induced variability as well as baroclinic processes, i.e. related to vertical shear of ocean density. The GRACE mission has been instrumental in quantifying such mass fluctuations, yet its lifetime is limited. However, the broader importance of non-tidal ocean mass variability for geodesy and oceanography, i.e. for understanding the time varying geoid, shape of the Earth's crust, centre of figure, Earth rotation, is obvious. Similarly, deep ocean processes can only be understood properly when not only sea surface height and upper ocean steric expansion are measured but deep ocean pressure anomalies are accounted for in addition. Only the knowledge of all three terms allows estimates of deep ocean warming and helps in understanding and predicting sea level rise. In "Consistent Ocean Mass Time Series from LEO Potential Field Missions (CONTIM)" we propose to combine expertise on precise satellite orbit determination, gravity field and mass modelling, and physical oceanography to retrieve, analyse and verify consistent time series of ocean mass variations from a set of low-flying Earth orbiters including GRACE. This information is used to further our understanding of oceanic movement, ocean warming and sea level rise.
Alexey Androsov | Sergey Danilov | Jens Schröter
Seamless Sea Ice Prediction
The BMBF Young Investigator Group SSIP (2017—2022) works towards advancing sea-ice prediction capacity on timescales from hours to years and beyond. To achieve this, SSIP develops and conducts research with a seamless sea-ice prediction system based on the recently developed AWI Climate Model. The unstructured grid of the ocean/sea-ice component of this model (FESOM) allows to use high resolution in the polar regions (plus other key regions) in a global setup, enabling a seamless application of the prediction system on a wide range of timescales. The group applies state-of-the-art techniques to initialize the prediction model using remote-sensing and in-situ observations; it optimizes and further develops the sea-ice component of the prediction model; and it applies the prediction system to address research questions related to sea-ice predictability, verification, and the impact of different observations on sea-ice prediction.
Marylou Athanase | Helge Goessling | Svetlana Loza | Bimochan Niraula | Simon Reifenberg | Lorenzo Zampieri
TRR181 is a DFG funded project about energy transfers in atmosphere and ocean.
The energy of a closed system is steady. It is not lost but rather converted into other forms, such as when kinetic energy is transferred into thermal energy or vice versa heat results in a force.
However, this fundamental principle of natural science is often still a problem for climate research. For example, in case of the calculation of ocean currents, where small-scale vortices as well as mixing processes they induce need to be considered, without fully understanding where the energy for their creation originates from. This is similar in the atmosphere, the only difference being that air is moving instead of water. Again, local turbulences can drive larger movements or vice versa waves on a larger scale can disintegrate into small structures.
All these processes are important for the Earth’s climate and determine how temperatures will rise in the future.
Deniz Aydin | Sergey Danilov | Thomas Jung | Stephan Juricke | Nikolay Koldunov | Martin Losch | Dirk Olbers | Patrick Scholz
The Year of Polar Prediction (YOPP) is a major international activity that has been initiated by World Meteorological Organization’s World Weather Research Programme (WWRP) as a key component of the Polar Prediction Project (PPP). YOPP takes place from mid-2017 to mid-2019. Its overarching goal is to significantly advance our environmental prediction capabilities for the polar regions and beyond.
As an internationally coordinated period of intensive observing, modelling, prediction, verification, user-engagement and education activities which involves various stakeholders, the Year of Polar Prediction will contribute to the knowledge base needed to managing the opportunities and risks that come with Arctic climate change.
Helge Goessling | Thomas Jung | Katharina Kirchhoff | Sara Pasqualetto | Felix Pithan | Kirstin Werner
Follow YOPP on Twitter @polarprediction or visit the YOPP YouTube Channel.
For more information also see the our leaflet and our infographics.
