Climate Modelling

We investigate the coupled Arctic climate system with the overall aim to (i) improve our understanding and model representation of key climate processes in the Arctic and (ii) to improve our understanding of Arctic – midlatitude linkages. We apply regional and global climate models for associated research. Our climate modelling research is part of national and international research projects, such as AC3, QUARCCS, REKLIM, Arctic CORDEX, and MOSAiC.


The phenomenon of Arctic amplification has a suite of causes, which include various interconnected processes and feedbacks, such as sea ice loss and albedo feedback, meridional atmospheric and oceanic energy fluxes, and radiation-climate feedbacks linked with temperature, water vapor, clouds and ozone. The relative importance of these different feedback mechanisms is still subject of debate. In parallel, climate models have difficulties in reproducing the observed drastic Arctic climate changes and the uncertainty in Arctic climate projections is large. That arises, in large part, from gaps in our understanding of key Arctic processes and feedbacks.

A large body of evidence demonstrates how changes in the climate of the Arctic impact lower latitudes (Cohen et al., 2014). For example, the decline in Arctic summer sea ice concentration is connected with atmospheric circulation responses in the following winter months and linked to anomalous cold winters over Eurasia and other regions of the Northern Hemisphere (Jaiser et al., 2013, 2016; Handorf et al., 2015). The Arctic is coupled with lower latitudes via horizontal advection of heat and moisture as well as through planetary waves in the coupled troposphere-stratosphere system. However, the coupling and impacts of the Arctic climate system to lower latitudes is not fully understood. To disseminate the knowledge on Arctic-midlatitude linkages we gained through our research we are contributing to the Earth System Knowledge Platform (ESKP) of the Helmholtz Association of German Research Centres.

Regional Modelling

HIRHAM atmosphere model

HIRHAM is a state-of-the-art atmospheric regional climate model (Christensen et al., 2007), which we apply to the circum-Arctic domain. The model uses the dynamics of the numerical weather prediction model HIRLAM7 and the physical parameterizations of the global model ECHAM5. We apply the model for various studies analyzing atmospheric energy budgets (e.g., Sommerfeld et al., 2015), key atmospheric processes (atmospheric boundary layer and clouds, e.g. Klaus et al., 2016). Furthermore, the model is used to help interpreting observed long-term trends in the Arctic.

HIRHAM-CLM atmosphere-land model

The original ECHAM5’s land-surface-soil scheme has been replaced in HIRHAM by the advanced land model CLM4.5 (Community Land Model version 4) (Matthes et al., 2017). This model is much more complex in its descriptions of vegetation and soil processes than ECHAM5’s inbuilt land component. CLM4.5 have been widely used in modelling permafrost-related processes. It significantly reduces the simulated bias in active layer thickness  and winter soil temperatures, which significantly feed back to the atmospheric circulation. Recently, an update with respect to CLM5 has been started.

HIRHAM-NAOSIM atmosphere-ice-ocean model

HIRHAM-NAOSIM model couples the atmospheric regional model HIRHAM and the regional ocean-ice model covering the North Atlantic/Arctic Ocean (NAOSIM; Karcher et al., 2003; Kauker et al., 2003). A detailed description of the model is given by Dorn et al. (2007). The model system was improved in key parameterizations, such as of sea ice and snow albedo, snow cover on sea ice, ice growth (Dorn et al., 2012) which lead to a more realistic onset of the summer sea ice melt and simulation of the snow/ice-albedo feedback. Further, regional feedback processes between summer sea ice anomalies  and atmosphere have been discussed (Rinke et al., 2013). Recently, an update of the model system has been started, including higher resolution and improved model components.


Global Modelling

Global Earth system models AWI-CM and AWI-CM-SWIFT

The earth system model AWI-CM has been developed at our institute in the framework of the project TORUS (TOwards Regionally focUsed modelling of decadal climate predictionS, funded by BMBF). The model consists in the Finite Element Sea-Ice Ocean Model (FESOM, developed at AWI) coupled to the atmospheric model ECHAM6 (developed at MPI Hamburg). A detailed description of the model is given by Sidorenko et al. (2015). The model has been applied to the study of climate variability in Rackow et al. (2016). By coupling the SWIFT fast scheme for simulating the chemistry of stratospheric ozone depletion in polar winter (Wohltmann et al., 2017) to ECHAM6, the atmospheric component of AWI-CM, we obtained an Earth system model which enables the simulation of interactions between the ozone layer and climate. Currently, we are using ECHAM6 and ECHAM6-SWIFT for the study of the role of tropo-stratospheric interactions and of stratospheric ozone for Arctic-midlatitude linkages.

Global atmospheric models with reduced complexity

In addition to complex global Earth system models, we developed and applied a hierarchy of global atmospheric models with reduced complexity. These models are quasi-geostrophic three-level models with varying horizontal resolution (e.g., Sempf et al, 2007; Labsch et al., 2015) and simplified physical parameterizations. The models simulate the spatiotemporal evolution of the Northern Hemisphere large-scale atmospheric circulation very well and thus serve as an idealized tool for the study of low-frequency atmospheric variability and atmospheric circulation regime behavior. In a recent study we investigated the role of quasi-geostrophic dynamics for Arctic-mid-latitude circulation linkages, which remains obscure when using comprehensive Earth-system models (Handorf et al., 2017).



Regional Modelling
Dr. Annette Rinke

Global Modelling
Dr. Dörthe Handorf