SODEEP

SODEEP - Study Of the Development of Extreme Events over Permafrost areas

SODEEP was a 2-years EraNet-Russia project that ran from November 2018 - January 2021. It was coordinated by HZG (Institute of Coastal Systems) and project partners were AWI Bremerhaven, Tyumen State University (TSU, Russia), University of Fairbanks Alaska (UAF) and West University of Timisoara (WUT, Romania). SODEEP was examining the impact of climate change on the permafrost degradation and related biophysical feedbacks over various spatial and temporal scales. The focus regions were major bio-climatic zones of the Russian Arctic and sub-Arctic, which are not only extremely vulnerable habitats to the permafrost degradation due to climate change, but also exhibit a threat for the global climate due to frozen soil-bounded carbon.

The project brought together data from long-term in situ observations and field studies as well as medium and highresolution satellite-borne and aerial observations with state of the art hierarchy of numerical models both in global and regional climate simulations as well as the permafrost standalone model. Novel techniques for the interpretation of remote sensing imagery as well as a permafrost temperature dynamics and active layer thickness database were developed. This allowed identifying missing or misrepresented permafrost-relevant key processes in numerical models providing adequate scale-dependent representations of these processes. This enabled the investigation of the future impact of extreme events on permafrost areas.

The project comprised four work packages:

WP1. Development of an integrated remote sensing methodology to process and interpret satellite imagery and aerial photography and establishing a permafrost dynamics database.

WP2. Specification of optimal permafrost parameterizations for the modelling and production of high resolution permafrost (~1km) maps.

WP3. Identification of key processes which determine the generation of extreme events.

WP4. Changes in extreme events and their impact on permafrost areas.

The diverse project consortium with expertise in in-situ observations, process modelling, satellite remote sensing data and Earh system modelling was  very important for the success of the project, which enabled the following main results:

• A collection of field data in the Russian European Arctic is available at TSU and parts of it are also included in the CALM database. However, due to lack of funding, these have not been transferred from TSU to a common database with the remote sensing data.
• An integrated remote sensing methodology for processing and interpreting satellite imagery and aerial photographs has been developed and utilized in two publications. Here, the methodology is evaluated and a comparison of identified changes in the satellite data with field measurements was conducted in several permafrost areas. One the one hand, 35 Years of Vegetation and Lake Dynamics were derived in the Pechora Catchment in the Russian European Arctic. The study highlighted the general agreement between climate data and satellite imagery derived indices, suggesting the applicability of the methodology to use remote sensing derived indices as an indicator of lake and vegetation dynamics for regions with sparse available field observations.
• On the other hand, the usefulness of very high resolution (VHR) imagery was demonstrated for detecting different types of disturbance permafrost areas using three example regions in different permafrost zones. The study focused on the detection of subtle changes in land cover classes, thermokarst water bodies, river dynamics, retrogressive thaw slumps and infrastructure in the Yamal Peninsula, Urengoy and Pechora regions.
• For the first time, a methodology has been developed to identify climatic conditions that can act as an extreme event relevant to permafrost degradation. Based on the ERA-Interim Reanalysis, four main types of events were hypothesised and classified. The agreement between in-situ observations (active layer thickness measurements and meteorological observations) with the ERA-Interim Reanalysis, ESA-CCI-LC land cover data and the remote sensing derived indices is strong evidence for the reliability of this methodology to identify extreme climate events associated with permafrost degradation. Fig. 1 shows a typical example from September 1995. A related joint publication is in preparation.
• The soil processes of the global earth system model MPI-ESM, i.e. in its land surface scheme JSBACH, and in the regional ESM ROM were improved in order to better represent the dynamics in permafrost areas. In particular, discussions within the project revealed that the existing vertical discretisation of the soil in JSBACH was insufficient for this purpose and needed to be refined. With the ESMs, one historical and two scenario simulations until the end of the 21st century were carried out. This allowed projections to be made of future changes in the circum-Arctic permafrost regions.
• In the observations and the ERA-Interim reanalysis, increasing surface temperature seems to be the main driver of permafrost degradation, while components of the hydrological cycle may change the intensity of thawing at the end of the summer season. In contrast, an analysis of the MPI-ESM simulation suggests that net-precipitation and the thickness and extent of snow cover may play at least as important, if not more important, a role in triggering abrupt ground thaw processes.

Other scientific results of SODEEP are:

• Improved, vertical discretisation of the soil model in JSBACH, which is better suited to represent permafrost-relevant processes. This can also be used in other land surface models.
• For the first time, historical and scenario simulations were carried out over the Arctic with a regional ESM that also includes a hydrological discharge model and ocean biogeochemistry.
• Projections of future extent and active layer in permafrost areas.

The project results will help to make climate projections over permafrost regions more reliable with climate and Earth system models.

Fig. 1 Extreme events and permafrost degradation