Analysis of the geo-ecological parameters controlling the spatial pattern of forest islands and discontinuous permafrost and its modification by forest fire, forest use and climate change in the forest-steppe of Mongolia

Droughts, wildfires and non-sustainable forest use have caused substantial loss of forests in Mongolia over the past decades. This still ongoing loss does not proceed in a uniform way. Various factors, especially topography, hydrology, permafrost, soil properties and human impact, lead to spatial differentiation. The aim of this project is to identify causal relationships between the local constellation of geo-ecological and anthropogenic factors on the one hand, and patterns of discontinuous permafrost, forest distribution, occurrence of forest fires and succession of vegetation after fire events (back to forest or into steppe instead) on the other hand. Also interactions between the factors (e.g. permafrost - forest / forest - permafrost) will be considered.

Based on six hypotheses, the causal chains underlying the loss of forests in Mongolia will be analysed by combining several methodological approaches. The local constellation of factors at selected representative sites will be assessed by geomorphological mapping, determination of the distribution and depth of the permafrost layer by ground penetrating radar, soil and vegetation mapping, analysis of remote sensing data, relief parametrisation, and biomass estimation. The selected study area, between the settlement Tosontsengel in the North and the main summit of the Khangai Mountains in the South, represents a region that experiences frequent forest fires. Extensive wood cutting is going on in the area since the mid-last century. Both, natural forests and forests under human influence in this area, will be subjected to analysis of forest-use and fire history, soil properties, depth of the permafrost layer, hydrological parameters and vegetation. Charcoal, palaeosols, and aeolian deposits, combined with luminescence and radiocarbon dating, will be used for reconstructing forest and landscape development prior to the onset of human influence. This palaeoenvironmental reconstruction will allow for evaluating the dimension of human impact within the bundle of interacting factors controlling the vegetation pattern.

In the next step, the identified and quantified geoecological parameters will be analysed geostatistically. This step will allow for identifying relationships between the local constellation of the factors climate, geology, soils and relief on the one hand, and resulting vegetation patterns on the other hand. The obtained relationships will be used for modelling potential forest distribution patterns for the whole area, using digital terrain models and multispectral satellite images. Finally, we will test the possibility of upscaling the modelling to regional scale, using satellite images of medium scale resolution. Based on the identified causal relationships, areas that are potentially threatened by drought stress, wildfires and succession barriers for fragmented forest stands will be determined, and projections of the further development of vegetation and permafrost will be developed.

Project leaders:

• Prof. Dr. Daniela Sauer, Institute of Geography in Goettingen, department of physical Geography
• Dr. Michael Klinge, Institute of Geography in Goettingen, department of physical Geography
• Prof. Dr. Manfred Frechen, Section of Geochronology and Isotope Hydrology, Leibniz-Institute for applied Geophysic, Hannover

Projectpartner Mongolia:

• Prof. Dr. Jamsran Tsogtbaatar, Institute for Geography und Geoecology, Mongolian Academy of Sciences
• Dr. Uudus Bayarsaikhan, Biology department, National University of Mongolia

Project cooperation:

• Dr. Choimaa Dulamsuren, Albrecht von Haller Institute for plant sciences, department of ecology and ecosystem sciences, Georg-August-University Goettingen
• Prof. Dr. Markus Hauck, Albrecht von Haller Institute for plant sciences, department of ecology and ecosystem sciences, Georg-August-University Goettingen
• Dr. Stefan Erasmi, Institute of Geography in Goettingen, department of cartography, GIS and remote sensing