COMTESS- Sustainable Coastal Land Management: Trade-Offs in Ecosyste

Coastal regions will be especially affected by climate change, i.e. rising sea- levels, increase of storms and winter precipitation. Ecosystems might change in ways that affect the stability of ecosystem functions and, thus, the provision of ecosystem services that coastal communities rely on today.

This project will investigate different land use scenarios for sites at the German North and Baltic Sea coast under climate change. The main focus is to evaluate ecosystem functions and services provided by these coastal regions and how they are affected by environmental change and different land management options. Work package 5 (“Modelling of biodiversity and plant- mediated ecosystem services (ESS) in response to land use management and environmental change”) is undertaken by the University of Potsdam. Part 1 of this sub- project lies with the Institute of Earth and Environmental Science, part 2 with the Institute of Biochemistry and Biology.

In detail, WP 5 will yield the ecological evaluation of the COMTESS scenarios by statistically modelling the functional and response diversity (WP 5.1) and developing a process- based model for the spatio- temporal dynamics of key species and plant- mediated ESS (WP 5.2). For this purpose, we will upscale the plot-level one (or two)-year measurements of WP 1 - 4 to the landscape scale and to time scales ranging from 2010 to 2100. The statistical niche modelling of WP 5.1 will be enhanced with the transient plant community dynamics captured by the individual- based model of WP 5.2. Finally, the insurance effect of functionally redundant species on the resilience of ESF and, thus, the provision of ESS will be analysed for variable environmental conditions. The predictions from WP 5 will be summarized in a GIS modelling shell used by later work packages for the ecological- economic analysis. More...

Researchers: Anett Schibalski, University of Potsdam; Katrin Körner, University of Potsdam

Duration: 2011 – 2014

Funded By: BMBF (FONA, Sustainable Land Management)


Spatiotemporal dynamics of shallow landslides and their interaction with the biotic environment

Subproject D7: Forest dynamics, landslide dynamics, and their interactions - a process-based landscape modelling approach for a mountain rain forest in South Ecuador

As part of a DFG research project (DFG-Research Unit/Forschergruppe FOR 816, Phase 2: Functionality in Tropical Mountain Rainforest) on the understanding of tropical species biodiversity, the project aims at understanding driving factors of landslides within undisturbed regions of tropical montane rain forests of the Ecuadorian Andes.

Shallow translational landslides represent the most important natural ecosystem disturbance within these areas. Since ecosystem disturbance regimes constrain patterns of biodiversity, it is important to understand the environmental factors that control the disturbance in order to predict the impact of possible future changes.

For the purpose we utilize statistical landslide susceptibility models as well as physically based models of landslides and forest dynamics. During a half year field campaign a lot of data has been collected that is now used to calibrate and test these models. Some measurements are still going on in the Andes of south Ecuador.

Within the ongoing phase of the project we create and upscale a spatiotemporal dynamic simulation model that allows to simulate and understand natural landslide patterns. The comprehensive model incorporates forest dynamics as well as landslide dynamics and will be an important contribution to a carbon balance for tropical montane forests. More...

Researchers: Peter Vorpahl, University of Potsdam; Andreas Huth, UFZ Leipzig-Halle

Duration: 2010-2013

Funded by: DFG

BIOPORE - Linking spatial patterns of anecic earthworm populations, preferential flow pathways and agrochemical transport in rural catchments: an ecohydrological model approach

Earthworms play a pivotal role in agro-ecosystem functioning by modulating soil structure that significantly influences soil hydraulic properties, organic matter dynamics, and plant growth. This project focuses on anecic earthworms like Lumbricus terrestris which create vertical semi-permanent burrows that function as preferential flow pathways. Preferential flow in macropores is a key process which strongly affects infiltration and may cause rapid transport of pesticides into depths of 80 to 150 cm where they experience a much slower degradation. Therefore, preferential transport is an environmental problem because the topsoil is bypassed, which has been originally thought to act as a filter to protect the subsoil and shallow groundwater. Assessing the environmental risk of pesticides in earthworm burrows and how human management practise feedbacks on that risk requires the development of an integrated eco-hydrological model. This model allows predictions of i) the spatiotemporal distribution and population dynamics of anecic earthworms, ii) the related pattern of connective preferential flow pathways (i.e., earthworm burrows), and iii) the space-time pattern of infiltration and travel depth distribution of solutes. This enables the understanding of how small-scale patterns regulate large-scale processes in rural landscapes and how feedbacks between earthworm engineering and transport characteristics affect the functioning of (agro-)ecosystems. We expect our final model to be applicable for catchment-scale risk assessment that may assist agrochemical registration. More...

Researchers: Juliane Palm and Loes van Schaik, University of Potsdam; Julian Klaus and Erwin Zehe, Technische Universität München

Duration: 2007-2011

Funded by: DFG

Assessment of Environmental Functions for Forest Landscape Restoration in Central Chile

Dryland forests are being threatened due to land use pressures and limited natural regeneration. The PhD-thesis focuses on dryland forest dynamics in Central Chile with the objective to identify areas of priority for forest restoration, following the principles of “Forest Landscape Restoration” (FLR). It implies the evaluation of existing and potential forest functions on a landscape scale to optimize the delivery of environmental services through forest restoration. On the basis of an analysis of land cover change using remote sensing and GIS, the proximate causes of vegetation cover change are assessed with statistical models including biophysical and socio-economic variables.

For the identification of forest restoration priorities, a spatial decision support system is used. Therefore, perceptions about the relevance of forest related functions identified in local workshops are being evaluated, taking into account experts’ opinions. Proxy maps for the selected potential forest functions are being developed using land cover classifications and available GIS-Data. On the other hand, areas suitable for restoration are being identified, and according to the relevance assigned to potential forest functions priority areas for forest restoration shall be identified using Multi-Criteria-Analysis.

Researcher: Jennifer Schulz, University of Potsdam

Duration: until 2011

Funded by: European Commission, REFORLAN Project, INCO Contract CT2006-032132


Understanding downstream migration of European eel (Anguilla anguilla) - Analysis and modelling of migration triggers –

The European eel stock is in steep decline and consequently the species has been added to the IUCN Red List of Threatened Species as critically endangered. In order to reduce the anthropogenic mortality caused by hydroelectric power plants turbines, it is absolutely necessary to identify the environmental triggers for downstream migration towards their maritime spawning grounds. The aim of this project is to identify and quantitatively describe these triggers and develop predictive models. Therefore, environmental data will be combined with catch data and infrared video data from several trapping sites in three “top-11 rivers” in southern Sweden by means of advanced statistical modelling approaches. This will help to identify environmental windows of optimal migration conditions, described by threshold combinations of triggers. The results will enable an enhanced turbine management for hydroelectric power plants which will consequently contribute to achieve the goals of the Swedish Eel Management Plan.

Researchers: Florian Stein and Boris Schröder, Technische Universität München; Olle Calles, Karlstads Universitet; Johan Östergren, Sötvattenslaboratoriet SLU Aqua; Uwe Brämick, Institute of Inland Fisheries in Potsdam-Sacrow

Duration: 03/2012-12/2013

Funded by: Elforsk ‘Krafttag ål’

Mechanistic niche modeling: The spatiotemporal distribution of earthworms and their effects on ecosystem functioning

Species distribution models are usually based on correlative or machine-learning methods and assume a quasi-equilibrium of the environment and niche conservatism of the species. The description of spatiotemporal distribution patterns along several environmental gradients is qualified to detect the realized niche of a species and thus an approximation of the potential distribution area of the species. However, the quasi-equilibrium approach is not firm concerning establishing or spreading species as also the so called “ecosystem-engineers” which steadily change their habitat, e.g. earthworms. A process-based approach to approximate species distribution is assumed to come closer to the fundamental niche of the species. Therefore a process-based niche model will be build that links population dynamics and distribution as also the effects of these dynamics on the soil environment. The effectivity of single or combined processes on different scales (spatially: from lysimeter to field scale, temporally: from seasonal changes to decades) and the quantification of the effects are in focus of this work. Furthermore, the model is assumed to afford the simulation of spatiotemporal changes in soil structure through earthworm activity which significantly influence water, carbon and nitrogen fluxes in the soil system.

Researcher: Anne-Kathrin Schneider

Duration: 2010-2011 

Funded by: ZALF


Biodiversity and Sustainable Management of a Megadiverse Mountain Ecosystem in South Ecuador

DFG-Research Unit/Forschergruppe FOR 816 (Phase 1) Functionality in Tropical Mountain Rainforest

Subproject A3.3: Spatial-temporal dynamics of landslides and their biotic & abiotic controls

Researchers: H. Elsenbeer, M. Märker and P. Vorpahl, University of Potsdam; A. Huth, UFZ Leipzig-Halle; B. Huwe, University of Bayreuth

Duration: 2007-2010

Funded by: DFG

University of Potsdam Graduate Initiative on Ecological Modelling (UPGradE)

Coordination together with Florian Jeltsch, Frank M. Schurr, University of Potsdam

Integrating dynamic processes into species distribution models to improve predictions for scenarios of environmental change 

Species respond to changes in climate, land use and land cover by distributional changes. Their geographical ranges may be contracting, expanding or shifting in space. Robust predictions of these changes are needed to inform dynamic and sustainable conservation strategies. Many studies have applied statistical species distribution models (SDM) to predict species range changes and extinction risks. SDM derive statistical relationships between species' occurrence and prevailing environmental factors to describe the environmental niche. They have comparably low data requirements, (free) software is widely available, and they allow rapid assessments of potential threats as projections of potential future species' distribution are made by directly transferring the statistical relationship to expected future environmental conditions. However, due to their static nature, transient responses to environmental changes are essentially ignored, and SDM incorporate neither dispersal, demographic processes nor biotic interactions explicitly. Therefore, it has often been suggested to link statistical and mechanistic modelling approaches. In this project, we show two different ways of such linkage that can both improve our understanding of species distribution patterns. 

(1) Mechanistic modelling as virtual playground for testing statistical models allows extensive exploration of specific questions: We call this the 'Virtual Ecologist' (VE) approach and promote VE as a powerful evaluation framework for testing sampling protocols, analyses and modelling tools (Zurell et al. 2010). Using a VE approach, we systematically test the effects of transient dynamics and ecological properties and processes on the prediction accuracy of SDM for climate change projections (Zurell et al. 2009). 

(2) Integrating dynamic processes into SDM: Results from (1) provide valuable insights about relevant processes accompanying species range changes. Consequently, we supplement projections of potential future habitat for black grouse in Switzerland with an individual-based model that describes demographic processes and dispersal, and improves our predictions by explicitly depicting persistence and extinction rates through time, as well as colonisation success in space and time (Zurell et al. in revision). 

Keywords: climate change, environmental niche, individual-based, metapopulations, range dynamics, spatially explicit, species distribution models, theoretical ecology, virtual ecologist

Region: Switzerland, Virtual

Researchers: Damaris Zurell, Boris Schröder, University of Potsdam

Collaborators: Volker Grimm, Dept. Ecological Modelling, Helmholtz Centre for Environmental Research - UFZ; Carsten F. Dormann, Dept. Computational Landscape Ecology, Helmholtz Centre for Environmental Research - UFZ; Eva Rossmanith and Florian Jeltsch, Plant Ecology and Nature Conservation, University of Potsdam; Niklaus Zbinden, Swiss Ornithological Institute; Niklaus E. Zimmermann, Dept. Land Use Dynamics, Swiss Federal Research Institute WSL;

Duration: 2007-2010

Funded by: University of Potsdam Graduate Initiative on Ecological Modelling - UPGradE

CarboZALF - Carbon budget of agricultural landscapes under global change

Researchers: J. Augustin, J. Hufnagel, C. Kersebaum, M. Sommer, G. Verch, S. Wirth, R. Ellerbrock, G. Churkina, D. Deumlich and W. Hierold, ZALF Centre for Agricultural Landscape Research Müncheberg

Duration: 2009-2012

Funded by: ZALF

White stork (Ciconia ciconia) in Brandenburg: Analysing environmental effects on long-term reproductive performance with GIS-based species distribution modelling

Researchers: Ute Eggers and Dieter Wallschläger

Duration: 2009-2011

Funded by: DBU

CLIMBIRD and Modelling Swiss Breeding Birds

Researchers: R. Maggini-Lehmann, N. Zbinden, Swiss Ornithological Institute, Sempach, Switzerland; N. E. Zimmermann, J. Bolliger, Swiss Federal Institute for Forest, Snow and Landscape Research WSL, Switzerland; J. Elith, University of Melbourne