SFB299 "Land-Use Options for Peripheral Regions" - Subproject B 3.1

Four project periods: 1997 - 2008

• 2006 - 2008  Spatially explicit modelling of phytodiversity in cultural landscapes:Model application, adaptation and extension (Project Homepage)
  
• 2003 - 2005  Modules for modelling land-use dependent floristic diversity in cultural landscapes (Report in German)
• 2000 - 2002  Functions and services of floristic biodiversity for future landscape changes (Report in German)
• 1997 - 1999  Spatial survey and assessment of deintensification processes with regard to their impact on floristic, community and process diversity. (Report in German)
• see Project overview 1999-2002 (large pdf-file)

Project management:

Prof. Dr. Dr. Annette Otte, PD Dr. Rainer Waldhardt

Dr. Dietmar Simmering, Dr. Birgit Reger, Dr. Camilla Wellstein, Dr. Katja Fuhr-Boßdorf

Land use concept, modelling of species occurrence and numbers, agricultural landscape, biodiversity, phytocoenotic processes, fallow, key species, landscape structur, ecosystem engineers, indicator species, aerial photograph interpretation, GIS, Central Hesse

Trade-offs between various landscape functions (e.g., economic and ecological measures) are of special interest in agricultural landscape research. The landscape ecological subproject within the SFB299 is therefore concerned with relations between land use and phytodiversity. In this context, we consider the spatio-temporal patterns of different land-use systems (cultivation, grassland, fallow) as well as abiotic site conditions and socio-economic variables; i.e. the most important driving forces of phytodiversity. The emphasis is laid on compositional aspects of phytodiversity and their measures (e.g. plant species richness, composition and abundance), but we consider structural and functional components of diversity as well.Our study area during the project phases between 1997 and 2005 was the highly fragmented, marginal mosaic landscape of the Lahn-Dill Highlands. Our studies in this region confirmed the significance of spatio-temporal land use dynamics of nonlinear and linear elements (habitats) as determinants of phytodiversity at the landscape scale. Further, the results clearly reflected the need to consider local species pools.The results eventually gave way to the development of the probabilistic model ProF (Prognosis of Floristic Ric

hness). The model is based on calculations of the probability of single species to occur in differently sized, heterogeneous spatial units with multiple habitat patches, i.e. we use patch-based field data to model species richness and species occurrence in multi-habitat environments (patterns) at the landscape scale. The link between the patch and pattern level is facilitated by a GIS-based classification of habitat types. For the Lahn-Dill Highlands (1.100 km2), we generated a map containing 500 habitat types (excluding linear elements). The modelling approach thus combines steps of bottom-up and top-down analyses to produce quantitative and qualitative estimates of phytodiversity. So far we received satisfying model results for landscape tracts varying in size from 0.22 to 650 km2. Validations of model results were accomplished by comparing them with field survey data.

We applied the model within the SFB-model network ITE2M to evaluate effects of land-use scenarios generated by the bio-economic simulation model ProLand (Scenario Agenda 2000, CAP reform) on phytodiversity, and compared ProF model outputs on species richness and the occurrence of arable weed species for both scenarios: The CAP reform will in this respect have considerable negative effects on species richness and the occurrence of arable weeds at the landscape scale.

For the fourth period of our project, we aim at the following objectives:

We will continue to apply ProF to highly fragmented regions, i.e. subareas (60 km2) of the Lahn-Dill Highlands and of marginal parts of the Nidda catchment (Hintertaunus and Vogelsberg ranges). We will use the existing habitat map for the Lahn-Dill Highlands and will prepare analogous data for the new study areas. Additionally, we will have to conduct patch-related floristic surveys to generate an appropriate data base for probability calculations. The model will produce estimates of present richness for certain species groups and predictions of the present spatial distribution of single species. Moreover we will analyse the impact of different grain sizes and pattern extents on model results. Also, the effects of local species pools on model applications will be clarified.

Secondly, we aim to adapt and extend the model to make it more appropriate for the application in highly intensive agricultural regions. For this purpose, we will conduct field surveys in the larger part of the Nidda catchment, the Wetterau region. Recognising the larger field sizes and a broader range of patch sizes (which result from an intersection of site and land use data), we will have to incorporate ?within habitat? species-area relations into the modelling approach. Eventually, we plan to extend the model by including the linear landscape elements and their contribution to local species pools. The significance of grain and extent for model results will also be object of our analyses. In this context, we will test differences between the options to either analyse habitat patterns with a standardised size, as presently implemented in ProF, or to use ?functional? or ?ecological? land units as the basic referential unit.

During the last project period, we will apply ProF to additional ProLand land-use scenarios generated in subproject A1. Complementary, we will reverse the hitherto existing approach by generating a land-use pattern that aims at the preservation or regeneration of phytodiversity in the Wetterau region. This scenario will then be used by ProLand to calculate the regional added value and thus the costs of sustainable phytodiversity management.

Funding: DFG

see publication database of the SFB 299