Biodiversity characterization at landscape level using geospatial modelling technique

Geospatial presentation of habitat status has become a key issue for planni ng conservation. Biodiversity characters of the habitat provide the basis o f prioritizing the sites in conservation effort. Since India is identified as one of the 12-mega biodiversity countries in the world, this task is of great significance. In India, biogeographers have long since recognized Nor th-East India as one of the most diverse regions with respect to flora and fauna, Since historic times. human interventions had started influencing th e landuse and landcover of this region. Deforestation is mainly attributed to shifting cultivation and commercial logging of timber. This has resulted in fragmentation of the landscape. Landscape fragmentation and degradation have been identified as fundamental reasons for biodiversity loss. The pap er uses landscape ecological principles for biodiversity characterization, Satellite remote sensing data have been used for characterization of the la ndscape and stratification for ground inventory. A geographic information s ystem has been used to spatially model the disturbance regimes and to integ rate the ground based nonspatial data with the spatial characters of the la ndscape. The vegetation types and landuse of Meghalaya have been mapped usi ng Indian remote sensing (IRS) time series data. The various parameters (vi z. patch shape. patch size, number of patches, porosity, fragmentation and juxtaposition) have been analyzed on the most recent landcover map to spati ally present the disturbance regimes. A spatial model incorporating ground based biodiversity attributes of the landscape elements, landuse change pat terns, disturbance regimes of the landscape and terrain complexity have bee n used to delineate the spatial pattern of biological richness. This has be en achieved using the landscape analysis package (LAP) developed at the Ind ian Institute of Remote Sensing. Habitat evaluation using ground based data and their spatial organization have been found to provide reliable informa tion on the biodiversity distribution pattern. The present approach for pri oritizing the biodiversity rich sites have the advantage of integrating spa tial, non-spatial information and horizontal relationships. The information is finally presented in space and time. This approach will facilitate cons ervation prioritization: systematic inventory and continuous monitoring. (C ) 2000 Elsevier Science Ltd. All rights reserved.