Biodiversity and agriculture in dynamic landscapes: Integrating ground and remotely-sensed baseline surveys.

Sustainable biodiversity and land management require a cost-effective means of forecasting landscape response to environmental change. Conventional species-based, regional biodiversity assessments are rarely adequate for policy planning and decision making. We show how new ground and remotely-sensed survey methods can be coordinated to help elucidate and predict relationships between biodiversity, land use and soil properties along complex biophysical gradients that typify many similar landscapes worldwide. In the lower Zambezi valley, Mozambique we used environmental, gradient-directed transects (gradsects) to sample vascular plant species, plant functional types, vegetation structure, soil properties and land-use characteristics. Soil fertility indices were derived using novel multidimensional scaling of soil properties. To facilitate spatial analysis, we applied a probabilistic remote sensing approach, analyzing Landsat 7 satellite imagery to map photosynthetically active and inactive vegetation and bare soil along each gradsect. Despite the relatively low sample number, we found highly significant correlations between single and combined sets of specific plant, soil and remotely sensed variables that permitted testable spatial projections of biodiversity and soil fertility across the regional land-use mosaic. This integrative and rapid approach provides a low-cost, high-return and readily transferable methodology that permits the ready identification of testable biodiversity indicators for adaptive management of biodiversity and potential agricultural productivity.

The impact of selective logging and forest conversion for subsistence agriculture and pastures on terrestrial nutrient dynamics in the Amazon

The conversion of primary forests in the Amazon via selective logging, slash and burn agriculture, and pasture establishment, will continue because of the increasing demand for timber and agricultural land. Forest conversion has major impacts on the stocks and flows of essential plant nurtrients. For example, the burning of slashed vegetation results in most of the carbon and nutrients in the biomass being volatilized during burning. Nutrient stocks are further depleted by the nutrient exports in timber and crop harvests. Poor crop and pasture mamagement practices can result in rapid declines in soil fertility and reduced vegetation regrowth potential. These changes can influence the rate of secondary forest regrowth and carbon sequestration on abandoned crop and pasture lands. The dominant land use practives following forest conversion in the Amazon are identified and the data on the impact of selective logging, subsistence cropping, and extensive pastures, on nutrient stocks and budgets in terrestrial Amazonian ecosystems are reviewed.