Dealing with soil variability: some insights from land degradation research in central Spain

Abstract Soil variability is often seen as problematic in land degradation studies in terms of sampling effort, data interpretation and for the extrapolation of results to other areas or time periods. Examples are given from land degradation research undertaken in central Spain which demonstrate some of these problems associated with soil variability. Geostatistics is presented as a useful tool for quantifying soil variability and in particular the variogram for interpreting and understanding spatial patterns. Soil variability is also seen to complicate the issues surrounding management strategies and makes monospecific management strategies less likely to be successful. In recent years several studies have suggested that variability in soil properties and vegetation cover may reduce the risk of land degradation by minimizing the spatial extent of runoff and erosion. These studies suggest that increasing soil variability may prove to be an effective strategy for reducing the runoff and erosion risk. This paper discusses these ideas and highlights the importance of increasing thresholds above which runoff and erosion occurs for the success of such strategies. This in particular applies to many semiarid environments where thresholds are deemed to be extremely low. Finally, these concepts are placed in the context of scale where soil variability may be viewed as existing at a multitude of nested levels varying from the micro‐ to the macro‐scale. Copyright © 2002 John Wiley & Sons, Ltd.