干旱区生态环境知识资源中心

Several approaches designed to enhance the understanding of the long-term dynamics of arid and semiarid ecosystems were proposed by botanists, zoologists, ecologists, geographers, and modelers from Argentina, Australia, Germany, Israel, Namibia, South Africa and the USA. Methods and approaches ranged from (1) ecological monitoring with traditional plot methods on transects through climatic or geological gradients, and more recent remote sensing methods, (2) the use of ecological archives, such as laminated sediments, fluviolacustrine deposits, or the colour banding pattern among the leaf bases of Australian grasstrees caused by fire, (3) modeling and spatial pattern analysis to (4) experiments with grazing enclosures. The focus of the studies varied from investigations of local (vegetation) dynamics, biological crusts, rodent and tenebrionod beetle communities; regional-scale water and nutrient dynamics to ecosystem dynamics and considered anthropogenic forces, land use, grazing, disturbance, fire and climate as the main factors of impact. Depending on the biological characteristics of the systems studied, the expression "long-term" required consideration of time spans < 10 yr., decades, centuries, or even historical time spans. The contributions to the symposium outlined the need for more long-term data to facilitate an understanding of the complex dynamics of arid and semiarid ecosystems, which is especially hindered by the mismatch between observation times and the inherent time scales of ecosystem dynamics. However, the complexity of these systems and the urgent thread of degradation requires sophisticated monitoring methods which need to be accompanied by experimental and modeling approaches. Particularly in mismanaged and/or over-explored rangeland areas, the local and scientific knowledge derived from long-term monitoring and experimental studies should be compiled and organized for use in restoration ecology. Although modeling can help to identify gaps in the current knowledge and to guide future field work, it is also a powerful tool for interpolating short-term knowledge about the driving processes to longer time scales and is able to explore the logical consequences of short-time interactions between the relevant factors on longer time scales.