Long-term dynamics in arid and semiarid ecosystems - synthesis of a workshop

Abstract

According to the United Nations (UNEP, Agenda 21), approximately one sixth of the world’s population, 70% of all drylands with a total area of 3.6 billion hectares, and a quarter of the total land surface of the earth is endangered by desertification. The most obvious consequences are increasing poverty and the growing damage to 3.3 billion hectares of rangeland. These impressive figures illustrate the importance of developing mechanisms for combating desertification. However, there is an increasing number of scientists who doubt the accuracy of these figures and who speak of ‘the myth of desertification’ as a publicity tool (e.g., Thomas & Middleton 1994). Their criticism is based on the idea that dryland ecosystems might after all be well adapted to disturbances and may exhibit good recovery characteristics. Nevertheless, it has been increasingly recognized that arid and semiarid ecosystems often exhibit complex non-equilibrium dynamics involving complicated nonlinear processes and stochastic event-driven behavior (Westoby et al. 1989; Walker 1993). Vegetation changes generally occur unpredictably in the short term (years) in response to rainfall, and episodically in the long term (several decades) in response to rare events, or due to grazing pressure, climatic change, altered disturbance regimes, or a combination of these factors. The complex dynamics of arid and semiarid ecosystems and especially the mismatch between observation times (years) and time scales of vegetation change (centuries) make it difficult to fully understand their long-term dynamics (Wiegand et al. 1995; Jeltsch et al. 1996). Hence, it is extremely difficult to establish whether an area is undergoing progressive, long-term decline in biodiversity and productivity (desertification), or whether it is merely suffering shortterm drought, from which the land may recover if human impact is reduced or eliminated. But for developing management strategies for sustained animal production and species conservation, it is necessary to know how resilient arid and semiarid ecosystems really are, and to assess their potential to recover from serious natural or human disturbances or stress. Long-term ecological monitoring and interdisciplinary approaches to interaction between rainfall, geology and ecology are required in order to gain a better understanding of long-term dynamics in arid and semiarid systems. These approaches include both field studies and modelling studies. The latter can help to extrapolate local information and knowledge gathered on shorter time scales to larger spatial and temporal scales. This knowledge originating from field and modeling studies forms an indispensable basis for identifying key factors and processes that shape the potential for both desertification and recovery therefrom. In the following we briefly summarize and link various papers (all this issue) on field and modeling studies that are based on contributions to the INTECOL symposium ‘Long-term dynamics in arid and semi-arid ecosystems’ in Florence, 1998, or that were inspired through discussions during and after the symposium