Water and Carbon Fluxes in Ecosystems

Abstract

One description of the aim of science is to achieve a sufficient understanding of the functioning of a system to be able to make predictions about the response of that system to a stimulus or perturbation. We may study and analyse ecological systems for many particular reasons and it is widely believed that this is both an intellectually stimulating activity and the way to the solution of practical problems. The question that I wish to address is whether we can expect ever to be able to understand an ecological system well enough to make predictions that are useful in the exacting con- text of ecosystem management. Is the understanding that we can achieve limited only by resources, or are there other more fondamental reasons why we may never be able to understand ecosystem functioning adequately? Shall we ever, for example, be in a position to predict the likely consequences of a major environmental perturbation? The difficulties involved in the retrospective analysis of the causes of forest decline are brought up repeatedly elsewhere in this volume, but could we have predicted the now evident resuit? In similar vein, there is much discussion at the present time about the likely consequences of a doubling in the concentration of atmospheric carbon dioxide for vegetation and climate. What are the prospects of ever being able to make useful predictions about the consequences of such a doubling for, say, a tropical, or even a simpler temperate, forest ecosystem, let alone making them now, in our present state of knowledge, before it actually happens? Our fondamental difficulty stems from the lack of empirical knowledge about the functioning of ecosystems. We may readily acquire empirical information on which we can base hypotheses, at the organisational level of cell, leaf or plant, but these difficulties increase at larger scales. Whilst we may legitimately obtain this knowledge by experimentation at the “micro-ecosystem” level of say a flower head (e. g. Bertsch, Part 3-H or Zwôlfer, Part 3-A) or the “mini-ecosystem” level of say a pond or small lake (e. g. Likens 1985), we are prevented by major practical problems and very substantial ethical difficulties from obtaining information by experiment at the scale of what is more generally thought of as an ecosystem.