Previous - Next
This is the old United Nations University website. Visit the new site at http://unu.edu
Abstract
Sensationalism
Localized
area of impact
Methodology
Objectives
Outcome
E. R. C Reynolds and F. B. Thompson
The case is made for an objective and quantitative assessment of the effects of vegetation on climate and its spatial extent. Reasons are given for focusing attention on changes in tropical vegetation. Previous studies have tended to be too restrictive in scale and sometimes unavoidably subjective. Improvements in experimental method and the collection of necessary data are discussed. In particular, computer models of atmospheric circulation appear to be the most feasible way of using the available data to simulate the climatic effects of vegetation change.
It is evident that there is a tendency to resort to overstatement and unbalanced reporting when presenting scientific theories and findings to the public through the media. The same can be true even of the scientist eager to impress the holders of the purse strings of research funds. The objective is to grasp the attention of the audience and make it so concerned about the matter that action is taken. Sooner or later the balance has to be restored by painstaking research and dissemination of the facts. Unfortunately, the earlier misinformation is often very difficult to dislodge.
This is the case with our subject. The informed public and the influential politician are convinced that felling or planting trees has a significant effect on the climate over a large area, if not globally Consequently, decisions to increase or decrease forest cover are supposed to implicate continental, or even world-wide, changes of climate. Indeed, it is only a small step to suppose that afforestation can be employed to ameliorate regional climate or that extensive deforestation will reduce rainfall at places far removed from the felling axe.
Claims that the global climate can be protected by a policy of forest conservation have become wedded to legitimate cases for conservation of the forest fauna and flora and the control of erosion and flooding by maintaining forest areas. Proper conservation arguments are pertinent to the world's remaining tropical rain forests and the scene for this is presented in chapter 2. In terms of climatic effects, through evaporation or fixation of atmospheric carbon dioxide, any tall woody vegetation might serve in place of the tropical rain forest. Thus, to distinguish between the effects on climate and other aspects of forest conservation it is important at this stage to establish what constitutes deforestation. An extensive but transient interference with the tree cover can have enormous and permanent effects on such factors as the flora, fauna, and the soil. In contrast, at current rates of forest clearance, unless there is no regrowth of shrubs and trees, the actual area without forest cover at any one time can hardly be expected to cause any climatic change even on a regional scale. The aim of this book is to approach the influence on the regional climate of vegetation change as quantitatively as possible. We hope that it will contribute an awareness of the magnitude of likely regional climatic changes resulting from land use policies.
Not only must we know the increase or decrease of the water balance components due to a change of vegetation, we must also know how extensive these effects may be. If the water balance is only materially altered within the confines of the area of changed vegetation, then the problem is greatly reduced. If, however, there are significant global effects, then the international community needs to be concerned in local land use policies. To underline the significance of the scale of the effects, we speak of "regional impacts" of the hydrological and climatological changes caused by vegetation change. The well-known effects on the microclimate and purely local water balances are adequately treated by others. At the other end of the scale, global climatic effects are unlikely to be significant if we are unable to demonstrate an effect at the regional scale.
It is inevitable that our considerations are concentrated mainly on extensive tropical deforestation and afforestation since there is a concentration of developing countries in these regions and a great potential for land use change. The exploitation of the forests of these regions to finance development and release land for agriculture and plantation crops is an attractive proposition in many ways. The fact that, relative to other regions, larger areas of the land surface may be altered hydrologically focuses attention on the tropics as being currently the most likely area to initiate regional or even global climate changes by land use change. The food demands of the increasing populations in tropical areas result in permanent land use changes following deforestation leading to changes in the radiation balance and evaporation. Additionally, if we wish to test the possibility that land use change affects the climate of distant areas, the nature of the general atmospheric circulation of the earth makes the tropical zone one of the more likely areas for such changes to lead to significant effects elsewhere. A more local effect of land use on climate in need of further study is that due to the internal convective circulation cells of tropical regions producing rainfall only some tens of kilometres from its evaporative source.
The subject has suffered more than most from uncontrolled observations leading to equivocal deductions. One problem is that large natural temporal variations of the rain climate (which have occurred since prehistoric times) are confounded with any climatic effects of land use changes. There is also the problem of land use itself affecting the systematic errors of rain gauges, especially in windy climates, so that corrections applied to the rain gauge catch may have to be changed with land use change (see chap. 5). Correlations of rainfall with the proportion of the area covered by forest are also suspect when there is no quantified value for land use change during the measurements since they may simply reflect the rainfall climates required by different vegetation types or land uses. Studies using critical experimental techniques have invariably been on a small scale due to the cost and the fact that large suitable sites (usually watersheds) that can be afforested or deforested for the experiment are simply not available. As a result, as we shall see (chap. 7), there is still a need for large scale studies using controlled methods to observe the effects of vegetation and land use changes to support other climatic research.
Historically it was probably due in part to the necessarily restricted scale of many hydrological and climatological investigations that two lines of thought evolved. One associated forests causally with greater regional precipitation and water yield, while the other denied this and even proposed a decrease in water yield. Over the last 20 years there has been some convergence of these conflicting points of view. There is now a wide area of agreement, in particular on the effects of forests on soil conservation, water quality, and the hydrograph. Many would agree that there are situations where a local or regional increase of rainfall and evaporation might be expected to follow afforestation. There is, however, still disagreement on both the magnitude and importance of these effects and on their likely locations; whether there would be an increase or decrease in the water yield following afforestation is even more in dispute. When the problem cannot be resolved by experiments, probably the next best approach is by using computer simulation models. These should be based on the physical systems involved, using as much real data as possible, and be tested against real catchment systems. This innovation for assessing the effect of land use change on climate has provided the thrust of this book. It is an objective way of combining our present knowledge to produce the best current predictions. Irrational outputs indicate our limitations in understanding the system and a lack of data. Sensitivity analyses show which parameters of the land surface, the atmosphere, and the radiation and water balances have insufficient precision.
Early general circulation models of the atmosphere lack relevance to our enquiry; they are intra- rather than interdisciplinary and too generalized to take into account the detailed effects brought about by vegetation changes. An interdisciplinary approach involving atmospheric physicists together with plant physiologists, hydrologists, and microclimatologists is needed to achieve the understanding and precision required to detect land use effects on regional climate. Scientists studying exchange processes at the earth's surface and those investigating general atmospheric circulation have been brought together by the workshop.
From the conclusions reached by the workshop (chap. 9), it is evident that the use of models to simulate the effect of vegetation change on regional climate is in its infancy. Very few situations have been simulated and those that have generally represent extreme changes. It is to be hoped that we are poised to assess to realistic degrees the significance to regional and local climate of man's manipulation of the earth's vegetative mantle.
Much depends on the reliability of the models and the constituent sub-models. Perhaps now is the time to test the models critically. This is the basis of the proposals for a mesoscale observatory study as discussed by Professor Dooge (chap. 7) and supported by the workshop (chap. 9).
So far we have considered simply the broad question of whether there is an effect of vegetation change and if so to what degree and to what extent. However, there are several points about the dependence of climate on the management of the surface vegetation that need to be understood to assist in the rational development of both national and international land use policies. This entails examination of two rather distinct phenomena. The first is the direct effect of vegetation change on the evaporation input to the general atmospheric circulation (see chaps. 7 and 8). The second is the less direct effect of deforestation increasing the carbon dioxide concentration of the atmosphere and so changing the radiation balance and, in turn, the air temperature, evaporation, and rainfall.
Although we have previously questioned the objectivity of the observational evidence attempting to relate regional climatic changes with deforestation or reforestation, which has been accumulating for the best part of a century, one of the tasks set for the workshop was to review and assess this evidence. In chapter 3 observations from tropical countries have been reviewed. Specific empirical investigations from India are reported by Meher-Homji in chapter 4, while Shiklomanov and Krestovsky give details of Russian studies in chapter 5. The reviewers were further asked to postulate what mechanisms could account for the observed climatic effects. Without an explanation that includes feasible hypothesized processes, it is difficult to pursue the matter further.
The evidence thus reviewed may not be definitive but is certainly indicative that there could be significant effects of forests on rainfall if not on the volume of runoff.
At the very least it indicates that we are right in looking for a different approach to resolve the arguments.
It has been possible for the workshop to agree to conclusions (chap. 9). They reflect the discussions of each of the review papers and are by nature compromises of the differing views of the participants (listed on page 212). These points of agreement have been discussed, drafted, discussed again by the workshop, redrafted and circulated to the members for comment and correction. Whilst they represent a consensus, the composition of the workshop will have been bound to produce some subjective bias.
Of the ten workshop conclusions, half related essentially to the investigative procedure. This emphasizes the need to improve our methodology before the precision is adequate to allow us to rely on the findings. The workshop presages a change in data collection away from small-scale observations to extensive methods. The scientific community is clearly on the lookout for new, more reliable tracers to use in large-scale hydrology. Throughout the field more rigorous attention to descriptions and scales is demanded; these need to be more relevant to the physical systems operating and the intended use of the information acquired.
The other conclusions incorporate as far as possible quantitative estimates of effects of changing forest cover. This is where greater precision is needed. On the larger to global scales, the workshop's conclusions are mostly qualitative, often indicating little more than whether changes will be positive or negative. However, there is optimism that improved simulation models will yield acceptable estimates of the magnitude of the effects.
It will be apparent that, although there are numerous uncertainties in the models that relate climate to vegetation via general atmospheric circulation, many of these arise from difficulties of parametrization and the quantitative values of inputs. Careful appraisal of the system and more reliable data should resolve these uncertainties. In one area the problems are rather a lack of proper understanding of the processes. These are the hydrological mechanisms occurring in the biosphere. The reason for our ignorance is the bewildering variety of ecophysiological adaptations directed towards the strategy of maintaining plant tissue moisture potentials to permit cellular metabolism and growth to proceed. The regulation of the distribution of plant roots and the resulting pattern of soil moisture extraction are as yet incompletely understood. The disposition, size, and behaviour of stomata controlling the rate of transpiration need further study. More information on the variations of albedo and the aerodynamic roughness of complete vegetation canopies is also required to provide a proper link with micrometeorology. Unless research into these biological phenomena is pursued, relationships between vegetation and climate will remain matters of empirical observation, and choices of land use to fulfil hydrological if not climatological objectives will be less certain.