|
|
|
||||||||||||
|
Global Change Effects
on Landscape and Regional Patterns of Plant
Diversity
Table of Contents A fine-scaled predictive
model for changes in species distribution patterns of high
moutain plants induced by climate warming Using an artificial neural
network to characterise the relative suitability of
environments for forest types in a complex tropical
vegetation mosaic Climate change in
conservation areas of South Africa and its potential impact
on floristic composition: a first assessment Monitoring temporal changes
in the spatial patterns of a Mediterranean shrubland by
Landsat TM images Feature description The fact that global and regional plant distributions are affected by climatic factors has long been recognised by ecologists (Holdridge 1947, Begon et al. 1986 pp. 53-63). Since the eighties the topic of plant climatic distributions and their changes has received revived interest stimulated by projections of global climatic change. A large amount of scientific activity has in particular been stimulated by international research programs such as the International Geosphere Biosphere Program (IGBP), and has resulted in significant progress in our capacity to understand and predict past, present, and future plant distributions (e.g. Cramer & Field 1999). Important results from this research were presented at the GCTE - LUCC Science Conference in Barcelona (Spain) in March 1998. Among these, a number of studies have focussed on analysing and projecting regional distributions of plant species and communities from climatic and terrain factors. In this issue, four papers are presented to illustrate some of the main approaches, which can be applied. A common underlying assumption is that plant present distributions reflect their climatic limitations, which would also apply under future conditions. Though this assumption ignores the potential differences between fundamental and realized niches (Rutherford et al. 1995), it is considered as operational over regional scales where topographical diversity is expected to offer refugia for species with low competitive abilities, or presently at the limits of their distributions (Gottfried et al., Rutherford et al.). Studies differ in whether the niche is seen as a species-specific characteristic where, following Huntley et al. (1995), climate response surfaces are derived for individual species distributions (Gottfried et al., Rutherford et al.). Alternatively, vegetation types can be considered as entities that will move as structural units, though their exact botanical composition may vary over time (Hilbert & van den Muyzenberg). Studies have taken a diversity of approaches to identify climatic niches and extrapolate them to predict future shifts in plant distributions. The most widespread approach consists in regression based techniques which can identify one or several climatic parameters providing the best fit for actual distributions. This approach can be used over large regional data bases, as presented by Rutherford et al. for conservation areas of arid South Africa, or over more focused areas where detailed botanical and microclimatic information can be recorded in the field, such as done by Gottfried et al. in the Austrian Alps. Satellite imagery offers a valuable intermediate level of detail, where, once technical difficulties have been overcome, vegetation and environmental parameters can be derived simultaneously over large landscapes such as the Betica mountains of eastern Spain (Viedma & Meliá). Large regional data bases can also be put to useful contribution with a completely different technique : artificial neural networks (Hilbert & van den Muyzenberg). This technique appears quite robust, as it makes it possible to override some of the heavy statistical assumptions underlying regression approaches. Results from these studies, and their discussed limitations, highlight sensitive points about our power to predict plant distributions under future climatic conditions. First, studies need to identify critical climatic parameters which are limiting in a specific study area. For example, Rutherford et al. identified a control of growing days by available moisture as critical to determine the effects of a temperature increase in arid South Africa. In contrast, altitudinal gradients in mean daily temperature were used to predict high mountain plant shifts in the Austrian Alps. Second, as shown by those studies which were able to explicitly address the effects of terrain (Gottfried et al., Hilbert and van den Muyzenberg, Viedma & Meliá), topographic complexity is an essential component which needs to be taken into account. Altitudinal gradients, aspect, and specific landforms (e.g. narrow gullies) can play a crucial role in the persistence of some species. With these parameters in hand, robust predictions of present species distributions can be derived and are applied as predictors of future (and possibly also past) distributions. However, several important limitations remain to be addressed. First, interspecific interactions and the potential limitation of some species by competition or predation can not be accounted for by these methods. This could in particular impose shortcomings to predictions of shifts of communities as a whole. A second, perhaps more problematic limitation regards the ability for species to migrate rapidly enough to track climate, especially over landscapes fragmented by human land uses (Pitelka et al. 1997). This issue is especially acute when addressing the case of locally rare species (Gottfried et al.), or for the design of climate change proof reserve networks (Rutherford et al.). Simulation modelling approaches which take interspecific interactions and species dispersal capacities explicitly into account will need to be included, along with response surface approaches, as part of the tool box for the prediction of future vegetation distribution. References M Begon, JL Harper & CR Townsend (1986) Ecology - Individuals, populations and communities. Blackwell Scientific Publications, Oxford. W Cramer and CB Field, editors (1999) Potsdam NPP Model Intercomparison Workshop. Global Change Biology 5 (Suppl. 1). M Gottfried, H Pauli, K Reifer and G Grabherr (1999) A fine-scaled predictive model for changes in species distribution patterns of hiogh moutain plants induced by climate warming. Diversity and Distributions, this issue DW Hilbert and J van den Muyzenberg (1999) Using an artificial neural network to characterise the relative suitability of environments for forest types in a complex tropical vegetation mosaic. Diversity and Distributions, this issue LR Holdridge (1947) Determinaion of world plant formations from simple climatic data. Science, 105, 367-368. L Pitelka and the Plant Migration Workshop (1997) Plant migration and climatic change. American Scientist . 85, 464-473 MC Rutherford, M O'Callaghan, JL Hurford, LW Powrie, RE Schulze, RP Kunz, GW Davis, MT Hoffman and F Mack (1995) Realized niche spaces and functional types: a framework for prediction of compositional change. Journal of Biogeography 22, 523-531. MC Rutherford, LW Powrie, and RE Schulze (1999) Climate change in conservation areas of South Africa and its potential impact on floristic composition: a first assessment. Diversity and Distributions, this issue O Viedma and J Meliá (1999) Monitoring temporal changes in the spatial patterns of a Mediterranean shrubland by Landsat TM images. Diversity and Distributions, this issue |
||||||||||||
|
|
|||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
