Humanity's experiment in exploring the resilience of the Earth System to large scale perturbations (e.g., modification of atmospheric composition or fragmentation of terrestrial vegetation cover) is now in full gear. First results of this experiment -- called Global Change -- have already been achieved, in particular, statistical evidence for anthropogenic global warming and the erosion of biodiversity. A full account of the climatic aspects is provided by the forthcoming IPCC Report [1]; the present state of the biosphere is described, for instance, in [2]).
Unfortunately, the investigators here are probing their own material support system, which cannot simply be replaced by a new specimen once the previous one has been worn out. The only way out this dilemma seems to be in modelling and computer simulation based on extensive non-destructive measuring and monitoring campaigns. The so-emerging virtual Earth Systems can be scrutinized safely in order to give deeper insights into the interactions of the various constituents and so avoid irreversible dead-end streets for the evolution of the original planet.
Big international research programmes like WCRP [3] and IGBP [4] are considered major stepping stones on our way towards such a fully-fledged Earth-System analysis. As a matter of fact, detailed 3D analogical models of the coupled geosphere-biosphere dynamics are supposed to transpire from these and related efforts within the next decade.
At present such models are not yet available, but even when they exist they may be almost as difficult to handle as their real counterpart. Therefore it seems to be wise to keep on studying also caricatures of the Earth System in the form of conceptual or even tutorial models (the different approaches to Earth-System modelling are discussed, e.g., in [5]). Such a strategy is well suited for uncovering qualitative phenomena and for exploring at any rate the topology, though not perhaps the geometry, of the system's phase space. The caricature may specifically reveal pertinent traits of the complex object under investigation and shatter conventional folklore based on conviction rather than analysis.
This modelling philosophy is further supported by recent insights of non-linear physics [6, 7]. They suggest looking for generic dynamical patterns of complex systems rather than striving for numerical predictions of details, which generally exhibit exponential sensitivity to conditional and computational errors. The identification of the major feedback mechanisms regulating or destabilizing the system in question is a crucial element of such a semi-quantitative analysis.
The evolution of the global ecosphere through billions of years was governed by the interaction of the biosphere and its geophysical environment as defined by the main factors insolation, plate tectonics, and the state of the atmosphere-hydrosphere system. A full explanation and reconstruction of the quaternary glaciation episodes, for example, seems to demand a thorough understanding of the planetary biogeochemical cycles as mediated by Life. The scientific paradigm behind these theories has been pioneered by J. Lovelock and his geophysiological approach to Earth-System analysis [8, 9].
A particularly useful ansatz for the investigation of geosphere-biosphere feed-backs is the Lovelock-Watson model (LWM) of ``Daisyworld'' [10, 11]. Despite its toy character, this model investigates possible mechanisms of environment stabilization through evolutionary adaption of terrestrial vegetation to varying insolation. We present here a 2D cellular automaton (CA) version of the original LWM, which takes into account a number of physical (e.g., lateral heat flow) and biological (e.g., competition and mutation) processes reflecting the dynamics within the real Earth System.
Our study has two main objectives, namely
Our material is organized as follows: in Section 2 we briefly review the original LWM. In Section 3 the extended model, which allows for infinitely many coexisting species, is introduced and discussed. The results of the quantitative analysis are presented in Section 4. In Section 5 the impacts of habitat fragmentation, i.e., the effects of heterogeneity, are analysed. The lessons to be learned from our geophysiological approach are summarized in the concluding section.