Integrating landscape processes: LAMOS, a Landscape MOdelling Shell

Sandra Lavorel ^{1, 2}, Ian D. Davies ², Ian R. Noble ²

¹ Centre d'Ecologie Fonctionnelle et Evolutive, CNRS UPR 9056, 1919 route de Mende, 34293 Montpellier Cedex 5, France
² Ecosystem Dynamics Group, Research School of Biological Sciences, Australian National University, Canberra ACT 0200, Australia

Context and purpose

Recent work has shown that the behaviour of landscapes with spatially linked processes such as dispersal, lateral flow of resources etc, is quite different from the behaviour of the same plants and animals represented as point models (e.g. Noble & Gitay 1996, Rupp et al. in press). Feedbacks, such as those between disturbance regimes and community composition, can lead to persistent, self-generated patterns that, in turn, change the overall composition and dynamics of the landscape communities. Many of these interactions can only be explored via modelling. The last 10 years have seen a proliferation of landscape simulation models, facilitated by the ever-increasing availability of computing power. However, this activity has been mainly uncoordinated, as it has developed in the absence of a general landscape theory, or at least of well-recognised benchmarks. This has in particular been the case for landscape fire models (Gardner et al. 1999).

LAMOS is a landscape modelling environment, designed to allow users to explore the role of different processes in the dynamics of landscapes with the minimum of programming effort. LAMOS was specifically designed to:

• Carry out formal comparisons of existing models: examining sensitivity to processes, to modelling methods, and to parametrization;
• Assemble new models from an existing toolbox by mixing and matching available methods for each process;
• Create new models by assembling some pre-coded methods for some processes with new methods for specific modules; for example fire modellers may wish to include specific methods for fire propogation while not having to re-invent a method for succession modelling.

Using LAMOS for comparative analysis of landscape models

LAMOS has been tailored to meet specific objectives for research on global change effects at the landscape scale. The challenges are:

• To understand and predict future disturbance regimes and their interactions with landscape vegetation patterns. Fire has been recognised as one of the most sensitive natural disturbances to climate, atmospheric, land use and biodiversity changes, and will receive priority attention.
• To examine migration of plant species in response to climate change; in the context of landscapes fragmented by human land use and increasing disturbed habitats.
• To assess changes in biogeochemical fluxes and feedbacks to the atmosphere.

To answer these challenges, landscape researchers involved in GCTE have recognised the need for a formal comparison exercise of landscape models, specifically landscape fire models and seed dispersal models. The aims of the comparison exercises are:

• To understand how much detail in modelling methods and data is required to account for:
  • spread and effects of disturbances (e.g. fire)
  • seed rain, including rare long-distance dispersal events
• To carry out detailed analyses of the interactions of disturbance and dispersal with landscape spatial pattern.

LAMOS is proposed as one of the tools to support these exercises.

The knowledge base and the models derived from the comparison exercises will be applied to run simulations for scenarios of land use and climatic change in target areas representative of the range of global situations.