Integrated studies of coupled human and natural systems reveal new and complex patterns and processes not evident when studied by social or natural scientists separately. Synthesis of six case studies from around the world shows that couplings between human and natural systems vary across space, time, and organizational units. They also exhibit nonlinear dynamics with thresholds, reciprocal feedback loops, time lags, resilience, heterogeneity, and surprises. Furthermore, past couplings have legacy effects on present conditions and future possibilities.

This paper presents an overview of multi-agent system models of land-use/cover change (MAS/LUCC models). This special class of LUCC models combines a cellular landscape model with agent-based representations of decisionmaking, integrating the two components through specification of interdependencies and feedbacks between agents and their environment. The authors review alternative LUCC modeling techniques and discuss the ways in which MAS/LUCC models may overcome some important limitations of existing techniques. We briefly review ongoing MAS/LUCC modeling efforts in four research areas. We discuss the potential strengths of MAS/LUCC models and suggest that these strengths guide researchers in assessing the appropriate choice of model for their particular research question. We find that MAS/LUCC models are particularly well suited for representing complex spatial interactions under heterogeneous conditions and for modeling decentralized, autonomous decision making. We discuss a range of possible roles for MAS/LUCC models, from abstract models designed to derive stylized hypotheses to empirically detailed simulation models appropriate for scenario and policy analysis. We also discuss the challenge of validation and verification for MAS/LUCC models. Finally, we outline important challenges and open research questions in this new field. We conclude that, while significant challenges exist, these models offer a promising new tool for researchers whose goal is to create fine-scale models of LUCC phenomena that focus on human-environment interactions.

Humans have continuously interacted with natural systems, resulting in the formation and development of coupled human and natural systems (CHANS). Recent studies reveal the complexity of organizational, spatial, and temporal couplings of CHANS. These couplings have evolved from direct to more indirect interactions, from adjacent to more distant linkages, from local to global scales, and from simple to complex patterns and processes. Untangling complexities, such as reciprocal effects and emergent properties, can lead to novel scientific discoveries and is essential to developing effective policies for ecological and socioeconomic sustainability. Opportunities for truly integrating various disciplines are emerging to address fundamental questions about CHANS and meet society's unprecedented challenges.

Agent-based simulation (ABS) is being increasingly used in environmental management. However, the efficient and effective use of ABS for environmental modelling is hindered by the fact that there is no fixed and clear definition of what an ABS is or even what an agent should be. Terminology has proliferated and definitions of agency have been drawn from an application area (Distributed Artificial Intelligence) which is not wholly relevant to the task of environmental simulation. This situation leaves modellers with little practical support for clearly identifying ABS techniques and how to implement them.This chapter is intended to provide an overview of agent-based simulation in environmental modelling so that modellers can link their requirements to the current state of the art in the techniques that are currently used to satisfy them. Terminology is clarified and then simplified to two key existing terms, agent-based modelling and multi-agent simulation, which represent subtly different approaches to ABS, reflected in their respective Artificial Life and Distributed Artificial Intelligence roots. A representative set of case studies are reviewed, from which a classification scheme is developed as a stepping-stone to developing a taxonomy. The taxonomy can then be used by modellers to match ABS techniques to their requirements. IntroductionAgent-based simulation (ABS) is being used in environmental modelling for many reasons that have already been discussed in the literature (e.g. Bousquet and Le Page, 2004;Ferber, 1999; Judson, 1994;Taylor and Jefferson, 1994). ABS provides a framework in which tractable techniques can be implemented that meet various requirements of environmental management modelling. First of all, ABS permits the coupling of environmental models to the social systems that are embedded in them, such that the roles of social interaction and adaptive, disaggregated (micro-level) human decision-making in environmental management can be modelled. It also permits the study of the interactions between different scales of decision-maker, as well as the investigation of the emergence of adaptive, collective responses to changing environments and environmental management policies.Whilst there are a number of practical problems associated with implementing ABS, such as dealing properly with floating point arithmetic and "ghosts in the model" (Polhill et al. 2006), we argue, however, that the efficient and effective application of ABS for environmental modelling is, at least for the beginner, more generally hindered by the fact that there is no fixed and clear definition of what an ABS is or even what an agent should be. Terminology has proliferated and definitions of agency have been drawn from an application area, Distributed Artificial Intelligence (DAI), which is not wholly relevant to the task of environmental simulation. This situation leaves modellers with little practical support for clearly identifying alternative ABS techniques and how to implement them. Those who are new to the field...

An agent-based model was developed as a tool designed to explore our understanding of spatial, social, and environmental issues related to land-use/cover change. The model focuses on a study site in a region of the Amazon frontier, characterized by the development of family farms on 100-ha lots arranged along the Transamazon highway and a series of side roads, west of Altamira, Brazil. The model simulates the land-use behaviour of farming households on the basis of a heuristic decisionmaking strategy that utilizes burn quality, subsistence requirements, household characteristics, and soil quality as key factors in the decisionmaking process. Farming households interact through a local labour pool. The effects of the land-use decisions made by households affect the land cover of their plots and ultimately that of the region. This paper describes this model, referred to as LUCITA, and presents preliminary results showing land-cover changes that compare well with observed land-use and land-cover changes in the region.