This chapter presents research carried out under the MASE project, including the definition of a conceptual model to characterize the behavior of individuals that interact in the dynamics of land-use and cover change. A computational tool for analyzing environmental scenarios of land change was developed, called MASE - Multi-Agent System for Environmental Simulation. MASE enables agent-based simulation scenarios and integrates the influence of socio-economic and political dynamics through the interaction of agents with rules of land-use and planning policies and the environmental physical and spatial variables. MASE simulator was extended to implement the Belief-Desire-Intention (BDI) model, called MASE-BDI. MASE and MASE-BDI are discussed including the conceptual model complexity and statistical techniques of map comparison to land change models. Two real cases of the Brazilian Cerrado validate quantitative and qualitative aspects of MASE and MASE-BDI simulators. Finally, the authors present some auto-tuning aspects of adjusting simulation parameters of MASE-BDI.
TopIntroduction
Models are crucial to represent complex and detailed reality since they can explicitly account for the state of knowledge, predict results, or act as objects of experiments. A model can represent real environmental patterns that embody essential and interesting aspects of the reality to be studied. Even simple conceptual models are critical components of biological disciplines, e.g., the earliest Lotka-Volterra model of species competition and predator-prey relationships (Lotka, 1925). Although this model is recognized as not being very realistic, it plays a role in organizing themes such as the dynamics of a hierarchically organized system, competitive exclusion, and the cycling behavior in predator-prey interactions.
Since all models are by definition incomplete, the central issue of modeling process is whether the fundamental aspects of the phenomenon are represented. Considering ecological phenomena, what is interesting and significant is usually a set of individuals and their relationships - from the interaction of two individuals to the behavior of a population in its environment. But the limited human comprehension of complex biological systems arises a problem when attempting to dissect a phenomenon into more easily understood components. This challenge is compounded by human’s current inability to understand relationships between the elements as they occur in reality. The real environment present multiple competing influences related to the broader context of time and space.
Models are used in land change science to better understand the dynamics of systems, to develop hypotheses to be tested empirically, and to make predictions or evaluate scenarios for use in assessment activities. The nature of the models, the selection of appropriate models and the nature of the abbreviations can be highlighted by the diversification of simulated models. In this sense, all modelers are using models whenever they use statistical tests or define a mathematical schema. The simulation of these models can help human comprehension of complex environmental systems. The land use and land cover change (LUCC) modeling strengthen the increasing need for tools able to capture the dynamics and outcomes of human actions through the use of individual-based and multi-agent models. Thus, LUCC dynamics are considered to remain a significant ecological challenge.
As cited by Brown et al. (2004), modeling is a major component of each of the three foci outlined in the science plan of the LUCC project (Turner II et al., 1995) of the International Geosphere- Biosphere Program and the International Human Dimensions Program.
- •
In Focus 1, on comparative land-use dynamics, models are used to help to improve our understanding of the dynamics of land-use that arise from human decision-making at all levels (households to nations). Surveys and decision maker interviews support these models.
- •
Focus 2 emphasizes the development of empirical diagnostic models based on aerial and satellite observations of spatial and temporal land cover dynamics.
- •
Finally, Focus 3 focuses specifically on the development of models of LUCC that can be used for prediction and scenario generation in the context of integrative assessments of global change.
This chapter presents the MASE (Multi-Agent System for Environmental simulation) project and focus on the importance of computational frameworks for modeling and simulating of environmental changes. The MASE project was developed in the Computer Science Department in association with the Biological Sciences Faculty of the University of Brasilia (UnB) during 2009-2015. The MASE project was defined considering Focus2 of LUCC project (Turner II et al., 1995) that emphasizes the development of empirical models based on satellite images that include observations of spatial and temporal land cover dynamics. The primary objective of MASE project is to define a conceptual model to characterize the behavior of individuals that interact in the dynamics of LUCC (Ralha et al., 2013).