Essa é uma revisão anterior do documento!
Tabela de conteúdos
GIS, Spatial Analysis, and Modeling
| David Maguire, Michael Batty, Michael Goodchild, editors | 2005 |
Transition Potential Modeling for Land Cover Change
Sérgio - goodLUCC
Approaches to Modelling in GIS: Spatial Representation and Temporal Dynamics
When dynamic predictive models are embedded within GIS, then the software in question is usually customized to the specific process in question. […] Softwares such as this [PCRaster], in fact, is conceived with physical rather than human process in mind.
In CA modelling, for example, regular cells are usually the storage bins for aggregations of individuals, […] This collection of agents act as an attribute of the cell rather than being a collection of agents in their own right. […] Diffusion is simulated by showing how cells states change rather by showing how agents actually move. CA models might be seen as ABM when the agents are fixed or immobile in the simulation sense.
Linking General-Purpose Dynamic Simulation Models with GIS
The software architecture required for ABM is significantly different from that required for just modeling the evolution of geographic entities in a static map.
Dynamic models typically produce two types of time sequences as results:
- longitudinal time histories of individual values
- snapshots of the entire system (movies)
Integration of Geographic Information Systems and Agent-Based Models of Land Use: Prospects and Challenges
This chapter summarizes the activities, concerns, and challenges of developers and end users of ABM/LUCC.
Many ABM/LUCC researches begin by developing models that operate over an abstract landscape, even when their longer-term goals include application to real-world landscapes.
Some interesting references about integration.
Types of integration:
- abstract cellular landscapes
- GIS initialization/output display
- integration via files written to disk
- run-time communication
- internal GIS functionality in ABM
A Wishlist for model integration:
| modelling functionality | object oriented structure. interface with other dynamic models. mobile objects. visualize in 3D |
|---|---|
| mathematical functionality | potentially needed for constructing agent decision-making, model calibration and coupling with environmental process. finite-element modelling, integer programming, genetic algorithms, simulated annealing |
| verification and validation | real-time visualization, on-the-fly sensitivity analysis by changing model parameters, generate and export graphics |
| build-in standard functions | well documented algorithms for all the above functionalities |
A call for a standard base modelling framework for ABM/LUCC.