Brimicombe: GIS, environmental modelling and engineering

Brimicombe defined engineering as "the need to design solutions more so in as much as those designs are often informed by computational or simulation modelling" (p. 4). He viewed geographic information systems (GIS) as tools "in the handling, integration, and analysis of spatial data" (p. 4). Given the static perception of nature that GIS provides, Brimicombe described the evolution of GIS to "geocomputation" (p. 37). Geocomputation, "spatial computation tools as a means of solving applied problems . . . and contributing to the development of theory" (p. 37), consists of sophisticated analytical applications, such as statistical modules, data mining, artificial intelligence, algorithms, and others. Together, environmental engineering and geocomputation allow simulation modeling and management of environmental phenomena and the construction of decision support systems by which policy makers can formulate environmental plans. Within his book, Brimicombe explained the nature of geocomputing and the data it contains, modeling from an environmental perspective, and the advantage gained by decision makers by combining the two.

Brimicombe characterized technology, such as GIS, as the application of science and syntax, operations, semantics, and uncertainty as the elements of geographic science.

Inductive or deductive modeling has many permutations--psychological, descriptive, normative, organizational, and exploratory. Likewise, models fall into various categories, according to Brimicombe, "analogues . . . hardware . . . mathematical . . . computational" (p. 55). Further classification can divide these groups into subsets, "static . . . dynamic . . . whitebox . . . greybox . . . blackbox . . . exploratory . . . prescriptive" (p. 56-57). Good modeling, Brimicombe contended, begins with a question, problem, or hypothesis of random, uniform, or clustered data.

Historical analogue, applying past events to "explain the present" (p. 55) and spatial analogue, "using events in one place to explain events in another" (p. 55) represent two types of analogue models. Hardware modeling attempts to replicate, on a small scale, a phenomenon "in order to study development, general behavior, changes in state, influence, or variables and so on" (p. 56). Mathematical models use its functions to solve singular (deterministic) or random, probabilistic (stochastic) situations. Computational models require computers "for symbolic manipulation using code and data in order to express phenomena and their workings" (p. 56).

Modeling topography stems from Tobler's first law of geography, continued Brimicombe, that is, "everything is related to everything else, but near things are more related than distant things" (p. 64). Devices to measure distances include IDW, inverse distance weighted" (p. 64).

In addition to space and time, models should reflect simplicity and completeness--conforming to William Occam's admonition, "it is vain to do with more what can be done with fewer" (p. 70) and Einstein's dictum, "the best explanation is as simple as possible, but no simpler" (p. 70).

The criterium sought for a model is that it as closely as possible represents reality. Interpolation and extrapolation extends the analytical range of models. Considerations of scalability and aggregation determine a model's flexibility. The ASCE Task Committee on GIS Modules and Distributed Models of the Watershed (1999) offers a review and evaluation of models.

Stakeholders, scientists, and policy makers employ GIS and models for the purpose of making sound environmental decisions. Decision support systems (DSS), according to Brimicombe, contains "database management system for accessing internal and external data, information and knowledge; modelling functions; user interface designs for interactive queries, graphical display and reporting" (p. 271). Spatial decision support systems (SDSS) adds to the power of the DSS; it will "allow handling of spatial data; allow representation of spatial relations (e.g. topology); include spatial analysis techniques (e.g. buffering, overlay); provide visualisation of spatial data" (p. 272). Combining GIS with environmental simulation models creates the basic components of a SDSS. What if scenarios, in the form of simulations, and feedback loops benefit environmentalists, economists, and stakeholders in communicating, collaborating, visualizing, and finding solutions to environmental problems.



Brimicombe, A. (2003). GIS, environmental modelling and engineering. New York: Taylor & Francis.