Geography Reference
In-Depth Information
combine rice paddies represented on the historic and the modern maps. It was then
possible to identify the common areas of rice paddies on the two maps and to
ascertain which areas no longer contain rice paddies.
Combining dif erent data layers may be problematic where polygon boundaries in
the input layers are supposed to be identical but in fact dif er. h e end result is termed
'spurious (or slither) polygons'—(usually) small polygons that represent the dif erence
between the two sets of boundaries. Various approaches exist for removing such spu-
rious polygons. If there is greater coni dence in the accuracy of one set of boundaries
than the other then the boundaries with greater accuracy may be retained in preference
to the other less accurate boundaries.
Multicriteria decision analysis
5.3
GIS allow the integration of diverse data sources and facilitate the exploration of the
relationships between them. At a simple level, it may be desired to identify a set of
areas that fuli l several dozen criteria (e.g. areas more than a given distance from some
feature, with gradual slopes, a particular soil type, and so on). Earlier sections intro-
duced tools, such as buf ers (Section 4.5) and overlay operators (previous section),
which can be used to help address such issues. h is section expands on some key top-
ics that relate to considering multiple criteria in this way. h e term 'GIS-based multi-
criteria decision analysis' (GIS-MCDA) encapsulates the processes involved in using
GIS to help decision making using multiple data sources. Malczewski (2006) provides
a simple dei nition of GIS-MCDA as 'a process that transforms and combines geograph-
ical data and value judgements (the decision-maker's preferences) to obtain informa-
tion for decision making' (p. 703). With GIS-MCDA, the relative importance of dif erent
criteria can be taken into account. For example, if, in some planning process, accessi-
bility by road is more important than the slope of the terrain then proximity to roads
may be given a larger weight than slope as a criterion in the decision-making process.
A means of selecting particular alternatives from a set of available options is a deci-
sion rule (Malczewski, 2006). h e weighted summation decision rule approach and
similar approaches are the most commonly applied in the literature (Malczewski,
2006). Malczewski (1999) outlines one simple additive weighting method. Using this
approach each criterion layer is standardized (e.g. by dividing each value in a layer by
its maximum value). h is is necessary to enable comparison of like with like as direct
comparison of, say, distance from roads with slope of terrain would be meaningless—
standardization means that all units will be comparable (using the approach suggested
above they will all range from 0 to 1; see Heywood
et al
. (2006), pp. 239-240, for
another example). Next, weights are determined (this may be quite a subjective proce-
dure). If a layer is to be assigned 40% of the weight this can be expressed as 0.4,
with the weights for the other layers totalling 0.6. h en, each of the standardized map
layers is multiplied by the weights and the weighted standardized maps are added
together. h e optimal alternative (e.g. most suitable area or areas for development) is
Search WWH ::
Custom Search