Customized models also became
more important with the realization that user-requirements
often determined which should be used (Matthews et
al., 2000). Awareness of the limitations of CSMs has increased
since their introduction. These include the difficulty
of providing input data, the stochastic nature of this input data
in a temporally and spatially continuous environment (Burrough,
1989a, 1989b), and the difficulty of representing complex
situations numerically (Passioura, 1973). Despite these
difficulties, there is evidence that CSMs can play an important
role in scientific research, decision support, and education
(Matthews et al., 2000).