Sensitivity and uncertainty analysis in spatial modelling based on GIS

Uncertainty analysis (UA) and Sensitivity analysis (SA) are prerequisites f or model building. UA aims to quantify the overall uncertainty associated w ith the model response as a result of uncertainties in the model input. SA studies how the variation in the model output can be apportioned to differe nt sources of variations. Both UA and SA have been successfully applied in many different fields. This paper describes the contribution that UA and SA can give in the field of GIS-based models. In the first part of the paper the fundamentals of UA and SA are introduced. The role UA and SA can play i n the spatial modelling based on GIS is then addressed. In the last part, a general procedure to perform the two kinds of analysis on GIS-based models is proposed. An example of analysis of a simple hydrologic model is descri bed. This research on GIS-based models, done at ISIS-JRC, provides effectiv e tools for both the assessment of the quality of the model output and the optimisation of the resources to be allocated for the implementation of GIS applications. (C) 2000 Elsevier Science B.V. All rights reserved.