A novel application of Sensitivity Analysis is presented. Useful applications of Global SA (GSA) already exist in the field of numerical modelling. In this paper, we explore the joint use of GSA, Geographical Information Systems (GIS) and Multi‐Criteria Evaluation. In this preliminary case study, 11 factors have been used to find the best place to locate a hazardous waste landfill. Two variance‐based methods (Sobol' and E‐FAST) are used to compute sensitivity indices in order to identify the factors that determine the variance of the model output. The results show that only three factors jointly account for 97% of the output variance. This information is employed to make a simplification of the original model, retaining only these three influential factors. In addition, if the SA is carried out in a pilot area where the spatial properties are similar to those of the whole region, we can infer the results to the whole area. This procedure achieves the goal of the study with an optimized allocation of resources for GIS data acquisition.

The aim of this article is to review how sensitivity analysis has been applied to models based on Geographical Information Systems (GIS) and Multicriteria Evaluation techniques (MCE). This kind of analysis is conceived as a stage in the model evaluation that examines the extent of output variation of a model when parameters are systematically varied over a range of interest. Twenty-eight studies related to land planning processes, environmental management, and location of noxious facilities have been examined. This review reveals that sensitivity analysis is a) not a common practice, b) is more widely carried out in location of noxious facilities, and c) that the analysis most frequently used is based on the variation of the weights of the factors implied in the process to test whether it significantly modifies the results obtained.

Over the last few years, urban expansion has occurred without much control or urban planning, causing both regional and global impacts on the natural ecosystems. This situation has spurred reflection on current planning policies and has also stimulated research into new tools and methodologies for the simulation of alternative territorial scenarios for the future. In this respect, the use of multicriteria evaluation techniques within a GIS environment has proved to be an efficient option for an optimal allocation of land uses. However, those techniques have not been explored thoroughly in the field of future scenario simulation. The aim of our research is to show and to reinforce the possibilities of these techniques in developing three different scenarios of urban growth (business as usual, crisis, and innovation and sustainability). All three scenarios discriminate between three land uses (residential, commercial, and industrial), and have been developed for the Madrid region. The base-line year of these scenarios is 2000 and the time horizon is 2020. In that process, fourteen spatial factors, which encompass environmental, economic, and social issues, have been considered. The results obtained show that some discrepancies exist between the business as usual scenario and European policies for transport and the environment. This situation has an impact on the territory and causes irreversible damage to the environment. The crisis scenario, in contrast, alleviates those pressures, if not because of the application of environmental policies to the planning process, at least because the situation of crisis causes the metropolitan model to change. Finally, innovation has been considered a driver of economic development and good planning practice, the positive effects of which have been highlighted in the third scenario.