Environmental Modeling and Assessment 6: 111–121, 2001.  2001 Kluwer Academic Publishers. Printed in the Netherlands. The need for scale sensitive approaches in spatially explicit land use change modeling A. Veldkamp, P.H. Verburg, K. Kok, G.H.J. de Koning, J. Priess and A.R. Bergsma Laboratory for Soil Science and Geology, Wageningen University, P.O. Box 37, 6700 AA Wageningen, The Netherlands E-mail: Tom.Veldkamp@geomin.beng.wau.nl; http://www.gis.wau.nl/∼clue This paper introduces some of the issues that are relevant to the spatially explicit modeling of land use systems. A short overview is given of the ways and means in which a number of different land use change models describe the land use system. Specific at- tention is given to the empirical modeling approach used in the CLUE (Conversion of Land Use and its Effects) modeling frame- work. This approach is demonstrated for three case studies in China, Ecuador and the Atlantic Zone of Costa Rica. These case- studies illustrate the methodology for multi-scale analysis of land use driving factors and their application in spatially explicit mod- eling exercises. Model functioning, performance and limitations are discussed. The presented case-studies indicate that empirical modeling results can contribute to a better theoretic imbedding of land use change research in scale sensitive and integrated theo- ries. Keywords: driving factors, interdisciplinary, land use change, modeling, scale 1. Introduction In general, land use is determined by the interaction in space and time of biophysical factors like soils, climate and topography, and human factors, like population, technology, economic conditions and planning by institutions [1–3]. In- terpretations of how such multi-dimensional land use/cover driving forces act and interact are still controversial, espe- cially with respect to the assessment of their relative impor- tance in specific cases. Insights in the interactions of land use and its driving factors can be gained by studying land use change in comparative studies. Unfortunately, relative few regional comparative studies have addressed the role of these combined driving forces in land use systems in a spa- tially and temporally explicit way. There is currently not a single theory that can be used to study the complexity of the land use system in all its aspects. Because land use change processes operate at the interface of social and natural systems, theories developed in the so- cial and natural sciences can only provide useful insights in the mechanisms of some driving forces. Unfortunately these theories are either only valid for one of the processes de- termining land use change or lack an explicit focus on the spatial dimensions. A common phenomenon in spatially explicit analysis of complex systems, is the fact that the scale at which an analy- sis is conducted affects the type of explanation given to the observed phenomena [4–6]. While pests and diseases might cause variation within a rice field, climate systems shape broad agro-ecological zones that determine the bio- physical constraints for rice cultivation. At coarse (aggre- gated) scales, the high level of aggregation of data obscures the local variability but can show patterns invisible at de- tailed scales, and vice versa. Furthermore, factors deter- mining land use (change) can operate at large spatial dis- tance from the area affected. Thus, for dealing with the issues of land use/cover change, it is necessary to use a multi-scale approach that identifies and quantifies land use driving forces and their interrelationships at various spatial scales [7]. Present studies of land use change are mostly limited to a single scale, depending on the disciplinary tradition of the researcher. Researchers in the field of the social sci- ences have a long tradition in studying individual behavior in human–environment interactions at the micro-scale, some of them using qualitative approaches [8,9] and others using the quantitative models of micro-economics and social psy- chology. Rooted in the natural sciences rather then the so- cial, geographers and ecologists have focussed on land cover and land use at the macro-scale, spatially explicated through remote sensing and GIS, and using macro-properties of so- cial organization in order to identify social factors connected to the macro-scale patterns [10,11]. Direct up-scaling of processes and qualitative understanding of land use driving factors obtained at the micro-scale to higher aggregation lev- els is not possible, because these relations are often subject to scale dependencies. This scale dependency can be caused by the non-linearity of the studied relation [12], feedbacks within the system and interactions at the micro-scale. Hu- man behavior varies with group size, for example, groups of people will make riskier decisions than individuals, so- cial conformity increases with group size; and inhibitions decrease with group size. Thus, it is not possible to de- rive the functioning of the system by solely studying the behavior of its micro-scale components. Theories that ex- plicitly deal with behavior at higher levels of organization (e.g., collective action theory [13]) are needed to understand the processes and driving factors of land use change at the regional or national scale.