Martin Paegelow

One of the major challenges that populations have to face is vulnerability to water: lack of water, flooding, pollution, hazard sensitivity and coping capacity. Owing to economic, social and environmental inequalities, this paper focuses on water-related vulnerabilities in the Metropolitan Region of Santiago (RMS) in Chile. Our main objective is to understand, through mapping, the multiscalar logics of water vulnerability. This study is carried out at two scales: at the regional level (RMS), we proceed with open access municipal statistical data and maps, while at the local level, a more detailed analysis focusing on the Chacabuco Province is based on the same type of data but with either a higher spatial resolution (census districts) or a spatially more intensive data processing in order to take into account intra-municipal differences. In this way, we put into perspective the discourse developed in the Chilean media and by the inhabitants of Chacabuco Province regarding the province as an environmental ‘sacrifice zone’ for the RMS. The vulnerability maps are carried out at different scales in a simple and reproducible way by multi-criteria evaluation (MCE). The results confirm the hypothesis of a sacrifice zone and show that high-resolution data and adequate data processing give, on average, lower vulnerability scores than using only statistical data on the municipal level. The results provide a cartographic decision support for stakeholders. Limitations of the study are discussed and required further research is pointed out.

This study focuses on various validation and error analysis techniques that are based on map comparisons. After a literature review, authors apply these techniques to analyze the accuracy of LUCC models in terms of quantity, pixel by pixel correctness and LUCC components such as persistence and change. In addition, the fidelity of the spatial patterns and the congruency of the simulation maps from different modelling tools are tested. Finally, an error analysis is conducted that focuses on the magnitude of allocation errors and the magnitude of errors in predicted land use / cover classes. In addition, the impact of training dates on Markov chain predicted LUCC is analyzed.Mentioned techniques of validation and error analysis are illustrated by modelling LUCC of a small study area in the Eastern Pyrenees (France), where current LUCC are driven by spontaneous reforestation, decreasing pastureland and minimal anthropogenic disturbance. This very simple data set is used with three different tools (CA-Markov, LCM and Dinamica Ego) that represent commonly used modelling approaches and there methodological characteristics are highlighted. Applied to this specific dataset, contrasting results occur for different software programs that can help users choose an appropriate modelling approach according to specific model objectives.

La revue est publiée par la Marburger Geographische Gesellschaft e.V. (l’association de la Société Géographique de Marburg), la société géographique allemande ayant le plus grand nombre d’adhérents, depuis sa fondation en 1985 sachant que les premiers numéros de cette revue de l’Université de Marburg remontent aux années 1950. Le dernier volume (147) dont le titre pourrait être traduit par « Recherches sur l’eau – fondements et applications. Contributions scientifiques et pratiques suprarégi...

(2013). Land change modelling: moving beyond projections. International Journal of Geographical Information Science: Vol. 27, Land change modelling: moving beyond projections, pp. 1691-1695.

La commune de San Juan de la Costa (province d’Osorno – 10^ème Région de Los Lagos) connaît depuis la fin des années 1980 une expansion remarquable des plantations de pins et d’eucalyptus. Ces boisements sont à l’origine de nombreuses perturbations environnementales mais aussi sociales dans une région considérée comme un hotspot de biodiversité. Si les processus de changement d’occupation et d’utilisation du sol (lucc) sont dans l’ensemble bien connus, les zones de forêts natives susceptibles d’être converties en plantations industrielles sont quant à elles inconnues. Cet article se focalise sur la validation spatiale de deux modèles de simulation afin de proposer des interprétations prospectives en se basant sur une approche supervisée (chaînes de Markov couplées à une évaluation multicritère) et une approche automatique (réseau neuronal multicouches). Les deux modèles intègrent la connaissance des dynamiques passées grâce à l’analyse diachronique de trois images satellites landsat (1986-1999-2008).