Global and local spatial analysis for applied goegraphical research

点击翻译按钮获取中文摘要

For many issues that greatly concern Canadians, action requires locational decisions: detection and monitoring of environmental hazards, and efficient response to epidemic outbreaks, are but a few examples where most key factors are located in geographical space. Presently, management decisions are supported by quantitative models; these models are potentially desirable tools, which can link, for example, disease incidence to health hazards, thus providing a realistic picture of where health care services will be most needed in the near future. Unfortunately, current models are often uncertain, therefore unreliable. In the best cases, unreliable models provide decision makers with a realistic, but blurry picture of the factors they need to manage, leading to ineffective decisions; in the worst cases the picture is so blurry that it is confusing, and may lead to management decisions that are not just ineffective, but harmful. The uncertainty stems from two properties of geographical factors: things vary unevenly in space, and near things are more similar than distant things. These two properties often occur simultaneously, whereas each existing modeling method can only address one of them; therefore, each method only partially clears the blurriness of a picture. Worse: each method, in trying to address one source of uncertainty, can aggravate the other, increasing the overall uncertainty.