This article is very interesting, and addresses what I think is a major issue within GIScience-the scale issue. Marceau (1999) lays out the “scale problem” (2), and provides a thorough review of solutions (and their limitations) from the literature. I also enjoy the before last paragraph of the paper, which suggests that the “methodological developments are certainly contributing to the emergence of a new paradigm: a science of scale” (12). While reading the paper, I wondered how this fit into the tool/science debate, and though I would tend to think of it as an important component within GIScience, I might not have considered “the science of scale” on its own, so it’s nice to see how the author clearly feels.
This issue seems omnipresent throughout geography (human and physical), and I know that I’ve had to deal with it within my own work. For example, my data collection will consist of me flying a UAV at a specific height (in order to achieve maximum photo resolution), thereby taking photos at specific scales. I will then create a model to make maps at specific scales. Beyond this, the maps I make will hopefully tell me things about the morphology of the landscape: will this be true only of Eureka Sound, or will it be generalizable to all of Ellesmere Island, or even all of the Canadian or International high Arctic? I do not find that any of the methods described in this paper provide a clear way to give a definitive answer on cross-scale inferences, which is to be expected. I think that as researchers, we must do our best to limit our inferences to the analyzed scales, and resist temptations to overgeneralize our results for increased importance. I am curious how things have changed in the nearly 20 years since this article has been published, what strides have been made, and what remains to be done.