Lessons and challenges in land change modeling derived from synthesis of cross-case comparisons

This chapter presents the lessons and challenges in land change modeling that emerged from years of reflection and numerous panel discussions at scientific conferences concerning a collaborative cross-case comparison in which the authors have participated. We summarize the lessons as nine challenges grouped under three themes: mapping, modeling, and learning. The mapping challenges are: to prepare data appropriately, to select relevant resolutions, and to differentiate types of land change. The modeling challenges are: to separate calibration from validation, to predict small amounts of change, and to interpret the influence of quantity error. The learning challenges are: to use appropriate map comparison measurements, to learn about land change processes, and to collaborate openly. To quantify the pattern validation of predictions of change, we recommend that modelers report as a percentage of the spatial extent the following measurements: misses, hits, wrong hits and false alarms. The chapter explains why the lessons and challenges are essential for the future research agenda concerning land change modeling.