Link to Session Platform

Chairs(s):

Robert Pontius, Clark University

; ,

Description:

Land change studies are getting increasing attention, as land change serves as both one of the causes and consequences of global environmental change. Recent advances such as the increasing availabilities of time-series remotely sensed data, the growing power of computing, and the rapid development of machine learning algorithms have opened up new opportunities for land change studies. However, opportunities also led to new challenges. Despite the growing number of change detection and land change modeling methods, robust change detection and modeling methods suitable for various land cover types and dynamic patterns are still lacking. Many methods for quantifying land change have limited abilities in characterizing land change effectively. Popular land change quantification methods often report only net gain or loss of a land cover, ignoring the gross changes, the transition patterns of a land cover across a time series, and the spatial-temporal pattern of land changes. Furthermore, there are a large number of machine-learning methods for detecting and modeling land change, but there is a lack of studies assessing the robustness and applicability of these methods.

This session invites research focusing on methods and applications on the detection, quantification, and modeling of land change, with an emphasis on addressing the above-mentioned challenges. Interested presenters should please send your title and abstract of no more than 250 words to session chair Robert Gilmore Pontius Jr at rpontius@clarku.edu and the session organizer Aiyin Zhang at AiZhang@clarku.edu.

Non-Presenting Participants Agenda