Remote sensing is gaining considerable traction in modern forest monitoring efforts, with the Carnegie Landsat Analysis System lite (CLASlite) software package and the Global Forest Change dataset (GFCD) being two of the most recently developed optical remote sensing-based tools for analysing forest cover and change. New research from the Coomes group recently published in the journal Remote Sensing evaluates the abilities of these relatively nascent technologies to classify land cover and forest dynamics against a more established classification approach.
In our study, we compared maps of forest cover and change produced by a traditional supervised classification approach with those produced by CLASlite and the GFCD, working with imagery collected over Sierra Leone, West Africa. Overall, we found that CLASlite maps of forest change exhibited the highest overall accuracies and, importantly, the greatest capacity to discriminate natural from planted mature forest growth. We concluded that CLASlite’s comparative advantage likely derived from its more robust sub-pixel classification logic and numerous user-defined parameters, which resulted in classified products with greater site relevance than those of the two other classification approaches. In light of today’s continuously growing body of analytical toolsets for remotely sensed data, our study elucidates the ways in which methodological processes and limitations inherent in certain classification tools can impact the maps they are capable of producing, and demonstrates the need to understand and weigh such factors before any one tool is selected for a given application.