Expert interviews about the use of visually depicted uncertainty for analysis of remotely sensed land cover change

Change analysis based on remote sensing data typically involves a high degree of uncertainty. Past research has shown that ignoring this fact can lead to questionable or unusable results. Although methods exist to describe and communicate geodata uncertainty the incorporation of uncertainty into analyses remains a challenge. We developed a concept to facilitate the use of uncertainty in change analyses including a change uncertainty measure, a technique to visualize uncertainty, and a software prototype for change analysis following a geovisual analytics approach. To discuss the practical use of the concept and identify its benefits and limitations we conducted semi-structured interviews with three expert groups conducting change analysis in different application fields: climate research, urban development, and vegetation monitoring. This paper discusses the methodology, summarizes preliminary results, and highlights first lessons learned from the interviews.