The Potential of ESA's Upcoming Sentinel-2 Satellites for Validating Deforestation and Forest Degradation in Guyana.
Galiatsatos, Nikolaos; Donoghue, Daniel; Mahmood, Jamil
Durham University, UNITED KINGDOM

Forest degradation is a land cover change which results from human activities that partially reduce the above ground carbon density without regeneration in a reasonable time frame. In other words, forest degradation is a non-temporary reduction of above ground carbon density. This paper aims to identify key challenges for quantifying forest degradation rates in humid tropical forests. It uses a case study to present results that help to identify strengths and weaknesses in the use of optical satellite imagery for mapping forest cover and tracking changes that can be attributed to forest degradation over a 30 year period from 1990 through to 2010. The paper has three objectives.
First we give a brief overview of the how satellite imagery can be used to assist with deforestation and forest degradation mapping in order to assess the quality of map product that can be typically produced. In this case, the focus is on the use of RapidEye imagery and automatic and manual methods of mapping. The second objective is to demonstrate an approach to evaluate the accuracy of mapping using very high resolution imagery and ground/aerial surveys. The paper describes sample design, response design, and estimation and analytical methods. Finally, the paper provides an evaluation of the strengths and weaknesses of Earth observation in estimating deforestation rates in cloud covered tropical regions. Guyana is the study area and it poses a challenge for satellite imagery as it is often cloud covered and acquiring a cloud-free image within a particular time-frame is extremely difficult. Satellite constellations such as RapidEye have proven very valuable in cloud covered tropical regions and a dataset covering 56% of Guyana was acquired thus focusing on the main deforestation and forest degradation areas within the country. We are comparing RapidEye and upcoming Sentinel-2 constellations and identify the possible issues that one may face when using Sentinel-2 for mapping and monitoring deforestation and forest degradation in persistently cloudy tropical areas like Guyana.