Quantifying Forest Change Using Remotely Sensed Time Series Observations – The Case of Northeastern Armenia

The talk will outline the methods of image time series analysis with the use of state-of-the-art, freely accessible satellite data for forest monitoring. The focus of the presentation will be the fusion of data handling from several sensors, methods for satellite image time series analysis, and the results of the author’s master’s thesis carried out at University of Münster (Germany) about forest dynamics in the northeast of Armenia. Continuous forest disturbances, mainly caused by illegal logging, have started from the early 1990s and have caused huge damage to forest ecosystems by decreasing forest productivity and rendering larger areas vulnerable to erosion. In order to gain insight about forest cover and disturbances over a long time period we used Landsat TM/ETM+ images. Google Earth Engine was used for data processing, which is a cloud-based tool enabling the access and analysis of large amounts of satellite imagery. To overcome the data availability problem caused by the gap in the Landsat series from 1988 to 1998, extensive cloud cover in the study area and missing scan lines, we used pixel-based compositing. In order to derive the disturbances only in forests, forest cover layers were generated using Classification and Regression Trees. The last part of the talk will focus on seasonality analysis and anomaly detection based on data from the Moderate Resolution Imaging Spectroradiometer (MODIS).

Language: English