Assessing Combinations of Landsat, Sentinel-2 and Sentinel-1 Time series for Detecting Bark Beetle Infestations

Bark beetle infestations are among the most substantial forest disturbance agents worldwide. Moreover, as a consequence of global climate change, they have increased in frequency and size number affected areas. Controlling bark outbreaks requires consistent operational monitoring, is possible using satellite data. However, while many satellite-based approaches been developed, full potential dense, multi-sensor time series has yet to be fully explored. Here, for first time, we used all available multispectral data from Landsat Sentinel-2, Sentinel-1 SAR data, combinations thereof detect Bavarian Forest National Park. Based on multi-year reference dataset annual infested areas, assessed separability between healthy forests various vegetation indices calculated We two compute infestation probability different datasets: Bayesian conditional probabilities, based best-separating index each type, random regression, type. Five sensor configurations were tested their detection capabilities: alone, Sentinel-2 combined, types combined. The best overall results terms spatial accuracy achieved with (max. accuracy: 0.93). detections also closest onset estimated year. detected areas larger contiguous patches higher reliability compared smaller patches. somewhat inferior those 0.89). While yielding similar results, combination did not provide any advantages over or alone 0.87), was unable 0.62). combined three achieve satisfactory either 0.67). Spatial accuracies typically probabilities than forest-derived but latter resulted earlier detections. approach presented herein provides flexible pipeline well-suited monitoring outbreaks. Furthermore, it can applied other types.