Search for a command to run...
Automatic identification of forest patches disturbed by the spruce bark beetle Ips typographus L. is crucial to reveal the rules of following bark beetle outbreaks on the landscape scale. Landsat imagery provides free resources to outline past and present gradations of bark beetle outbreaks (BBOs). The objective of this study is to identify the most sensitive vegetation index through different method of vegetation index differencing to identify past and actual bark beetle outbreaks. Six Landsat Thematic Mapper (TM) images, from 2005–2009 and 2011, were converted into selected vegetation indices (VIs) sensitive to conifer tree health in a Norway spruce–dominated forest in the High Tatra Mountains. The Vegetation Condition Index (VCI), Moisture Stress Index (MSI), Normalised Difference Moisture Index (NDMI), Normalised Difference Vegetation Index (NDVI),  Disturbance Index (DI) and Changed Disturbance Index (DI´) were calculated separately for every year, and the methodology of vegetation index differencing was applied to multiple two-year time periods (2005–2006, 2006–2007, 2007–2008, 2008–2009 and 2010–2011), thus producing the Changed Vegetation Index (ΔVI). A set of thresholds was established on ΔVI to classify disturbed and undisturbed forest due to BBOs; the sensitivity of different VIs to identify BBO was equally evaluated. The highest accuracies of classifications were reached in 2007 and 2011 (kappa index of agreement >70% and >40%, respectively), which were characterised by an epidemic phase of a BBO. All selected VIs were highly sensitive to BBOs, except for NDVI. The stable threshold value for change detection is not widely applicable to detect past forest disturbances caused by bark beetles, however. Finally, for further research of the epidemic phases of BBOs, we recommend the utilisation of the vegetation indices VCI, MSI and NDMI to detect BBOs because of their simplicity and easy interpretability.