Landsat Imagery Spectral Trajectories—Important Variables for Spatially Predicting the Risks of Bark Beetle Disturbance

Abstract

Tree mortality caused by bark beetle infestation has significant effects on the ecology and value of both natural and commercial forests. Therefore, prediction of bark beetle infestations is critical in forest management. Existing predictive models, however, rarely consider the influence of long-term stressors on forest susceptibility to bark beetle infestation. In this study we introduce pre-disturbance spectral trajectories from Landsat Thematic Mapper (TM) imagery as an indicator of long-term stress into models of bark beetle infestation together with commonly used environmental predictors. Observations for this study come from forests in the central part of the Šumava Mountains, in the border region between the Czech Republic and Germany, Central Europe. The areas of bark beetle-infested forest were delineated from aerial photographs taken in 1991 and in every year from 1994 to 2000. The environmental predictors represent forest stand attributes (e.g., tree density and distance to the infested forest from previous year) and common abiotic factors, such as topography, climate, geology, and soil. Pre-disturbance spectral trajectories were defined by the linear regression slope of Tasseled Cap components (Wetness, Brightness and Greenness) calculated from a time series of 16 Landsat TM images across years from 1984 until one year before the bark beetle infestation. Using logistic regression and multimodel inference, we calculated predictive models separately for each single year from 1994 to 2000 to account for a possible shift in importance of individual predictors during disturbance. Inclusion of two pre-disturbance spectral trajectories (Wetness slope and Brightness slope) significantly improved predictive ability of bark beetle infestation models. Wetness slope had the greatest predictive power, even relative to environmental predictors, and was relatively stable in its power over the years. Brightness slope improved the model only in the middle of the disturbance period (1996). Importantly, these pre-disturbance predictors were not correlated with other predictors, and therefore bring additional explanatory power to the model. Generally, the predictive power of most fitted model decreases as time progresses and models describing the initial phase of bark beetle outbreaks appear more reliable for conducting near-future predictions. The pre-disturbance spectral trajectories are valuable not only for assessing the risk of bark beetle infestation, but also for detection of long-term gradual changes even in non-forest ecosystems.

Evaluation of the Influence of Disturbances on Forest Vegetation Using the Time Series of Landsat Data: A Comparison Study of the Low Tatras and Sumava National Parks

“…When the three sites affected by wind calamities in Low Tatras and Sumava NPs were compared, similar result could be seen. The vegetation indices based on the use of the SWIR band (e.g., the NDMI and wNDII index) show similar trends as well [46]. A bark beetle calamity has a much more complicated progress with a specific effect on the vegetation.…”

“…The time series results obtained can be applied to generate predictions of future states and to observe the spontaneous and economically managed forest regeneration. The use of satellite data provides progressive opportunities to monitor and evaluate the state and changes in the forest vegetation for the purpose of the protection and management of national parks [46,47]. Figure A2.…”

Post-disturbance recovery of forest cover and tree height differ with management in Central Europe

“…2, respectively). For further characterizing the pre-disturbance spectral characteristics, we also calculated the slope of all observations before disturbance (pre-disturbance slope; Hais et al 2016).…”

Forest health monitoring in transition: Evaluating insect-induced disturbances in forested landscapes at varying spatial scales

“…The information can be used in management operations to control large-scale outbreaks (Senf et al 2017b). Research include hyper-spectral detection of changes in chlorophyll absorption (Lausch et al 2013) and detection of stand level stress from Landsat time-series as an indicator of susceptibility to bark beetle infestations (Hais et al 2016).…”