University of Twente Student Theses

Login

Risk assessment of bark beetle outbreak in the Schwarzwald national park

Fernandez Perez, Fernando (2020) Risk assessment of bark beetle outbreak in the Schwarzwald national park.

[img] PDF
1MB
Abstract:The main threat of Norway spruces (Picea abies L.) in Europe is the bark beetle (Ips typographus L.). This insect is able to colonize weakened trees, but when there is a mass-attack, it can also breed in healthy trees. Forest management in Europe tries to minimize the severe effects of this pest. By knowing in which areas the outbreak probability is higher, the monitoring resources can be optimized. Modeling the distribution of bark beetles has been done before, obtaining different results in each study area. Therefore, our objective is to identify which spatial variables can be used for predicting the bark beetle outbreak in the Schwarzwald National Park. To that aim, a model created with boosted regression trees was applied. In order to avoid spatial autocorrelation, ten sub-models were calculated. The most important variables for predicting bark beetle outbreak in the study area were number of Norway spruces, height of Norway spruces, altitude, soil depth, slope, and percentage of Norway spruces compared with other tree species. These variables, in combination with other predictors, were used for creating the outbreak probability map. By using AUC as a validation method, the accuracy was 0.8. The model performance was also assessed with TSS, obtaining an accuracy of 0.49. This research provides insights into the spatial variables that can predict bark beetle outbreaks, which will support the decision-making process carried out by the National Park committee.
Item Type:Essay (Master)
Faculty:ITC: Faculty of Geo-information Science and Earth Observation
Programme:Spatial Engineering MSc (60962)
Link to this item:https://purl.utwente.nl/essays/84933
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page