Use of point patterns to model distribution of 2021 British Columbia wildfires

Bc. Polina Moisseyenko

Master's thesis

Abstract:

This diploma thesis deals with wildfires in boreal forests. Wildfires are one of the most frequent sources of tree cover loss in boreal forest regions. The pattern of wildfires has been changing globally. The author of this diploma thesis is aimed to evaluate the pattern of wildfires that occurred in British Columbia during the 2021 wildfire season in terms of 1st and 2nd order effects. The chosen external factors (mean wind speed, distance from roads, vegetation zones, and average drought level) were considered during the calculations. The results show that the distribution of wildfires had a clustering pattern in the southern parts of the province. All of the considered factors are associated with wildfire incidents. The author attempted to model the relationship between wildfire incidents and external factors. The final model was used to visualize a prediction of wildfires throughout the study area. The prediction was further compared to the wildfire events that occurred one year later. These results highlight that point pattern analysis of wildfire incidents can be used to predict areas of high wildfire risk.

Keywords

wildfires, kernel density, L-function, G-function, Poisson process

Language used: English

Date on which the thesis was submitted / produced: 31. 3. 2023

Thesis defence

• Date of defence: 17. 5. 2023
• Supervisor: D.Sc. Olga Špatenková
• Reader: Vojtěch Barták, Ph.D.

Czech University of Life Sciences Prague

Faculty of Environmental Sciences

Master programme / field:
Engineering Ecology / Nature Conservation