Main content

A Study of Methods for Spatial Interpolation of Fire Weather in the Canadian Prairies

Show full item record

Title: A Study of Methods for Spatial Interpolation of Fire Weather in the Canadian Prairies
Author: Cheng, Yue
Department: Department of Mathematics and Statistics
Program: Mathematics and Statistics
Advisor: Ali, AyeshaNadeem, Khurram
Abstract: Thousands of wildfires occur in Canadian forests every year and it is very challenging for fire management agencies to predict weather conditions and fire risk especially in the areas with low weather station density. This thesis compares several existing interpolation models (ordinary kriging, ordinary cokriging, and thin-plate spline smoothing) to the inverse distance weighting which is used by Canadian fire management agencies using weather station data from Manitoba and Saskatchewan on a daily basis. The North American Regional Reanalysis (NARR) data extracted from physical models, which has a strong correlation with weather station data, is integrated into spatial interpolation models. This thesis also integrates elevation into ordinary cokriging and the thin-plate spline smoothing models. Results show that integrating NARR into ordinary cokriging and thin-plate spline smoothing model increases prediction accuracy in areas with low station density and could be potentially useful for Canadian fire management agencies.
Date: 2020-05
Terms of Use: All items in the Atrium are protected by copyright with all rights reserved unless otherwise indicated.

Files in this item

Files Size Format View
Cheng_Yue_202005_Msc.pdf 8.595Mb PDF View/Open

This item appears in the following Collection(s)

Show full item record