Evaluation of Fire Ignition Probability Model by Using Spatial Autocorrelation Method: A Case Study in Yunnan Province

DOI

10.2991/icammce-17.2017.35How to use a DOI?

Keywords

spatial autocorrelation analysis; model residuals; Yunnan Province

Abstract

The forests in Yunnan province are dense and precious. Meanwhile, the forests fire happens frequently. There are ten models generated with two modeling methods in the study area, Yunnan province. In this paper, the spatial autocorrelation method was applied to analyze and evaluate the performance of fire ignition probability models. The results show that: 1) the models generated by geography weight logistic regression method are better than the ones created by binary logistic regression method; 2) compared with other models, the Model 7 is optimal in terms of evaluating by the quantitative spatial autocorrelation method. The results are of significance for making decisions of establishing early warning systems.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Wei Wang
AU  - Zhanqiang Chang
AU  - Xiaomeng Liu
AU  - Jie Zhu
AU  - Wen Yu
PY  - 2017/04
DA  - 2017/04
TI  - Evaluation of Fire Ignition Probability Model by Using Spatial Autocorrelation Method: A Case Study in Yunnan Province
BT  - Proceedings of the 2017 2nd International Conference on Advances in Materials, Mechatronics and Civil Engineering (ICAMMCE 2017)
PB  - Atlantis Press
SP  - 166
EP  - 171
SN  - 2352-5401
UR  - https://doi.org/10.2991/icammce-17.2017.35
DO  - 10.2991/icammce-17.2017.35
ID  - Wang2017/04
ER  -