TY - JOUR
T1 - Combining ungrouped and grouped wildfire data to estimate fire risk
AU - Hernandez-Magallanes, I.
N1 - KAUST Repository Item: Exported on 2020-10-01
Acknowledgements: The author is a postdoctoral fellow at the Institute for Applied Mathematics and Computational Science at Texas A&M University. The research was perform as a part of a PhD dissertation at the University of California, Berkeley. I would like to thank Professor D. Brillinger for his support and guidance. Thanks also to my committee members Professors D. Nolan and G. Judge at UC Berkeley and Dr. H. Preisler at USDA Forest Service Pacific Southwest Research Station for providing useful comments. R. Burgan at the USDA Forest Service, Rocky Mountain Research Station kindly provided the data. This work was partially funded by the UC MEXUS-CONACYT Doctoral Fellowship and IAMCS-KAUST Postdoctoral Fellowship.
This publication acknowledges KAUST support, but has no KAUST affiliated authors.
PY - 2013/10/11
Y1 - 2013/10/11
N2 - © 2013 John Wiley & Sons, Ltd. Frequently, models are required to combine information obtained from different data sources and on different scales. In this work, we are interested in estimating the risk of wildfire ignition in the USA for a particular time and location by merging two levels of data, namely, individual points and aggregate count of points into areas. The data for federal lands consist of the point location and time of each fire. Nonfederal fires are aggregated by county for a particular year. The probability model is based on the wildfire point process. Assuming a smooth intensity function, a locally weighted likelihood fit is used, which incorporates the group effect. A logit model is used under the assumption of the existence of a latent process, and fuel conditions are included as a covariate. The model assessment is based on a residual analysis, while the False Discovery Rate detects spatial patterns. A benefit of the proposed model is that there is no need of arbitrary aggregation of individual fires into counts. A map of predicted probability of ignition for the Midwest US in 1990 is included. The predicted ignition probabilities and the estimated total number of expected fires are required for the allocation of resources.
AB - © 2013 John Wiley & Sons, Ltd. Frequently, models are required to combine information obtained from different data sources and on different scales. In this work, we are interested in estimating the risk of wildfire ignition in the USA for a particular time and location by merging two levels of data, namely, individual points and aggregate count of points into areas. The data for federal lands consist of the point location and time of each fire. Nonfederal fires are aggregated by county for a particular year. The probability model is based on the wildfire point process. Assuming a smooth intensity function, a locally weighted likelihood fit is used, which incorporates the group effect. A logit model is used under the assumption of the existence of a latent process, and fuel conditions are included as a covariate. The model assessment is based on a residual analysis, while the False Discovery Rate detects spatial patterns. A benefit of the proposed model is that there is no need of arbitrary aggregation of individual fires into counts. A map of predicted probability of ignition for the Midwest US in 1990 is included. The predicted ignition probabilities and the estimated total number of expected fires are required for the allocation of resources.
UR - http://hdl.handle.net/10754/597801
UR - http://doi.wiley.com/10.1002/env.2235
UR - http://www.scopus.com/inward/record.url?scp=84908167468&partnerID=8YFLogxK
U2 - 10.1002/env.2235
DO - 10.1002/env.2235
M3 - Article
SN - 1180-4009
VL - 25
SP - 365
EP - 383
JO - Environmetrics
JF - Environmetrics
IS - 6
ER -