Forecasting and Summarizing Wildfire Hazard in California

• Activity Number: 250
• Type: Contributed
• Date/Time: Monday, August 2, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics and the Environment
• Author(s): Kevin Edward Nichols*+
• Companies: University of California, Los Angeles
• Keywords: point process models ; prototype

Abstract

We will explore the use of multi-dimensional point process models for forecasting wildfire hazard in California. In particular, we extend previous space-time point process models for forecasting wildfire hazard in Los Angeles County using daily weather covariates by adding distances from major Los Angeles County roadways as an explanatory variable, and we attempt to quantify the relationship between these distances and wildfire hazard. In addition, we use point process prototypes to summarize the patterns of wildfire occurrences in California in particular months, seasons, and years. Such summaries provide a convenient description of the typical realization of wildfire occurrences in California for each time period and highlight the major seasonal trends in wildfire behavior in California. Methods of constructing standard errors for these point process prototypes are also investigated.