Premium
The application of prototype point processes for the summary and description of California wildfires
Author(s) -
Nichols Kevin,
Schoenberg Frederic Paik,
Keeley Jon E.,
Bray Andrew,
Diez David
Publication year - 2011
Publication title -
journal of time series analysis
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.576
H-Index - 54
eISSN - 1467-9892
pISSN - 0143-9782
DOI - 10.1111/j.1467-9892.2011.00734.x
Subject(s) - point process , point (geometry) , cluster analysis , process (computing) , computer science , hierarchical clustering , cartography , meteorology , geography , mathematics , statistics , machine learning , programming language , geometry
A method for summarizing repeated realizations of a space‐time marked point process, known as prototyping, is discussed and applied to catalogues of wildfires in California. Prototype summaries are constructed for varying time intervals using California wildfire data from 1990 to 2006. Previous work on prototypes for temporal and space‐time point processes is extended here to include methods for computing prototypes with marks and the incorporation of prototype summaries into hierarchical clustering algorithms, the latter of which is used to delineate fire seasons in California. Other results include summaries of patterns in the spatial‐temporal distribution of wildfires within each wildfire season.