Introduction
- Whitlock et al. suggests that the future fire condition could be more severe in northwestern US based on the simulation of potential future climate and vegetation. (C. Whitlock et al., 2003)
- Odion et al. implies that fires in multiaged, closed forests, the predominant vegetation, was less severe than open forest and shrubby non forest vegetation in western United States during 1920 to 1987. (D. Odion et al., 2004)
- According to Vadrevu et al., "Use of vegetation fire statistics including records of ignition sources and the number of fire occurrences, is an effectual method to quantify the temporal and spatial characteristics of fire regimes" (K. Vadrevu et al., 2011)
Objectives
- To show distribution of hotspots of fire in U.S. from 2001 to 2011
- To identify vegetational characteristics of the hotspots
- To discuss temporal changes of the hotspots
Study Area
Continental United States
Data Description
- MODIS Fire Detection GIS Data Cumulative fire detection data from 2001 to 2011 Location of Active Fires, Fire Radiative Power(FRP) Spatial Resolution: 1 km http://activefiremaps.fs.fed.us/gisdata.php
- Land cover GlobCover 2009 land cover map Global Land Cover Product (2005-06) Spatial Resolution: 290 m x 260 m http://ionia1.esrin.esa.int/
Method
- Compute annual mean center and standard deviation ellipse and Kernel Density of fire from 2001 to 2011
- Compute and Global Moran’s I
- Cluster analysis to identify hotspots of fires
- Divide Point data into two groups
- Group1: 2001 to 2006
- Group2: 2007 to 2011
- K-Means Partitioning Clustering using CrimeStat(Clusters:5, Separation: 4, Number of Standard Deviations for the Ellipses: 1X)
- Overlay analysis to identify vegetational characteristics of clusters
- Cross tabulation(Tabulate Area) using hotspots ellipse and vegetation data
- Discuss the temporal change of the clusters
- Changes in location, frequency, and vegetational characteristics
Result
Annual Change of Mean Center
Temporal Trend was not observed by computing mean centers and standard deviation ellipse.
Kernel Density
Areas with high density can be observed near Idaho, Kansas-Oklahoma, and Georgia-Florida.
Global Moran’s I
The Number of Fire
The Average FRP of Fire
Both the number of fire and the average FRP are spatially autocorrelated in state level.
Conclusion
- Fire occurrence and Average FRP in US is spatially autocorrelated in the level of state.
- Hotspots are observed in near Idaho, Southern California, Kansas-Oklahoma, Georgia-Florida in each period of years
- Temporal change of location of annual hotspots are not significant. However, most of the hotspots for 2001 to 2006 are located along the border or edge of the continent, an additional hotspot is observed in the middle of US for 2007 to 2011.
- In the western hotspots near Idaho and Southern California, "Closed to open shrubland" occupies large portion.
- In the southwestern hotspots near Georgia and Florida, "Closed broadleaved deciduous forest" occupies large portion.
Further Considerations
- Regression analysis of number of fire cases and land use
- Regression analysis of severity of fire and land use to find out the relationship between fire severity and vegetation
- Methods to observe temporal change of fire severity
- Employ Spatiotemporal Pattern Analysis
References
- C. Whitlock, S. Shafer, J. Marlon, "The role of climate and vegetation change in shaping past and future fire regimes in the northwestern US and the implications for ecosystem management", Forest Ecology and management, 2003, Vol. 178, Issue 1-2, p 5-21, DOI: 10.1016/S0378-1127(03)00051-3
- D. Oddion, E. Frost, J. Strittholt et al., "Patterns of Fire Severity and Forest Conditions in the Western Klamath Mountains, California Patrones de Severidad de Fuego y Condiciones del Bosque en las Montañas Klamath Occidentales, California", Conservation Biology, 2004, Vol. 18-4, p. 927-936, DOI: 10.1111/j.1523-1739.2004.00493.x
- K. Vadrevu, I. Csiszar, E. Ellicott et al., "Hotspot Analysis of Vegetation Fires and Intensity in the Indian Region", IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING, 2011, Vol. PP, Issue. 99, p. 1-15