Analyzing Southern California Wildfires with Spatial Data
Analyzing Southern California Wildfires with Spatial Data
by Heather Robbins
Embedded Files
Spatial Questions
Spatial Questions
Attribute Queries
Attribute Queries
How many wildfires have burned more than 10,000 acres in Southern California between 2013 – 2023?
Which cities in the study area have a high population (100,000 or more people) and high population density (over 10,000 per square mile)?
Spatial Queries
Spatial Queries
Which reservoirs and lakes have been affected by large historical fires?
How many large fires burned within 5 miles of a major highway?
Spatial Joins
Spatial Joins
Which populated area experienced the most wildfires from 2000 – 2023?
What is the average distance from an air attack base to past wildfires?
Combinations of Queries and Joins
Combinations of Queries and Joins
Which city owned fire stations serve the largest number of fires based on proximity?
How many people live within 5 miles of an area that has burned more than once?
Study Area
Study Area
In California, wildfires are a well-established natural hazard occurring statewide. The combination of hot, dry summers and strong seasonal winds allows fires to ignite and spread rapidly—sometimes on a near-daily basis (South Ops GACC, n.d.). The widespread urban-wildland interface further intensifies risk by placing large populations in direct proximity to fire-prone areas.
Data Collected
Data Collected
Historical wildfire perimeter data was obtained from Cal FIRE and includes all documented wildfire events through the year 2023. Unfortunately, data beyond 2023 was not available at the time of this analysis. Although the dataset spans events as far back as the 1800s, only fires occurring between 2000 and 2023 are considered to ensure relevance and recency.
Additional datasets were collected to support spatial analysis, including California county boundaries, lakes and reservoirs, census-designated populated areas, fire support facilities, and the National Highway System. These layers were used to explore spatial relationships through a combination of spatial queries, attribute queries, and spatial joins.
For purposes of the analysis, southern California refers to Imperial, Kern, Los Angeles, Orange, Riverside, Santa Barbara, San Bernardino, San Diego and San Luis Obispo counties. Operations Southern California (OSCC) works in partnership with the US Forest Service and Cal FIRE to help in the preventative efforts and suggests 50 acres is the threshold of what is considered a large fire. For this analysis, only larger fires are considered.
Methods
Methods
The study area boundary was created by manually selecting the relevant counties and exporting them to a new shapefile. All collected vector datasets were projected to UTM Zone 11N using the Project tool to preserve shape and distance accuracy. These datasets were then clipped to the study area using the Clip tool.
To ensure data quality, each dataset was reviewed for incomplete or missing attribute fields that could affect analysis. A definition query was applied to the wildfire perimeter dataset to include only fires larger than 50 acres and occurring between 2000 and 2023, resulting in a total of 1,313 fire events. A similar query was applied to the fire support facilities dataset to include only active facilities.
The Cal FIRE facilities layer includes a variety of support sites (e.g., command centers, nurseries, headquarters), but only fire stations and air attack bases were used for this analysis. These were extracted using the Make Feature Layer tool to create two new layers.
Attribute Queries
Attribute Queries
Attribute queries were processed using the Select by Attributes tool:
Wildfires: Year >= 2013 AND GIS_Acres > 10,000
Populated Areas: Population >= 100,000 AND POP_SQMI > 10,000
Spatial Queries
Spatial Queries
To assess the impact of wildfires on local water sources, a spatial query is used to select water features (Lakes and Reservoirs) that intersect with fire boundaries using the Intersect tool. This creates a new layer, Impacted Water Bodies.
Major highways were selected from the National Highway System using NHS Code = 1. A 5-mile buffer was created using the Buffer tool, resulting in a new layer (HighwayBuf5). Fires within this buffer were extracted using the Clip tool to generate a layer, Fires Close to Highway.
Spatial Joins
Spatial Joins
A Spatial Join was used to relate wildfire events to populated areas. The Populated Areas layer served as the target feature, and Wildfires as the join feature. Non-essential fields were removed, retaining only city name, population, population density, and fire name.
To determine the average distance to the nearest air attack base, a Spatial Join was performed with Wildfires as the target and Air Attack Bases as the join feature, using the "Closest" match option. The resulting distance field (in meters) was converted to miles by creating a new field and applying the calculation:
Distance / 1604.34
The Summary Statistics tool was used to calculate the mean distance in miles.
Combinations
Combinations
The Fire Stations layer was queried by owner to select city-owned stations, which were exported as a new layer (CityFireStations). A Spatial Join was then performed with wildfires using a 15-mile distance threshold, using "Within a distance" as the match option and a one-to-one join operation.
The Count Overlapping Features tool was used on the wildfire perimeter layer to identify areas that had burned more than once, with a minimum overlap count of 2. This new layer (MultipleFires) was then buffered by 5 miles, dissolving all features into a single output. An Intersect operation was used to extract the Populated Areas layer within the buffer, and the attribute table was used to calculate the total population affected.
Assumptions
Assumptions
Southern California is widely recognized as a region of high wildfire risk. This analysis serves to highlight communities and infrastructure vulnerable to recurring or severe fire events, which may inform future prevention and response funding strategies. Given its population size, Los Angeles is expected to be heavily represented in the results, particularly in terms of the number of people directly impacted by proximity to wildfire-prone areas.
Results
Results
Large Wildfires (2013–2023)
Large Wildfires (2013–2023)
A total of 33 wildfires burned more than 10,000 acres between 2013 and 2023. The largest among them was the Thomas Fire, which began on December 4, 2017, and burned 281,709 acres.
High-Population-Density Cities
High-Population-Density Cities
Five cities have a population exceeding 100,000 and a population density greater than 10,000 people per square mile:
Water Bodies Impacted by Fires
Water Bodies Impacted by Fires
There were 17 intersections between wildfire perimeters and water bodies. A combined area of 8,744 square meters (approximately 0.009 km²) was affected, as calculated from the Shape_Area field. Notable impacted water sources include:
Castaic Lake
Diamond Valley Lake
Elderberry Forebay
Isabella Lake
Lakes Cachuma, Casitas, Nacimiento, Piru
Pyramid Lake
San Vicente Reservoir
Fires Near Highways
Fires Near Highways
A total of 435 wildfires occurred within 5 miles of major highways, indicating widespread interaction between fire perimeters and transportation infrastructure. In many instances, fire perimeters crossed highway corridors entirely.
Populated Areas with Frequent Fires
Populated Areas with Frequent Fires
An assessment of the Populated Areas with Fires layer revealed that Los Angeles experienced the most wildfires since 2000 (48 total), followed by:
Thousand Oaks – 40 fires
Santa Clarita – 33 fires
Distance to Air Attack Bases
The average distance from historical wildfire events to the nearest air attack base is 24.5 miles. Additional statistics include:
Distance to Air Attack Bases
The average distance from historical wildfire events to the nearest air attack base is 24.5 miles. Additional statistics include:
Median Distance: 20.8 miles
Maximum Distance: 169.7 miles
City Fire Stations and Proximity to Fires
City Fire Stations and Proximity to Fires
Analysis of the CityFireStations layer indicates that the Highland 541 fire station is located near the highest number of historical fires. An attribute query revealed this station is within 15 miles of 85 wildfires.
Population Near Reburn Areas
Population Near Reburn Areas
A 5-mile buffer was created around areas where wildfires occurred more than once. The total population within this buffer is approximately 16.9 million people.
Conclusions
Conclusions
This analysis confirms that Southern California counties face a disproportionately high risk of wildfire events, with densely populated urban areas often located within or near impacted zones. The presence of more than 33 large fires since 2013—and over 430 fires in proximity to major highways—demonstrates how wildfire threats intersect with critical infrastructure, water resources, and population centers.
The proximity of populated areas to repeated burn zones is particularly alarming. The estimated population of more than 16.9 million people living within five miles of overlapping wildfires highlights both the widespread nature of these threats and the urgency for proactive mitigation and response strategies.
One especially revealing result is the maximum distance of 170 miles between a historical wildfire and the nearest air attack support base. While the average and median distances are more reasonable, the extreme outlier indicates potential gaps in aerial wildfire response infrastructure and suggests the need for deeper evaluation of spatial coverage and response times.
This analysis supports the need for targeted fire prevention strategies, improved emergency infrastructure, and sustained investment in public safety measures—particularly in regions such as Los Angeles County, where population exposure and fire frequency are most pronounced.
Limitations
Limitations
Several limitations should be acknowledged. First, the fire data used in this analysis extends only through 2023; more recent wildfire events from 2024 onward are excluded and may reflect important shifts in fire frequency, size, or location. As climate variability continues to influence fire behavior, incorporating up-to-date data would allow for a more comprehensive and current assessment.
In the population exposure analysis (Q8), the use of the Intersect tool may overestimate the number of residents affected, as population values reflect entire city boundaries, not just those portions within the five-mile buffer of overlapping fires. A more accurate approach might involve spatially weighted population data by recalculating population based on the portion of area compared to the original area.
Recommendations for Future Work
Recommendations for Future Work
Several avenues for further analysis emerged during this project:
Burn Area Ratios: Calculating the percentage of each county that has burned over the past two decades could highlight regions at risk of ecological degradation or long-term habitat loss.
Suitability Analysis: A suitability model could help identify optimal locations for new fire support facilities, such as air attack bases or city fire stations, especially in regions that experience long response times.
Agency Demand Assessment: Further analysis of the wildfire dataset’s "Assigned Agency" attribute could help determine which local, state, or federal agencies are responding most frequently. This insight could support more equitable resource allocation across jurisdictions.
Funding Prioritization: Facilities like the Highland 541 fire station, which is near an unusually high number of wildfire events, could serve as case studies for identifying sites that may warrant additional funding or infrastructure upgrades.
References
References
CAL FIRE facilities for wildland fire protection (facility24_1). (2020). [Dataset]. CA Governor’s Office of Emergency Services. https://data.ca.gov/dataset/facilities-for-wildland-fire-protection
California Department of Forestry and Fire Protection. (2019). California County Boundaries [Dataset]. https://gis.data.ca.gov/datasets/8713ced9b78a4abb97dc130a691a8695_0/about
California Department of Forestry and Fire Protection. (2024). California Fire Perimeters (ALL) [Dataset]. https://gis.data.cnra.ca.gov/datasets/CALFIRE-Forestry::california-fire-perimeters-all/about
California Department of Water Resources. (n.d.). NHD Major Lakes and Reservoirs Shapefile [Dataset]. In National Hydrography Data. https://data.cnra.ca.gov/dataset/national-hydrography-dataset-nhd
Federal Highway Administration [FHWA] & Bureau of Transportation Statistics [BTS]. (2024). National Highway System (NHS) [Dataset]. In National Transportation Atlas Database (NTAD) (2024.05.01). U.S. Department of Transportation. https://data-usdot.opendata.arcgis.com/datasets/usdot::national-highway-system-nhs/about
O’Sullivan, D., & Unwin, D. (2010). Geographic Information Analysis.
Price, M. H. (2019). Mastering ArcGIS Pro.
South Ops GACC. (n.d.). Large Fire Risk Probability Graphic User Guide/FAQs. https://gacc.nifc.gov/oscc/predictive/weather/myfiles/firerisk_guide.pdf
U.S. Census Bureau & Esri. (2013). U.S. census populated place areas [Dataset]. Esri Content - ArcGIS Online. https://services.arcgis.com/P3ePLMYs2RVChkJx/arcgis/rest/services/USA_Census_Populated_Places/FeatureServer
Page updated
Google Sites
Report abuse