Nearing a Decade of Los Angeles Fires

Lab Reports, Maps

This map is an interactive visualization comparing the perimeters of Los Angeles County fires (2010-2018) with CAL FIRE Fire Hazard Severity Zones.

(If you experience any difficulty viewing the embedded map above, click the button below to view the map directly on


After the recent 2019 wildfires that spread throughout California, I wanted to explore what areas had been affected in my home county, Los Angeles. I was particularly inspired by a visualization created by Geospatial @ UCLA, in which they classified all of the buildings in California by their CAL FIRE Fire Hazard Severity Zone.

I like Geospatial @ UCLA‘s map because it’s accessible to an everyday user. The map provides a search function for users to directly look up an address, community name, or business to find what fire hazard zone the building belongs to. The zoom feature also allows for browsing specific areas of interest.

A screenshot of Geospatial @ UCLA’s interactive map


I used CARTO, an online GIS software platform, to merge the 2007 adopted State Responsibility Area (SRA) LA County GIS layer with the 2012 recommended LRA LA County GIS layer. I combined the two in order to provide a more comprehensive representation that encompassed all areas under risk of wildfires.

I then layered the Small Fire Perimeters (<5000 acres) and Large Fire Perimeters (>=5000) shapefiles over the map to show where fires actually occurred. These shapefiles encompass fires that occured from 2010 and up, however they have not yet been updated with the 2019 fires.


Base Layer
My visualization focuses on Los Angeles County and consists of four GIS layers. For the basemap, I used a style called “HERE Terrain Day”, which is a template option provided by CARTO. I decided on displaying the terrain so that a user who was just browsing, and perhaps unfamiliar with the orientation of the LA area, could glean insights based on the physical land and where the fire zones were located.

Middle Layers
The next layers depicted are the fire hazard severity zones. I chose to include a legend for the fire zones and changed the 1-2-3 severity scale used in the data to the severity terminology used in the CAL FIRE draft maps (Very High, High, and Moderate). I colored the zones a dark orange, light orange, and yellow to correspond to weight of their severity.

This image has an empty alt attribute; its file name is image-3.png

Top Layers
The final layers are the 2010-2018 fire perimeters, which are separated by large (>= 5,000 acres) and small fires (<5,000 acres). For these layers, I had to apply an “Intersect and Aggregate” Analysis in CARTO because the perimeter shapefiles showed fires that happened all over California, not just within LA County. This CARTO analysis removed any fire perimeters that were not overlapping with the fire hazard severity zones.

This image has an empty alt attribute; its file name is image-4.png

For the fire perimeters, I opted to outline the large fires in black and the smaller fires in gray. The thick outlines allow the areas to be seen while continuing to show the middle and base layers beneath.

Each perimeter also has a hover pop-up providing details about the fires of 2010-2018. Specifically, the year it occurred, the name of the fire, how many acres were burned, and the cause. I made the pop-up boxes different colors in order to differentiate the large fires from the small ones.

The goal of the visualization was to highlight the fire zones, but I thought it was also useful to acknowledge when the fires occurred. To highlight this, I added widgets based on the “year” column of my data, which appended “year” filters to the side of the map. I renamed the widgets accordingly so that the user could understand which layer they were interacting with.


As expected, many of the fire hazard zones are near forests or other areas with high vegetation. Juxtaposing the fire hazard severity zones with the actual fire perimeters reveals large patches of hazard zones that have gone unaffected for the past near decade, and others that were unaccounted for in the designated zones. It would be worth taking a closer look into how these zones are determined and what factors — if any (such as power lines, which were a huge focus in fire prevention this 2019) — are missing from the evaluation process.


Visual Congestion
I don’t regret selecting terrain as a base layer, however the specific template I used has particularly wide, white roads, which can obscure the light green map when zoomed out.

Condense layers
I decreased the opacity of the fire hazard severity zones in order to emphasize the perimeter outlines and maintain the terrain’s visibility. In doing so, the overlapping LRA and SRA areas create a darker hue that, albeit doesn’t ruin the map, potentially infers meaning when there isn’t necessarily greater fire hazard in that area. The overlap/darker hue simply means that the area is both under local and state responsibility. In the future, I would experiment with different terrain layers so that I could use full opacity on the fire hazard severity zones.

Furthermore, I left the large and small fires as separate layers, which display as separate widgets on the right side of the map. The map’s legibility could be simpler if I combined the two shapefiles rather than provide the separate options/widgets. There are different filter capabilities when you have two widgets available rather than one, however it is unlikely a user would, for example, want to display the 2018 small fires next to the 2016 large fires.

Searching locations
I tried using the search function on my completed visualization twice, but both times it glitched and distorted the terrain display on my map. In the future, I would like to find a way to include the search function because I found it very useful in the Geospatial @ UCLA visualization.