Picture any regular map that has a large set of point data. Now imagine there is some kind of gridded layer over it, where the grids are in the shape of a rectangle or hexagon. To put it simply, binning is a handy way of aggregating points that fall within a grid (or bin) of such type of layer, and the quantities can be categorized and color coded the same way a choropleth map would be.
However, when the grid is in the shape of a hexagon it is much more efficient at aggregating the data around the bin center because hexagons are structurally more similar to a circle than a square, and also because a single hexagon is adjacent to more grids compared to a single square.
To demonstrate how much easier it is to visualize patterns in large point datasets using this method, USFS data for all fire occurrences recorded in all 50 states and territories of the US between 1984 and 2018 were used to showcase how hexbin mapping compares to mapping clustered point data.
This map was created with a bin size of 75 miles.
Click and drag to pan the map to see the full extent.
Use the slider to compare the point cluster map with the hexbin map.
At a glance, it is evident that the largest concentration of fires are in the midwest and south east regions of the country. The most prominent areas of the midwest are around Topeka, Kansas and Tulsa, Oklahoma. In the south west, Florida has an outstanding number of fires compared to any other states in the region.
The western third of the US has a moderate number of fires, with the most observations made in the area between Idaho and Nevada. Fires are the least prominent in Alaska, Hawaii, and Puerto Rico.
The areas with very high and moderate fire occurrences generally have rich wildlife and a warm, dry climate which causes the vegetation in these regions to serve as kindling for fires. Other most common specific natural causes include global warming, lightning strikes, and volcanic eruptions. Human activity such as arson, discarded cigarettes and debris are also common causes.
Even so, it is extremely alarming to see so many fires concentrated in the state of Florida. Let's take a look at a hexbin map that focuses on the area for a detailed analysis.
This map was created with a bin size of 25 miles.
Click and drag to pan the map to see the full extent.
Use the slider to compare the point cluster map with the hexbin map.
Fires are most prominent in the north west and the stretch along the central to southern areas of the state. In the north there is Apachicola National Forest, and in the south there is Osceola County which has Kissimmee Prairie Preserve State Park and Three Lakes Wildlife Management Area. These areas have lush vegetation, which explains why there are higher incidents of fires compared to other parts of the state.
Even after taking a closer look, it is evident that Florida has an outstanding number of fire occurrences compared to anywhere else in the US. What could be the reason for this? Simply said, it is the result of a combination of factors which makes the whole state highly susceptible to wildfires.
Florida has experienced a growing population over time which generates a higher risk due to more urban settlements created within its vastly abundant wildlands. Additionally, the frequency of lightning strikes in Florida is unparalleled in the nation, which coupled with extended droughts can create incredibly destructive fires on a large scale.
Point cluster maps are great at indicating specific trends at various scales, given that they are interactive. The clustering sizes and steps must be carefully considered as having too little range of clusters can make it harder to observe trends on a smaller scale.
Hexbin maps are great at illustrating the big picture trend of the data at a very large scale. This makes them a great tool for creating static maps but the analysis is done at a fixed scale and can be labor intensive to provide dynamic findings.
return to see all projects