kepler.gl is a great way to experiment with mapping. It’s easier to learn than full GIS (Geographic Information System) software which have an overwhelming amount of functionality, and there’s no software to install. You can find it on the web at https://kepler.gl/.
kepler.gl has some included data you can explore to learn to use the website. Go to kepler.gl and click the blue “GET STARTED” button in the middle. On the popup window, click “Try sample data” near the top right. There are a number of datasets to explore, but I suggest starting with California Earthquakes one because it includes both data and location information.
When you load the California Earthquakes set, you’ll see a map with purple points of different shades and sizes. When you mouse over those points, a popup will tell you statistics about the earthquake it represents (date, location, depth, and magnitude).
In the menu to the left, you can see the data that the map represents labeled “new dataset.” If you click on “new dataset,” a window will appear that shows you what data is included. You’ll see that the data from mousing over the map points is in this table, and the “Latitude” and “Longitude” fields are what kepler.gl uses to create points on the map. To map your own data, you’ll need to create a similar spreadsheet (more on that later). For now, close the dataset table with the ‘X’ in the upper right.
kepler.gl lets you make maps based on data, and even though this example already has data loaded, we can still experiment with different representations. When you load the California Earthquakes example, the data has already been added to the map as the purple points we explored, which is shown in the left menu under “Layers.” The earthquake data is simply labeled “Point,” but you can change that to a more meaningful name by double-clicking. I decided to rename this layer “Earthquakes.” You can also toggle hiding the layer by clicking the eye icon. To change the way that this layer appears on the map, click the ‘v’ on the right-side of the layer bar, and display options will appear. You can change the “Fill Color” to change the color of the earthquake points on the map, and you can add an outline and choose its color. If you click the three stacked dots to the right of “Fill Color,” you’ll see that “Color Based On” is set to “Magnitude.” This means that the color of circles for this layer on the map varies by the Magnitude field in the data. You can try changing that value to another field like “Depth” to compare where the deepest and shallowest earthquakes occur in California. As you make changes to the layer’s appearance, you’ll see the map update with your choices.
You can also change the representation of a data layer in kepler.gl. When you expand a layer’s display options with the ‘v,’ the first option listed is “Basic,” which here was set to “Point.” Clicking on “Point” reveals other ways to represent the data. Because we have point data (single latitude/longitude coordinates) for each earthquake and because there are so many earthquake points so close together on this map, I suggest trying to visualize the data as with “Cluster” or “Heatmap.” Both of these help reveal geographical clumps of earthquakes, with Cluster expressing groups as circles whose larger radius reveals more earthquakes in the area and Heatmap showing groups via color. Both Cluster and Heatmap allow you to adjust the “Radius” to control how close individual data points have to be to be combined together.
As a historian, I often want to know when events happened relative to one another, and kepler.gl has some great functionality to show that. The California Earthquakes example starts with a representation of when earthquakes occurred open on the bottom of the map, labeled “DateTime.” This is one version of what kepler.gl refers to as a “Filter.” And you can use it to change what data appears on the map. When you load it, the map is only showing earthquakes that happened between January 1991 and March 2011. You can increase or decrease this range by dragging with the white bars that appear over the histogram (the bar-graph representation of when earthquakes occurred). You can also move the window by dragging from between the white bars. One of kepler.gl‘s neatest features is that you can easily animate this display. Clicking the play button in the bottom left will start the time slice you’ve specified moving over the data, and there are other buttons to change its speed and other characteristics. When you play the filter, you’ll see the map animate to show the earthquakes that happened at that time. To see, add, and modify filters, click on the funnel icon near the top of the menu on the left side, just under the “kepler.gl” text.
To map your own data, you’ll need to create a spreadsheet in a csv format (you can save Excel and Sheets documents as csvs via Save As…) that includes latitude and longitude columns that locate your data geographically. Unfortunately kepler.gl does not provide an easy way to do this. There are, however, a number of services that can help you get latitudinal and longitudinal coordinates for places you want to map. https://www.mapcoordinates.net/en allows you to search for locations and get the coordinates for any point by clicking on the map. You can do the same by right-clicking in Google Maps. With either of these methods, you’ll need to copy the coordinates into your spreadsheet. To do more than one location at a time, I recommend https://geocode.localfocus.nl/, and a list of other options can be found here.
When I tried kepler.gl for the first time, I used some WPA Slave Narrative data prepared by Stephen Robinson. I plotted both the locations of where the interviewees were enslaved (in red) and the locations of their interviews (in blue). I then added an “Arc” layer that connected these points to show the movement of the interviewees since their enslavement. I matched the color of the “Source” and “Target” to the enslavement and interview locations to provide a visual link representing the freed people’s movement. I also showed this map in 3D (with the button at the top right) to make the arcs easier to follow. You can preview my map above or explore it interactively here: https://seanmsmithphd.net/maps/kepler.gl.html.
