Mapping Earthquakes… and Nuclear Tests?

Project Description

Mapping Earthquakes… and Nuclear Tests?

As you probably know by now, last night North Korea tested a nuclear bomb in defiance of international bans, as reported in this Guardian article.

Aside from the main story, what strikes us as interesting is the way the world found out about it: thanks to the machines that detect earthquakes. Indeed, a nuclear explosion is so powerful that it makes the earth shake, just like earthquakes do.

Naturally, we wanted to know more.

We found data from the US Geological Survey listing all earthquakes for the past week, and it does show, just before 3 am UTC on Tuesday, a 5.1 magnitude earthquake in North Korea:

us,c000f5t0,A,”Tuesday, February 12, 2013 02:57:51 UTC”,41.3015,129.0664,5.1,1.00,137,”North Korea”

But what surprised us most in this data was just how many earthquakes occur every week, as you can see from this interactive cluster map of earthquakes below:

It seems that the disasters that make the front page news are only a very small portion of the total. Thankfully, the vast majority of seismic events are much less powerful, largely harmless, and few non-geologists take much notice of them.

As you can see below in the screenshot of the data from the US Geological Survey, earthquakes actually happen remarkably frequently, and all around the world:

Screen shot of original CSV data

This kind of data, while it is a true treasure trove of information, is not particularly easy to make sense of. By visualizing it with Mapsdata though, we can start to understand it better.

Thanks to the cluster map view of our data above, we can quickly see the precise number of earthquakes that have happened in different locations around the world. With hundreds of quakes in the past week, the West Coast of the US seems to have been the most shaken.

Changing the view of our data to a heat map of earthquakes, it becomes clear that seismic activity is very localised. The vast majority of earthquakes happen in just a handful of regions, such as in and around California and Alaska, in the Caribbean  or near the Solomon Islands in the south-west of the Pacific Ocean:

Earthquake heat map

In order to explore a specific region in more detail, or even get more information about a particular earthquake, mapping our data with markers is the best option:

Finally, we can use a bubble map to display magnitude in addition to the location of each quake. Here, the size of the bubble is proportional to the magnitude, allowing us to quickly compare events. Clicking on a bubble displays the exact magnitude of the corresponding earthquake:

Feel free to explore this data visualization through the four different maps we created, and let us know what you think on Twitter (@Mapsdata).

Read on to discover how we created these maps and understand how to create your own visualizations.

Try now


How we created this visualization: Getting data

The first step in creating a data visualization is to find the data (click here for tips). In this case, we found it on the US government’s open-data portal: The precise page is here.
This page allows users to see and download a list of all earthquakes (above 1 degree of magnitude) that have happened in the past week. This great data set is updated in real time with each new earthquake reported.

US government data portal earthquake data

Structuring data:

The second step in order to create a data visualization is to structure the data correctly (click here for help on how to do this).
In this case, the data is available from in CSV format (comma-separated values), which looks like this:
Screen shot of original CSV data

Mapsdata supports this CSV format, so we could have used the original data, however, there was a lot of information there that was not relevant to our visualization, so we decided to give it a quick clean-up in a spreadsheet.

We simply copied all the data and pasted it in the spreadsheet. Then we selected it and used the “Text to column” tool (found in the “Data” menu in Excel) to put the different comma-separated values each in their own cell, which already looks a lot better.

Then we deleted the columns that were not relevant, and kept only the key data: date and time, latitude, longitude, magnitude and region.

Click the button below to download this data set:

Download Data

Visualizing data:

The third and last step is to visualize this with Mapsdata (click here for details on the different visualizations):

We simply connected to the Mapsdata app, clicked import and added our spreadsheet. In a few seconds we had our visualization.

As explained above, the four different ways to visualize this data each have their own advantages.

In a single glance, a heat map lets us make sense of where earthquakes are happening, while using clusters we can compare the actual number of earthquakes in each region. Thanks to markers, we can see the precise location of each event and interrogate each data point to display further information, like date and time or magnitude. Last but not least, the bubble map lets us display magnitude for each individual earthquake in addition to its location. We can thus easily get a sense of their relative strength, and click on each to know the exact magnitude.

Let us know what you think of this visualization via Twitter (@Mapsdata).

And of course you can create your own data visualizations on the Mapsdata online app:

Try now