Let’s get stuck into Day 3 of Dashboard Week!

The Data

Today’s topic was about historical tornado tracks from 1950 – 2018 using data from National Oceanic and Atmospheric Administration – NOAA.

The Process

I didn’t do any data prep with this week’s data, so I was able to get stuck in straight away. However, I quickly encountered some issues when trying to work out the granularity of the data. At first I assumed that every row was a tornado, but then I realised this was not the case. After discussing this with my cohort, I decided to do a “count distinct” on a field which returned the total number of tornadoes for each year.

When I was exploring the data, I noticed that Texas experiences more tornadoes than any other state. Upon further research which confirmed this, I wanted to dig deeper into this. Why does Texas experience more tornadoes? What are the consequences of this? Where are the hotspots? When I had these questions in mind, I started building out my charts to tell the story.

The Dashboard

I wanted to stick with a simple colour scheme, and I was inspired by this viz by Diego Parker (which is amazing)! I opted for yet another longform viz, as I think it helps to tell a story and allows for each section of the dashboard to be seen individually before moving to the next.

I decided to add some BANs at the top with important figures from the dataset. I also wanted to show the comparison between Texas and other states in terms of the number of tornadoes, and I think the first chart highlights just how many more do in fact occur there. I tried to find out more about key events which I annotated on both the "Injuries" and "Fatalities" charts. I thought it would be interesting to see tornado hotspots and where they were more likely to start in Texas, so I created a map (using mapbox) with the points plotted.

I had a bit more time left and I really wanted to know why exactly Texas experiences so many tornadoes. After some researching, I discovered the "Tornado Alley" and added an image of what this is to the end of the viz. If I had more time, I would have definitely looked into the temperature and humidity in Texas and found data regarding this to help with my analysis. Overall, it was very interesting to find out more about tornadoes in the US and narrow in on one specific location.