DS22 Dashboard Week Day 2 | Hail Storms in America

by Simon Evans

Day 2 looks at us analysing Hail Storms in America.  The data comes from the National Oceanic and Atmospheric Administration - NOAA and has been nicely formatted by the Oasis Hub here. NOAA has some additional spatial data that we need to use.

Our brief:

  1. Work alone unless you're stuck
  2. Prep the data how you see fit
  3. Include the spatial data (one or both of the data sets)
  4. Write a blog post
  5. The dashboards MUST be built with the Alteryx Reporting Tools. You can use Tableau to explore the data, but the final dashboard must be built in Alteryx.
  6. Presentations at 3pm

Firstly I brought in the .csv file in Alteryx. I noticed there were a lot of headers with strange acronyms so I spent some time understanding which each of these meant using the documentation available, re-named the fields to make more sense to me in Alteryx, and got rid of some which I didn't think I would need.

After about 30 minutes messing around with the spatial files I realised that I could just use one of these. My workflow is shown below:

After a bit of exploring the data, I realised Texas was the state with the highest amount of storms, so I decided to focus my analysis on Texas.

The first part of my flow uses big numbers to summarise the key stats over 1955-2019

I used a summarize tool to reduce my data set to the sum of all of the key numbers in the dataset, except for storm number which I used a count.

I could then use a report text tool to put all these in big number formats, and was then able to put them together in  a horizontal container, similar to how you might use in Tableau.

The next part was to look at the trend for number of hailstorms over time so I used another summarize tool to look at the count of tornadoes over years and then used an interactive chart tool to produce the following chart

I then wanted to look at the seasonality of hail storms in texas, and was able to do this by summarizing by month and creating the following view with another interactive chart

The next part was to look at the number of storms with different hailstone sizes, so i created a calculation based on the hail size field that put that record in a bin, which allowed me to produce the following view:

Given that most hailstorms had an average stone size lower then 2 inches, I wanted to filter those above 2 and create a map which showed these across the state.

Using the map report I was able to produce this:

The last part was to union all these together and place them in containers to create the following: