June 30, 2021
Data Vis, Tornadoes in CT, D3 & Observable

While working on a prior project, I learned some D3.js, a data visualization library, and I discovered ObservableHQ, a platform for data visualization. I recently completed my first ObservableHQ data vis project: Tornadoes in Connecticut.

There have been multiple tornadoes that tore paths through landscapes near me, in recent years. I know that greenhouse gas-caused warming is destabilizing the climates, I wanted to know if tornadoes are getting worse in my area.

weather.gov

Looking at data in a table does not provide a clear picture. That’s the beauty of data visualization: it is technology of insight.

Tornadoes in Connecticut at ObservableHQ

The scatter plot seemed to show that Tornadoes in CT are not getting stronger. But are they getting more frequent?

Tornadoes in Connecticut at ObservableHQ

The bar graph makes apparent that the data does not support the conclusion that there is a significant trend of increasing frequency.

That’s good news.

But what if the data is flawed? I was able to include multiple datasets, so that the user can switch between them and compare data from different sources.

If anyone wanted to wrangle an additional dataset, edit their own copy of the notebook, and add it in, ObservableHQ makes that simple. The charts will update the axes to fit the selected dataset.

ObservableHQ is nice to develop in. It is a visionary, innovative, and intuitive website. It is easy to make a notebook: you code it right on the page. It gives you virtually everything you need. It is a code editor, data flow runtime, host, and platform.

D3, the flexible JS library, was built by the same person who built Observable and ObservableHQ. They work together, seamlessly, and Observable can be used with other JS libraries, as well.

I was able to quickly hack out a unique, interactive, responsive data visualization that offers a vivid window into an important topic that was otherwise opaque.

Contact

If you have an idea for a data visualization project, please email me, especially if you have good data. If it is an impactful idea then I may be willing to do it for free. Browsing the charts at ObservableHQ will give you ideas about the kinds of charts that are possible. Virtually anything is possible with D3 & Observable.

Links

Share this page