Dashboard Week Day 1 - The Change in Water Temperature around the Gulf of Mexico

I started dashboard week with the intention that I would not rush my planning process. I know that this can impact me, especially when I jump into something without truly understanding the data and the granularity of it. Therefore, I gave myself a good hour - hour and a half to work on this. I wrote out my questions as follows.

When - Time series. How much has the water temperature risen over the duration in years. A year on year analysis could be conducted here, or a simple percentage difference from the year previous.

Where - Where has the water temperature risen the most. I knew that this could be mapped but also noted that it could be achieved through a simple bar chart .

What - At what time is the water at it's warmest, shallow or deep? I wasn't entirely sure about this question or what in general. I noted that it was a maybe in terms of exploration.

How - Compare measures, add different levels of detail. Does anything stand out? With this question, I thought it may be interesting to look at the depth of the water and temperature. I wondered if there were any correlations. I know that the deeper the water, the colder it gets, but wondered if that was truly the case and what the data showed us.

Who - I thought it may be interesting to look at the organisations and the % of samples they conduct however, again, was unsure how to tie this back to the question.

With this plan in mind I thought to look at the data and figure what to exclude and drill down to as it was a very large dataset. In the end, I looked at 2004 - 2014. I have to say that the dataset was still very large and causing issues in terms of speed and load time. I then extracted the data which took several minutes. I thought this would help with the speed but it was still very slow. After tableau freezing, closing and giving me error messages, I finally got to a stage where I could start dragging fields in. I realised very quickly that there were some extreme values, which I then excluded. This caused tableau to freeze once more and so I had to start again with my exclusion process. After an incredibly frustrating morning, I had to get myself together and produce something to present in the 2 and a half hours I had remaining.

Below is my final output.

Link to viz: https://public.tableau.com/profile/alisha7755#!/vizhome/Dashday1/Dashboard1

I didn't have much time to really even think back to my questions and see how they answered my main question about the change in water temperature. I would have loved to supplement this with extra data to say what the change in water temperature was doing to the environment. I'd also have loved to have a fact card with core statistics, supplemented with information. The water is getting warmer - but why and what is the impact that it is having on us. These were the issues I truly wanted to dig deeper into. Time was something that I really struggled with. The issues I had with tableau really threw me off my thought process and plan.

I guess, in the end, I did produce something to present, even if it wasn't my finest work. However, I need to become better with time and understanding that unfortunately, things do go wrong, and it's okay to ask for help in those cases. Taking a step back to re-evaluate is also okay. Especially when things don't work as you think they would.

Author:
Alisha Dhillon
Powered by The Information Lab
1st Floor, 25 Watling Street, London, EC4M 9BR
Subscribe
to our Newsletter
Get the lastest news about The Data School and application tips
Subscribe now
© 2025 The Information Lab