A group called the Marine Mammals Exploring the Oceans Pole to Pole (MEOP) Consortium has been tracking oceanic data since 2004 by attaching tags to marine animals who migrate thousands of miles to find food. They are able to learn about their foraging habits but they can also collect data from the oceans–even in remote polar regions. As a result, they’ve been able to collect more than 300,000 temperature and salinity profiles.
MEOP has made this data available from its website for anyone to use, so after I learned about it from an article on the Discovery website, I wondered what the data could tell us if it were uploaded into Watson Analytics. If you’re wondering the same thing but are pressed for time, you can view this short video that summarizes what I found. But I hope you’ll stick around and read about my discoveries.
A little help from my friends
I worked with the Watson Analytics user experience and design team to study a subset of the data. We started with the data for Canada from April 2010 – January 2011. We also combined the data sets for Australia and China from January – September 2014. The data sets had to be converted to a .csv file before we could use it in Watson Analytics, but that’s a simple thing to do. Also, the data feature was used to create data groups to help make analysis a little easier.
(If you haven’t already signed up for Watson Analytics, you can do so here for free.)
Oh… Canadian seal?
The first visualization created shows the readings from the Canadian stations. After 34.5 psu on the salinity scale (x-axis), there is a strong positive correlation between salinity and temperature. Interestingly, this visualization looks like a marine animal!
Moving away from this format to a more conventional one, I looked at the temperatures between April 2010 and January 2011 and noted a big difference between April and May.
A sea (animal) change for China and Australia
Were these anomalies? What about the China and Australia group in 2014? The visualizations told me a similar story, right down to the form of yet another sea creature.
Drilling down even more, I examined China’s salinity and temperatures. Imagine my surprise when Watson Analytics provided me with another data visualization shaped like a swimming ocean animal. I also was able to discern that China has the deeper readings of the two countries.
Also, the readings from January to April created a big cluster that looked like a sea turtle hiding in his shell.
A bigger picture but without the animals
I used Watson Analytics to create more traditional visualizations such as bar graphs and a spiral graph and learned that depth and temperature affect salinity for the years I analyzed. It would be interesting to examine a much larger subset of data—or even all of it—to see if this plays out for more years and in other countries and regions in the world. You could also plug this data into Watson Analytics to see if salinity has increased since 2004 or to see if depth, salinity, and temperature are changing the foraging patterns of marine animals.
The amount of data available to the public from the web increases every day. Watson Analytics can be used to identify patterns in that data that inspire further investigation. What will you investigate?
Get started on your own data journey
If you haven’t used Watson Analytics yet, now’s a great time to start by signing up for your free account.
Go pro and get more of what you love about Watson Analytics. View the Introduction to Watson Analytics Professional video to learn more.