Earth’s Oven Story

We all are aware of the fact that earth is burning up with the rising heat in our environment. Now, this rise in temperature can be due many reasons, but before deep diving into the reasons, there is still one curious questions. And to answer this question we have to start looking into the history because sometime, answers to the future lies in the past.

When and Which part of the world started warming up and which part is contributing to this overall global warming?

To answer this questions, we have leveraged a dataset from “NASA Goddard Institute of Space Studies”.
You can access the dataset via this link.

Before jumping right into the analysis part of the data, I would like to mention that we’ll be using Tableau Public as a visualization tool and to some extent for data munging as well.

Now, let’s jump right into the dataset and find out what exactly it has to offer us!

Step 1: What this dataset is all about?

Overall we have 20 columns in the dataset which talks mean temperature change in the respective month of the year between 1880–2020.
Below are a brief understanding about each of them.

1. [Hemisphere] — which part of the world are we referring to North, South or Global
2. ]Year] — years for which we have data points are available
3. Jan, Feb, Mar, Apr, May, Jun, Jul, Aug, Sep, Oct, Nov, Dec — months for which data points are available.
4. [J-D] — January to December
5. [D-N] — December to November
6. [DJF] — December, January, February
7. [MAM]- March, April, May
8. [JJA] — June, July, August
9. [SON ]— September, October, November

Step 2: Data Pre-Processing

Since we have data on monthly basis and spread out horizontally over different columns, we need to arrange them vertically so that we can easily plot them on the graph. To do that we have to pivot the data for all the month and for now hide the other combination month fields like J-D, D-N… etc.

Instead of the existing month combinations, we created our own combinations in Tableau itself, using it’s grouping feature.
So we created 3 categories, J-A (Jan-Apr), M-A (Mar-Aug), S-D (Sep-Dec).

Step 3: Analyze the anomalies in the dataset

To check whether are any outliers in the dataset and also in which particular part of the earth and majorly in which months in expectation to see major variations in the temperatures, we plot our data points using a “Boxplot”.

Plot 1: In the above plot we can easily see that temperature variations are majorly happening in Northern Hemisphere and that too majorly in the starting and ending of the year.

Seems like Northern Hemisphere is seeing a rather major shift in the temperature change and contributing the most in the global changes in temperature.
One more observation we have from this activity is, there are no outliers in our dataset, which is good to know information as well.

Step 4: When it all started to make a difference?

From above visual we can see the Northern hemisphere is contributing the most in increasing the global temperature, now, from the below visual we will see the how Global and Northern Hemisphere are moving on average over the years.

MakeoverMonday Viz. Vs. This Viz.

Visualization shown on MakeoverMonday is all about factor which is contributing the most in the increasing the effects of Global Warming, which is Green House gases.
What we are trying to understand here through this visualization is that when and where it all started to happen and how alarming the situation is given the numbers are too high compared to the previous decades.

I think through above visualization, we are getting a clear picture that, temperatures are rising with an alarming rate in Northern hemisphere and tipping point was in starting of1900’s (around 1920’s) and major variations in the temperature are basically in Winter seasons (Jan-Apr) and (Sep-Dec).