Carbon dioxide from burning fuels causes global warming, a process capable of changing the world’s climate significantly.

As you can see from the graphs, the amount of carbon dioxide in the atmosphere has increased steadily over the past 150 years, and so has the average global temperature. Some of this is due to human activity.Along with other gases such as methane and water vapors, carbon dioxide is a greenhouse gas. It absorbs heat energy and prevents it escaping from the Earth’s surface into space. The greater the amount of carbon dioxide in the atmosphere, the more heat energy is absorbed and the hotter the Earth becomes.

 Results of global warming

A rise of just a few degrees in world temperatures will have a dramatic impact on the climate:

  • global weather patterns will change, causing drought in some places and flooding in others.
  • polar ice caps will melt, raising sea levels and causing increased coastal erosion and flooding of low-lying land – including land where major cities lie.

Data Collection

All the data for this analysis and visualization was collected from Links for the data are below:

Data transformation

Data for both Co2 emission and world population were collected from 1980 – 2011 but for our study we considered the data from 1991-2011. Co2 emission per metric ton ranges from 0-70 metric ton per year and it was very difficult for us to visualize it in Carto. So as to make the visualization simple, we derived some variables which made the visualization simple and also more understandable. Some of the derived variables are mentioned below:

  • Mean of Co2 Emission by Country – we took the average of Co2 Emission by each country and then assigned a value of 1 the Co2 emission in that year is more than the average emission of all the years and 0 otherwise.
  • Level_Score – We divided the countries by giving them a score from 1-10 as per the level of Co2 emission in a year, by giving a country emitting in range of 0-5, a score of 1 and a country emitting more than 50, a score of 10.

We automated the procedure of deriving a variable by writing a r script for both the variable derivation, so as to save the time.

Geo Spatial Analytics

For geospatial analysis, we used Carto and by changing the sql query and putting some filters in Carto itself, we came up with some very beautiful analysis. This analysis is done for the year of 2011 only.

From the graph below we can clearly see that only a very little part of world i.e.  Oman and United Arab Emirates are the only countries which are emitting Co2 at a very alarming rate in 2011 i.e., more than 50 metric tons a year and also more than any of the other countries.


Inspite of being such a small nation, we can see that Oman and United Arab Emirates are emitting more Co2 in the environment than many of the big nations of the world which emit less than 40 metric tons each year.



Point Visualization(Animated)

For Geovisualisation, we used 2 methods one by dropping points at each country so as to animate it and other by marking the whole country with a polygon which could not be animated because of the Carto constraints.


This visualization is animated and is showing the Co2 emission by each country from 1991-2011. We have used 2 layers in this case:

  • 1st Layer is made using Torque Cat which shows the level of Co2 is above or below the Country’s average Co2 emission. Green markers are to show that the Co2 Emission of the country has gone below the average and Red Markers are to show that the Co2 Emission has gone above the average.
  • 2nd layer is made using Simple wizard which places a marker on each Country showing the country’s name on hovering over the marker and the level of Co2 emission.

There was a need of using 2 layers because when we use Torque Cat the Infowindow is disabled and also there was a need to showing county’s name on hovering as it is not user friendly to zoom all the time to see which country’s marker is red/green. So we put a Simple Wizard layer over the Torque Cat layer to solve this problem.

  • We have put our custom legends and label so as to make the map and points more understandable.
  • We also changed the duration of animation in CARTOCSS to 30 seconds and the frame count to 20(as there are only 20 years), so that it can show visualization of each year just once and we kept torque data aggregation method to linear only as linear is more appropriate than cumulative method in this case. Then we also changed the color of each marker by using green for marker[value=1] and red for marker[value=2].

Polygon Visualization(Static)


This visualization is better than the point visualization, but this cannot be animated using Carto as animating polygon feature is not available in Carto. We were able to achive this polygons using the carto update query only which is mentioned in the data transformation technique mentioned above. But we can always change the SQL query or put filters using SQL to get this kind of visualization.

For this visualization, we derived a variable LEVEL_SCORE (described above) and defined different color for each category (green for the least Co2 Emitting country to Dark Red and Black for the most Co2 emitting Country).

  • We used category wizard to make this visualization by selecting Level_Score as the differentiating feature.
  • Used Infowindow to put country’s name and the level of Co2 Emitted on hover.
  • Changed the legend tag to put custom legends showing the range of Co2 emitted by each country.
  • Updated the CARTOCSS to change the color of each polygon from green to Dark Red.