This project is one of the projects from the Udacity Data Analyst Nanodegree Investigate a Dataset lesson. I chose the Gapminder dataset for this project and was curious about two things:

• how carbon emissions has changed by country & region (continent) over the years,

• how food production index, agricultural land, life expectancy and renewable water vary with carbon emissions.

Introduction:

Carbon emissions make up the highest greenhouse gas emissions (GHG), which causes global warming and climate change. Carbon dioxide is released into the atmosphere by burning fossil fuels, respiration, volcano eruptions, etc. These emissions contribute to extreme weather/temperatures, low life expectancy, increased wildlife migration, disruption in food supply, and shrinking water supply.

Data Wrangling:

I gathered, assessed, cleaned and analyzed six raw datasets from Gapminder: carbon emissions per person, food production index, agricultural land, life expectancy, renewable water and population.

The wrangling steps taken are:

• dropped data from the years 1800 to 1961 and 2017 as most countries were missing data during those periods.

• used the melt function to unpivot the years from individual column to one column

• renamed specific column headers to more descriptive names

• classified each country into a region (i.e. continent)

• Merge all datasets into one dataset for analysis

Conclusion:

I analyzed the clean data and used Matplotlib and seaborn to visualize the results. Due to variability in the data, I used the moving averages.

• Carbon emissions per person have increased in all regions, with the Asian Region having the largest increase of 5 tonnes/person from 1961 to 2013. Oil-producing countries' presence could explain this and the fossil fuels emitted from the industry activities.

  • Kuwait, Qatar, Bahrain and Brunei make up most carbon emissions per person in the Asian region.

  • Qatar, UAE, Saudi Arabia, Brunei, and Bahrain carbon emissions per person has increased from 1961 to 2013.

  • Though Kuwait has the highest carbon emission per person, it has reduced its carbon emission by 3.17 tonnes from 1961 to 2013.

• Africa, Oceania and South America consistently have the lowest carbon emissions per person across all years, with Africa having the lowest carbon emissions of three regions. However, its population is the highest of the three areas. Africa's population is higher than Oceania for all years but higher than Europe during the last 10 - 12 years. A reason for the lower emissions in Africa could be less industrial presence in Africa than in more developed regions..

• There's been a decline in renewable water in all regions with Oceania and South America having the highest drop

• Food production index and life expectancy have increased in all regions.

• Except for Europe, all regions have an increase in agricultural land.

• Carbon emission is negatively correlated with food production index, agricultural land and renewable water and positively correlated with life expectancy and total population. However, none of these correlation coefficients are strong; life expectancy (coeff = 0.46) is the only variable with a moderately positive correlation with carbon emissions.