Group13 Report

ISSS608 Visual Analytics and Applications

Group 13 : Report

Economic Growth And Climate Change

Proposal

Analysis Report

Poster

Application

Abstract

In recent years, there has been rising incidents of climate disasters such as Hurricane Irma in Haiti, intense monsoon rains and flooding in south-asia, Hurricane Harvey in United States among many more. This paper presents exploratory analysis of vital indicators on greeenhouse gas emissions, air pollution, and economic development of OECD member countries. The indicators data were collected from OECD data portal for years 2002-2012. To gather deeper understanding of envionmental impact caused by these emissions, we employed advanced analytics concepts such as geospatial analysis, hierarchical clustering, and time-series analysis, mainly driven by interactive visualisation in RShiny Visualisation Framework. As part of analysis, we uncover some startling facts such as US, Australia, Canada being the leading economies of all emissions, Norway and Iceland generate most amount of their energy supply from renewable energy sources, and Japan having reported large number of respiratory diseases in recent years.

Introduction

The Organization for Economic Co-operation and Development (OECD) is an intergovernmental economic organization with 35-member countries, founded in 1961 to stimulate economic progress and world trade. Most OECD members are high-income economies with a very high Human Development Index (HDI) and are regarded as developed countries. As of 2017, the OECD member states collectively comprised 62.2% of global nominal GDP (US$49.6 trillion) and 42.8% of global GDP ($54.2 trillion) at purchasing power parity.

However, has this economic progress come at a cost? During this period, the reported greenhouse gas emissions has shown a rising trend. For instance, there was a large decrease in GHG emissions in 2009 due to economic recession, further enforcing our fears that human activities could have drastic impact on climate change.

Overall trend analysis of OECD

This section provides high-level overview of trends shown in indicators for OECD as group.

• Economic Development

Most of the OECD countries have shown constant growth in GDP, with slowdown in 2009 possibly due to global economic recession. While few countries such as Greece experienced consistent decline from 2009 onwards, Turkey showed significant improvement in its GDP during same period. However, FDI inflows and outflows showed inconsistent behaviour, with US, UK, Australia, Japan, Italy heavily involved in economic activities.

• Greenhouse Gas Emissions

The overall trend showed gradual decline in greenhouse gas emissions for OECD countries. However, HFC (Hydrofluorocarbon) observed significant rise during this period. In September 2016, the New York Declaration on Forests urged a global reduction in the use of HFCs, reported to have impact on global warming.

• Air Pollution

Most of the air pollutants such as Carbon Monoxide, CO2, Nitrogen Oxide, Non-Methane Volatiles and Sulphur Oxides have shown declining trend over the years, except for unconventional spikes in years 2005 and 2008. Nevertheless, PM2.5 and PM10 pollutants showed consistent growth during this period. The continued exposure to PM2.5 and PM10 have shown drastic impact on human health.

Country-wise analysis of individual indicators

In addition to overall trend analysis, we analysed individual indicators to understand their underlying patterns for individual countries using heatmap. For remaining indicators, we recommend visiting our R Shiny dashboard with interactive features.

• Carbon Dioxide

Most of the OECD countries experienced sudden decline in CO2 levels in 2009. Though Japan and United States have consistently produced large amount of CO2, few European countries such as Hungary, Portugal and Sweden showed consistent improvements in CO2 levels.

• Methane

In addition to United States and Australia, Mexico is one of the largest producers of Methane gas. Few countries such as Estonia and Iceland, which heavily rely upon renewable energy sources, produced very less amount of Methane in comparison to other OECD countries.

• Nitrous Oxide

Most of the OECD countries have shown consistent improvement in N2O levels. In comparison to other greenhouse gases, nitrous oxide is considered to be less harmful.

• Sulphur Hexafluorocarbons

In addition to United States, Korea showed consistent increase in SF6 gas emissions, mostly attributed to waste generated by electronic industry. SF6 is considered to be most potent greenhouse gas, with a global warming potential of 23,900 times of CO2 when compared over a 100-year period.

• Perfluorocarbons

Canada, Japan, Korea, and US have consistently produced large amount of PFC, mainly produced by aluminium smelting industry.

Clustering Countries using HeatMaply in R

To differentiate groups of countries having shown similar patterns in terms of greenhouse gas emissions, air pollution, and economic growth, we performed multivariate analysis of selected indicators. Considering the small size of dataset (30 observations), hierarchical clustering seemed appropriate method. The approach taken was as follows –

• Aggregate data at country level, averaging measure values across years • Remove indicators with missing values (>50%) for most countries • Remove countries having missing values (>50%) for most indicators • Apply appropriate data transformation – scaling, normalisation, percent • Apply different distance methods, clustering methods to figure out optimal number of clusters with high average silhouette width

Below is the consolidated list of indicators after data cleaning –

• Methane • Carbon Monoxide • Carbon Dioxide • Forest, % of land area • HFC • NO2 • NON-Methane Volatiles • Nitrous Oxide • Perfluorocarbons • PM10 • PM2.5 • Renewable • Sulphur Hexafluorocarbons • Sulphur Oxides Based on Euclidean distance measure, average clustering method gives the highest optimal value. Below chart suggests 3 clusters based on average clustering method –

The 3 clusters can used visualised using heatmaply package in R as follows –

This exercise can be repeated for different distance methods, clustering methods, and number of clusters using our R Shiny dashboard.

Conclusion

As we saw in earlier sections, the visual analysis of indicators helped us better understand the overall growth of OECD countries, based on greenhouse gas emissions, air pollution, and economic growth. Clearly, the advanced economies such as US, Australia, Canada, Japan have had largest impact in terms of emissions and pollution. This has not only led to increase in respiratory disease cases in these countries, but also in neighboring countries such as Norway and Iceland.

The visual analysis also helped us find out hidden pattern in indicators for individual countries. For more information, please check our R Shiny dashboard.

Acknowledgements

We extend our sincere thanks to Prof. Kam Tin Seong for his enormous help and guidance on this project.