- Yuyao Chen (ychen856@wisc.edu)
- Yuang Zhou (yzhou377@wisc.edu)
I. Introduction
Along with the popularity gained by Tableau, increasing attention is now paid to the technique of Visualization, when it comes to Data Analysis. Visualization allows the illustration of the condition shown by the data in a direct, visible and intelligible way rather than the abstract numerical expressions. Visualization definitely provided a widely accepted channel of communication, especially when showing the data analysis report to those who are not familiar with statistics and math.
Sustainable energy, as a keystone of pollution reduction and countermeasure for global-warming, has been emphasized and discussed for several years. However, rather than generate an idea of sustainable energy implementation, it is of more importance to study the quantitative distribution and evolution of such a renewable source around the countries in today’s world, considering the difference with anomalies and uncertainties in these countries.
How is today’s renewable energy being implemented globally? This is the problem we want to give an answer to through visualization. A well-known form of visualization is the global map and different shades of colors to show the value of one index among different areas. In this report, the time-mannered evolution study on particular countries and the fixed time point comparative analysis around the world will be presented.
II. Introduction of the Database
This project is to study the recent development of renewable energy all around the world.
“Sustainable Energy for All” from World Bank
Description: The world bank website has provided a set of data from 84 different databases, concerning the 11 variables describing the development and application of sustainable energy within a scope of more than 10 years among over 259 countries.
Link: https://databank.worldbank.org/source/sustainable-energy-for-all#
Under this data set, Renewable Energy Share of TFEC (Total Final Energy Consumption) (%) is focused.
“World Development Indicators” from World Bank
Description: World Development Indicators (WDI) is the primary World Bank collection of development indicators, compiled from officially recognized international sources. It presents the most current and accurate global development data available, and includes national, regional and global estimates.
Link: https://databank.worldbank.org/reports.aspx?source=world-development-indicators
Under this data set, these series are focused:
- Electricity production from nuclear sources (% of total)
- Electricity production from oil, gas and coal sources (% of total)
- Electricity production from renewable sources, excluding hydroelectric (% of total)
- PM2.5 air pollution, mean annual exposure (micrograms per cubic meter)
- GDP (current US$)
- Fuel exports (% of merchandise exports)
III. Approaches to Answer the identified questions
a. Project Design and Flow Chart
In order to get as complete records as possible, the data from 2005 to 2014 are selected from our origin source. This selection is decided due to the missing data after 2015, or in some unstable areas in south east Asia and Africa around the 1990s and 2000s.
The first part of our analysis is a ranking of renewable energy’s global implementation through the comparison of the share it takes in the total final energy consumption. This ranking is supported by visualization on the world map: the degree of color showing the level of renewable energy share directly.
Then, several typical countries are selected for a more detailed study: The type of energy they consumed, and how the ratio of these types change over years, are studied. Since the implementation of sustainable energy is in fact influenced by the countries’ economy and policies, the G20 countries, a collection of twenty of the world’s largest economies formed in 1999, are worth further analysis. Here we didn’t include the European Union (EU) because EU’s members such as Italy, Germany and France are already on the list.
Another group of countries are selected as well after the ranking of the global renewable energy implementation. Realized that, due to the relatively simple and monolog structure of energy, the disadvantaged countries yet to be developed can easily gain high renewable energy share through one or two green plants. So, the countries who have good rank and not be a less-developed country are selected for further study as well.
In order to gain more information about the current situation, the correlation between the variables are also studied. And for the global countries, a clustering method will be conducted and visualized. It is possible to interpret these clustering and divide the counties into different groups. The characteristics for each group are discussed as well.
b. Selected Group 1: G20 Members
Argentina, Australia, Brazil, Canada, China, France, Germany, India, Indonesia, Italy, Japan, Republic of Korea, Mexico, Russia, Saudi Arabia, South Africa, Turkey, the United Kingdom and the United States.
c. Selected Group 2: Other Well-Performed Countries
Thailand, Canada, Switzerland, Iceland, Norway, Tajikistan, Pakistan, Sweden, Brazil, India, Finland, Vietnam, Indonesia and Denmark.
IV. Data Visualization result and interpretations
a. Global Map on Sustainable Energy Share
of Total Final Energy Consumption in 2005 with the Top 5 Countries
of Total Final Energy Consumption in 2014
The first interesting point we noticed is the dark color around mid Africa, which indicates high renewable energy share in total final energy consumption. Considering the relatively low development level around this area, the usage of sustainable energy is rather unexpectedly high. This phenomenon might be explained by the special electricity condition in Africa: Most countries rely on one or two hydro-power plants, (for example, Democratic Republic of the Congo* relies on the hydro-electrical plants on the Congo river to power the whole country).
*https://www.andritz.com/hydro-en/hydronews/hydropower-africa/democratic-rep-congo
Here in Figure 5, based on the selection group 2, shows how the selected top countries perform. The North Euro area is outstanding.
b. G20 Energy Type Evolution
Legends:
Oil, Gas, Coal, (Grey)
Nuclear, (Yellow)
Renewable Energy without Hydro, (Green)
Hydroelectricity and Others (Blue)
This Visualization result shows the energy types in 19 of G20 countries (The other one is the EU) and how it changed from 2005 to 2014. Considering that the development of Hydroelectricity has a strict geographical limit, hydroelectricity is divided from other renewable energies as an individual component.
In general, there is a trend of increasing sustainable energy share in most G20 countries, especially for Germany, Italy and the United Kingdom. Russia and Saudi Arabia seem to have fallen back in the past 10 years. One possible explanation might be the abundance of their natural resources and the importance of fossil energy business in their economy made it harder to apply the usage of sustainable energy.
Some key features we noticed from the visualization is about Japan. In 2014, all nuclear plants in Japan were temporarily shut down* for maintenance and security checks, which means zero nuclear energy consumed that year. The visualized data confirmed this fact. Similarly, Germany has reduced the proportion of nuclear power a lot in the 10 years from 2005 to 2014, just as its government claimed**.
*https://www.japantimes.co.jp/news/2015/04/13/business/2014-nuclear-power-output-zero/
**https://www.france24.com/en/live-news/20211231-germany-to-close-nuclear-reactors-despite-energy-crisis
c. Line Graph – Overall trend of Renewable vs. Renewable Electricity Production Among G20 Members
In this graph, we have the blue line represent the percentage of electricity produced from nonrenewable and the red line represent the percentage of electricity produced from renewable sources. This graph shows the overall trends of the percentage of electricity produced from renewable sources vs. nonrenewable sources. We can see that the percentage of electricity produced from oil, gas, and coal sources is slightly declining, whereas the percentage of electricity produced from renewable sources is drastically increasing over the decade.
As we can see from the graph, even though it shows an upward trend in renewable energy consumption, there only appears to be a slight decrease in nonrenewable energy consumption, meaning that most countries are not ready to forgo their reliance on traditional energy production (from oil, coal, etc.). Therefore, these traditional approaches would still be their top choice in terms of efficiency and cost-effectiveness.
d. Growth Rate of Renewable Energy Consumption among G20 Countries
These integrated tableau bar plots show the annual rate of growth in renewable energy consumption (% of total energy consumption) over the period of 2005 to 2014 (10 years) for each member of G20. Each entity of the categorical variable (countries) is represented by a row of bar-plots filled with the same color so that users can easily identify the trend for a country’s growth rate throughout the decade and also make comparisons between countries’ performances within the same year. By clicking on the bars in a country, users can get the information on the growth rate for a specific year. For example, the growth rate for the United States in percentage of renewable energy consumption from 2005 to 2006 is 0.0951. The graph also shows a reference line for the average growth rate of each year so users can easily measure the growth rate against the average among all 19 countries.
The bar-plot in the first column represents the annual growth rate of percentage of renewable energy consumption from 2005 to 2006, and it is arranged in a descending order with Germany, UK and Italy being the top performers with the highest growth rate and China, Turkey, and Saudi Arabia having the lowest growth rates. However, if we look at Germany specifically, it shows a very interesting trend with it starting off strong but showing a decade-long decline in growth rate. However, on the other hand, the rapid growth of the UK and Korea has outstripped increases in percentage of renewable energy consumption. We discovered that the growing stable of wind farms* was one of the main reasons for the UK’s renewable record. Also, the Korean government put more emphasis on requesting domestic businesses/companies to consume renewable energy due to its lack of natural oil reserves and coal supply and its ambition in ending the reliance on coal imports**. Note that Germany has long been one of the leading countries in green energy transition, and the decrease in its growth rate does not represent its overall decrease in green energy consumption – it could only mean that the rise of renewables share became more stable. 2005-2014 might not be the critical energy transition period for Germany developing its renewable energy technologies. Also, from the average reference line, we can see that 2012-2013 has the highest average growth rate in renewable energy uptake compared to other years, which could imply that the world was inclined to a more conducive environment for renewable energy technologies and investments.
*http://www.sd-commission.org.uk/data/files/publications/Wind_Energy-NovRev2005.pdf
**https://www.jstor.org/stable/26909944?seq=1#metadata_info_tab_contents
e. Correlation Matrix (HeatMap)
For the next step of our project, we want to find out what kinds of social or societal variables/indicators are associated with renewable energy consumption or production; in other words, we hope to uncover the characteristics of a country or an economy with higher renewable energy production and consumption. For this graph, it shows a Pearson correlation matrix between eighteen variables where the correlation coefficients are depicted by colors: color blue represents positive correlation whereas orange represents negative correlation, and the brightness of colors represent how strongly the two variables are correlated. For this graph, we are mainly studying the correlation between “percentage of renewable energy consumption”, “electricity produced from renewable sources”, “electricity produced from nonrenewable sources” with other social indicators, such as GDP, CPI, population growth (annual %), tax revenue (% of GDP), etc.
We noticed that there is a -0.607 degree correlation between “Renewable Energy Consumption” and “Starting a Business Score”, the latter measuring the easiness of starting a business in the particular country designated. It might be rational to explain that, if the specific country is in a fast developing period, while the cost, including electricity is low, the starting business will be very common and the difficulties will be relatively low. In such a period, taken the example of the United Kingdom in 19th century, Germany in 20th century and China in early 21th century, the structure of electricity supply is supposed to be mainly based on traditional energy sources, like fire power of oil, gas and coal, which can yield relatively concentrated and economically efficiency electricity to support the fast expanding business, as well as the supervision of environment protection is not simultaneously emphasized in such a special period discussed.
Another interesting finding is that “suicide mortality rate (per 100k population” is negatively correlated with “percentage of renewable energy consumption” and “electricity produced from renewable sources” with correlation coefficients of -0.404, and -0.307, which are relatively prominent compared to other variables. One possible explanation behind this finding is that high suicide mortality rate may be an possible indication to a high-stress and highly-competitive soceity with poor living standards/low quality of life, which might be also associated with high unemployment, more concerns about money, work, economy and job stability, and so on. These are more likely to happen in developing countries where the society would rather prefer efficiency and cost savings, and they need more reliable energy sources in order to provide more opportunities to their people. Therefore it’s less likely for them to transfer their energy source and invest heavily in green technology that produces sustainable energy.
Moreover, we found that “percentage of renewable energy consumption” and “account ownership at a financial institution or with a model-money-service-provider” (% of population with age 15+) is positively correlated with a coefficient of 0.411, making it the most prominent correlations between renewable energy consumption and other indicators. More people having financial accounts means high financial engagement and inclusion – more individuals and enterprises have access to useful and affordable financial products and services that meet their needs, such as savings, credit, insurance. These financial products are presented to them in a sustainable and responsible way, which could be an indicator for high accumulation of capital and production of goods and services – a strong economy as a developed stage of the country. It’s possible that during this stage of transformation into more sustainable approaches of development for the country, people would potentially have more money to invest and support the development of clean energy.
The strong positive correlation between “health expenditure (% of total GDP)” and “percentage of renewable energy consumption” at 0.371. Healthcare spending can lead to greater health opportunities, which can help enhance human capital and potentially increase productivity, boosting economic performance. More emphasis on healthcare would show the government’s inclination for a more humanitarian approach – health care’s prosperity measures the government’s enthusiastic mind on the caring of people’s health, which is a indicator of the improvement for people’s living condition and the matching developed stage of the country: people are caring more about advanced health rather than basic survival. In such a stage, is also the time point for the evolution of energy structure, from traditional based to, in a gradual conversion process, a healthier sustainable energy based system.
Additionally, we found that there’s a negative strong correlation between “female labor force (% of total label force) ” and “percentage of renewable energy”. We would associate having a larger percentage of the female labor force with a society that emphasizes gender equality and women empowerment. This more liberal approach on gender equality would be associated with a more liberal perspective on human welfare, and therefore it contributes to the likelihood of consuming more sustainable energy sources as opposed to non-renewable energy.
f. Clustering
The clustering data used is the dataset collected in 2014. The renewable energy share in total energy consumption is compared with three other indicators in order to get a direct view of clustering, the characteristics of clusters are discussed as well.
GDP v.s. Renewable Energy Share
GDP is a widely used measurement of development for a country. In this subsection, GDP and the percentage share of renewable energy consumption in total energy consumption are used as the variables for clustering. Global countries are clustered into four groups.
As the world’s biggest economic entities, China and the United States earned their place in one individual group that is marked in red. They both have outstandingly high GDP and relatively low renewable energy percentage (around 10%). This might be because the traditional energy industry has developed for many years in both of them, and renewable energy, although emphasized in many reports and policies, still has a potential market to discover and a long way to go in the process of replacing traditional energy.
The next cluster painted in blue has relatively higher than average GDP performance with a low renewable energy percentage. Typical countries in this cluster are Germany and Italy. We can see that among the major countries they have good performance in the implementation of renewable energy.
The third cluster painted in yellow shows moderate renewable energy percentage. In contrast, its average GDP is lower than the previous two. Typical samples are Sweden and Finland, who are both Northern European countries. They have really good performance among developed countries: 40% to 50% of the energy consumption is from renewable energy.
And the last cluster painted in green presents the lowest GDP index and the highest rate of renewable energy. These countries are most described as less-developed countries or human-resources-rich countries. The reason why they have such a high percentage is explained earlier: for most countries in this cluster, when considering their total demand, a single hydroelectric plant, at the appropriate juncture, is able to power most of the country.
PM2.5 Pollution v.s. Renewable Energy Share
PM 2.5 pollution is widely considered as a consequence of consuming major fossil energy. It is believed that the increasingly serious PM 2.5 pollution will stimulate the demand of clean renewable energy. However, it has to be stated that, raw plant fuels, like straw, are renewable, but their consumption produces considerable amount of PM 2.5 particles.
Sustainable energy also has its advantage when it comes to the pollution problem, which inspired this clustering study based on the PM 2.5 pollution index and renewable energy percentage. It can be seen that Iceland has a really good performance with high sustainable energy use and low air pollution.
Also, Canada is a special case when it comes to pollution and sustainable energy. Canada has *15% energy generated from nuclear power, which has low air pollution but is not considered as sustainable energy. That explained why Canada’s location is far left from Finland, Norway and Iceland on the cluster graph, as their similarity exists.
*https://world-nuclear.org/information-library/country-profiles/countries-a-f/canada-nuclear-power.aspx
Fuel Export v.s. Renewable Energy Share
If a country relies on fuel export* as one of its key economy industries, will it be harder for it to promote sustainable energy? In order to answer this question, a clustering based on fuel export proportion and renewable energy share is established.
The orange cluster has a strong characteristic of low renewable energy use and high fuel export. Qatar, Kazakhstan and the Russian Federation are all major oil/natural gas export countries. High fuel export indicates a high nature’s fuel resource in a country, which means that fossil energy will be competitive in price. Also, the nation’s economy relies on this business a lot and many employment opportunities are provided in related industries. And that might be a reason why the development of renewable energy is relatively below average.
*https://www.worldstopexports.com/worlds-top-oil-exports-country/
V. Summary and Comments
Notwithstanding the fact that some countries are still highly reliant on traditional energy and some countries are still facing it, the future development of sustainable energy is hopefully expected to be worthwhile. Our study has revealed that some developed countries with moderate GDP performance, like Germany, Finland and Italy, have better performances in renewable energy implementation. We are glad to find that some countries suffering from heavy air pollution have also realized the importance and the advantage of renewable energy and seek more applications. Another trend we find is that if a country has a high dependence on its natural resource and fuel export, their progress in promoting renewable energy will be limited.
We can conclude from the correlation analysis that the introduction of renewable energy is associated with other human rights affairs: like gender equality, emphasis on healthcare and social security, especially for those developed countries. Better development stage means more discussion and investment about sustainable energy subjects. In general, the overall global distribution of renewable energy share is still in an unbalanced and undeveloped stage, which requires more effort and resolution for further implementation.
Sustainable World
A data visualization present of the global sustainable energy implementation