Home > Environmental studies essays > Population density is positively correlated with CO2 emissions

Essay: Population density is positively correlated with CO2 emissions

Essay details and download:

  • Subject area(s): Environmental studies essays
  • Reading time: 7 minutes
  • Price: Free download
  • Published: 15 October 2019*
  • File format: Text
  • Words: 1,797 (approx)
  • Number of pages: 8 (approx)

Text preview of this essay:

This page of the essay has 1,797 words. Download the full version above.

The Earth is not an unlimited natural resource. The CO2 emissions witnessed today are drastically higher than those in the past, and if this exponential increase continues then the Earth’s natural resources will dissipate before we do. To prevent this utter destruction of the Earth, the key factors resulting in increased CO2 emissions need to be identified. It is widely known that the consumption of fossil fuels is one of the leading factors of this global warming, primarily in more urban areas. In this study, I try to uncover the relationship between urbanization and CO2 emission rates. Furthermore, I look at how different population densities within these urban environments can alter CO2 emissions and the extent to which urbanization lessens the effect of population densities on CO2 emissions

Data and Methods

The data for this project was taken from the World Bank’s World Development Indicators (WDI) database. The WDI database contains a variety of information that includes economic, environmental, and social variables. This data comes from every country in the world, ranging anywhere from the year 1960-2016. However, each variable’s data differs by country, depending on the year the data was acquired. For this project, we will solely focus on the set of variables acquired in the year 2010 because there are nearly no missing values. The ones with missing values have been expunged from this project. The final dataset consists of 175 countries, with 42 countries having been removed.

For this project, to measure the CO2 emissions data had to be collected from a laboratory. The Carbon Dioxide Information Analysis Center at the Oak Ridge National Laboratory in Tennessee accomplished this task. The laboratory could use their instrumentation to measure the metric tons of carbon dioxide emissions per capita. By including this measurement of per capita, the overall effect of the population on CO2 emissions can be “controlled for” within this dataset. Table 1 specifies overall measures of this variable, and Figure 1 exhibits a histogram of the range of CO2 emissions determined. The mean amount of CO2 emissions per person was around 4.87, but Figure 1 shows that the distribution is heavily right-skewed. This implies that most people produce less than 10 metric tons of CO2, with a few environmental degraders who emit as much as 40 metric tons.

Urbanization, however, can alter these CO2 emission levels. For this study, urbanization was the percent of the population that lived in urban areas. This data was acquired from the United Nations. It is also important to note that for the context of this project “urban areas” refers to country-specific lands which can vary between countries. Figure 2 is a histogram that exemplifies the urbanization of the countries chosen for this study. The quite symmetric histogram and the mean urbanization of 56.19, implying that no significant outliers skew the data. This outcome means that nearly all types of urban environments are represented within this data set, allowing the results to pertain to a wider range of environments.

To measure the population density, people were counted per square kilometer of land and grouped into 1000s. To calculate this number, a populations size was divided by the country’s total land area. By undergoing this division, water bodies could be excluded and the results could be more accurately acquired. The histogram in Figure 3 shows the distribution of population density sizes within the sample used. It is clearly portrayed that almost all the populations had less than 3000 people. The population density data had a mean of 0.32, which corresponds to Figure 3’s results. Most people fell into this category, with only a few outliers of about 7,000 and 18,000 people.

In addition, I used OLS regression models to approximate the relationship between the key variables mentioned and the amount of CO2 emissions released. Furthermore, I incorporated multiple control variables in the models in Table 1. These control variables included overall gross, the number of exports, and the percentage of the population that was a “working age” (15-64 years old).


With all these variables combined, the main causes of higher CO2 emissions can be calculated. Figure 4 shows a scatterplot of the CO2 emissions per capita based on population density. The distribution of CO2 emissions resembles the distribution witnessed in Figure 3. The medians of the two are not too different which is why most the lower population density countries emit most of the CO2 emissions. The means of the two variables differ slightly more from each other, which makes sense due to the outliers in the population density data set. In contrast, when comparing urbanization effects on CO2 emissions the data is not so evenly distributed. The means and the medians of the two variables are drastically different, by at least fifty. This results in a right-skewed histogram as seen in Figure 2.

Most of all, the difference between the effects of population density and CO2 emissions can be seen in Figure 5. Figure 5 shows the bivariate relationship between urbanization and CO2 emissions per capita. Although there is still variability among the data, the scatterplot suggests that there is a positive correlation between urbanization and CO2 emissions. This positive relationship can be supported by the correlation coefficient which is 0.49. Within this data, there are outliers for CO2 emissions, but these data points are not strong enough to contradict the overall pattern of the data. From Figure 5, it is also apparent that the data follows a linear path.

The bivariate relationships mentioned above can also be altered by including independent variables and other control variables. This relationship can be exemplified by using the example of the percent of people in the sample of working age. This variable had strong correlations with both the amount of CO2 emissions (r = 0.62) and urbanization (r = 0.58). Therefore, the relationship between CO2 emissions and urbanization is likely because people within this working age range are the ones emitting the highest emission rates. This is in comparison to CO2 emissions being solely factored by urbanization levels, thus altering the data outputs and making them more realistic and logical.

In Table 1 you can see the results that were calculated from four separate OLS regression models that predict the amount of CO2 emitted. Model 1 and 2 merely show the effects of each variable on CO2 emissions, but the data this study cares about comes from model 3. In Model 3 the key variables population density and urbanization are evaluated together. The model predicts that, among people with the same urbanization frequency, a one unit increase in population density is associated with 0.38 less metric tons of CO2 emitted, on average. In comparison, the urbanization model predicts that, among areas with the same population density, a one percent increase in urbanization is associated with 0.14 more metric tons of CO2 emitted. The effects of urbanization and population density are not very far from each other. The main difference is that urbanization has a positive correlation with CO2 emissions, while population density has a negative one. This implies that more urban areas emit more CO2, while more populated areas emit less CO2.

Model 4 brings into account the other control variables. Within this model, the effects of the two key variables are increased noticeably due to the inclusion of controls for income levels, percent of working age, and export levels. The largest decrease can be witnessed in the population density data which dropped from -0.38 to -0.97 (almost 60% reduction in emissions). In comparison, the effect of urbanization on CO2 levels decreased from 0.14 to 0.017 (12% reduction in emissions).

Model 5 introduces interaction terms, pertaining especially to the relationship between income levels and the percent of urbanization. The inclusion of this interaction term enables the effect of income levels on CO2 emissions to vary based on the income level of the person in the dataset. For example, the 0.035 interaction term within Model 5 implies that people within the middle-income level have a less positive effect (0.140-0.035) on the amount of CO2 emitted than by those within lower income levels. This same correlation can be witnessed from people within the upper income-level range, they exhibit CO2 emissions 0.034 less. Both interaction terms are positive, suggesting that people living in more urban environments with higher income levels release less CO2 into the atmosphere. However, the interaction terms are not significantly different from each other, so I cannot be completely assured that there is a difference in CO2 emissions between people in the upper-middle classes and the lower classes.


Throughout the context of this report, the consensus is that population density is positively correlated with CO2 emissions, while urbanization maintains a negative correlation with CO2 emissions. Pertaining specifically to urbanization, higher income levels emit fewer metric tons of CO2 than lower classes. This could be due to environmental care being an expensive project and only those that can afford it can reduce their ecological footprint on the Earth. The effect of population density was still important, with decreased amounts of CO2 emissions with the addition of more control variables. The two key variables both decreased as you look right across Table 1, but urbanization levels never reach those of population density.

As for the interaction between urbanization levels and income levels, the correlation is quite faint. While Model 5 does reveal drops in CO2 emission levels, they are only by a mere 3.5% and that is nothing in the bigger scheme of things. Had this number exceeded even 10% then a correlation could have further been investigated. The results, therefore, suggest that more urbanized and lower class environments produce the most CO2 emissions. But if it were this easy then why hasn’t this problem been fixed yet? This idea leaves room for further research, due to the fact that every year CO2 emission rates change. To have the most up to date and accurate results, this study needs to be reciprocated as human populations continue to grow and expand their areas of consummation. With additional research, the included variables will change, but they should still relate to human density and land use. The inclusion of these variables will narrow the data, and hopefully, assist researchers in tackling the leading cause of the consistent increase in CO2 emissions.

Overall, the combined results implicate that CO2 emissions can be the result of both increased populations and urban environments. In contrast to the belief that the rich are using most of the Earth’s resources and thus abusing them, this statement was proved false. As mentioned above, the accessibility to money allows the upper class to invest in renewable resources and hybrid vehicles, limiting their CO2 emissions. This is a problem for lower class people because both examples are expensive and lower classes of people cannot afford them. In order for people to start reviving the Earth, and prevent further degradation, more affordable environmentally friendly methods need to be invented. The Earth is not an unlimited resource and at this rate, it will dissipate before we do.

...(download the rest of the essay above)

About this essay:

If you use part of this page in your own work, you need to provide a citation, as follows:

Essay Sauce, Population density is positively correlated with CO2 emissions. Available from:<https://www.essaysauce.com/environmental-studies-essays/2017-12-7-1512684723/> [Accessed 13-07-24].

These Environmental studies essays have been submitted to us by students in order to help you with your studies.

* This essay may have been previously published on Essay.uk.com at an earlier date.