Impact of Forest Coverage on Urban Air Quality: A Global Analysis¶
Author: Avsar Vora¶
+Table of contents¶
-
+
- Introduction +
- Dataset Overview +
- Data Load and Data Preparation +
- Forest Area vs Air Quality Analysis +
- Conclusion +
+
Introduction¶
This research delves into the complex relationship between the forest areas of various countries and environmental factors, specifically focusing on Urban Air Quality metrics such as AQI, CO levels, and Ozone AQI, among others. Through comprehensive dataset analysis, the study seeks to identify patterns and correlations that highlight the broader impact of forest coverage on climate resilience across different nations.
+The motivation for this research stems from the urgent need to understand the role of forest coverage in mitigating environmental challenges, particularly in urban areas where air quality issues are most pronounced. Forests play a crucial role in carbon sequestration, pollutant filtration, and overall climate regulation. By investigating the correlation between forest areas and urban air quality metrics, this study aims to provide valuable insights that can inform policy decisions, promote sustainable urban planning, and enhance climate resilience. Understanding these relationships is critical in the context of global environmental changes and the increasing pressure on urban ecosystems to adapt and thrive.
+Goals¶
The main objective is to examine and illustrate the connections between:
+-
+
- World's Forest Area Ratio Change since 1992 +
- Forest Area Ratio Change per Country +
- Top 10 Countries with Expanding Forest Area Ratio +
- Top 20 Countries with Most Forest Area Ratio (2021) +
- Comparision of contries Air Quality with Forest cover (2021) +
+
Dataset Overview¶
Datasource1: World Forest Area¶
-
+
- Metadata URL: https://www.kaggle.com/datasets/webdevbadger/world-forest-area/data +
- Data URL: https://www.kaggle.com/datasets/webdevbadger/world-forest-area/download +
- Data Source & Data Type: Kaggle - CSV +
- License: Creative Commons Attribution 4.0 International License. +
The dataset comprises 34 columns, encompassing detailed information such as the country name, country code, and the percentage of forest area as a proportion of the total land area. The dataset spans from 1990 to 2021, with individual columns representing the annual percentage of forest area for each year within this period.
+Datasource2: World Air Quality Index¶
-
+
- Metadata URL: https://www.kaggle.com/datasets/adityaramachandran27/world-air-quality-index-by-city-and-coordinates/data +
- Data URL: https://www.kaggle.com/datasets/adityaramachandran27/world-air-quality-index-by-city-and-coordinates/download +
- Data Source & Data Type: Kaggle - CSV +
- License: Creative Commons Attribution 4.0 International License. +
This dataset contains detailed information on various countries, including the number of cities, their geographic coordinates (latitude and longitude), and different air quality indices recorded in 2021.
++
Data Load and Preparation¶
+Import Packages¶
+import os
+import subprocess
+
+import matplotlib.ticker as mtick
+import numpy as np
+import pandas as pd
+import plotly.express as px
+import pycountry
+import seaborn as sns
+from matplotlib import pyplot as plt
+from sqlalchemy import create_engine
+
+plt.style.use('ggplot')
+percent_formatter = mtick.StrMethodFormatter('{x:,.0f}%')
+percent_d_formatter = mtick.StrMethodFormatter('{x:,.00f}%')
+Data Load: Load data from sqlite database for Forest Area and Air Quality Index¶
+# If database doesn't exist, run data_collection pipeline.
+if not (os.path.exists("../data/processed/world_air_quality_data.sqlite") and os.path.exists(
+ "../data/processed/world_forest_data.sqlite")):
+ subprocess.run(["./data_collection.sh"])
+
+# Create sqlite engine to make connection
+air_quality_engine = create_engine(f"sqlite:///../data/processed/world_air_quality_data.sqlite")
+forest_engine = create_engine(f"sqlite:///../data/processed/world_forest_data.sqlite")
+
+# Load data from sqlite to pandas data frame
+air_quality_df = pd.read_sql_table('world_air_quality_data', air_quality_engine)
+forest_df = pd.read_sql_table('world_forest_data', forest_engine)
+air_quality_df.head()
+| + | Country | +AQI Value | +CO AQI Value | +Ozone AQI Value | +NO2 AQI Value | +PM2.5 AQI Value | +
|---|---|---|---|---|---|---|
| 0 | +Afghanistan | +86.333333 | +0.333333 | +42.000000 | +0.000000 | +86.333333 | +
| 1 | +Albania | +77.111111 | +1.000000 | +42.555556 | +0.555556 | +76.555556 | +
| 2 | +Algeria | +106.250000 | +4.000000 | +35.000000 | +25.750000 | +106.250000 | +
| 3 | +Andorra | +32.000000 | +1.000000 | +32.000000 | +0.000000 | +24.000000 | +
| 4 | +Angola | +85.000000 | +3.380952 | +23.190476 | +2.238095 | +82.523810 | +
forest_df.head()
+| + | Country Name | +1990 | +1991 | +1992 | +1993 | +1994 | +1995 | +1996 | +1997 | +1998 | +... | +2012 | +2013 | +2014 | +2015 | +2016 | +2017 | +2018 | +2019 | +2020 | +2021 | +
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | +Afghanistan | +1.852782 | +1.852782 | +1.852782 | +1.852782 | +1.852782 | +1.852782 | +1.852782 | +1.852782 | +1.852782 | +... | +1.852782 | +1.852782 | +1.852782 | +1.852782 | +1.852782 | +1.852782 | +1.852782 | +1.852782 | +1.852782 | +1.852782 | +
| 1 | +Albania | +28.788321 | +28.717153 | +28.645985 | +28.574818 | +28.503650 | +28.432482 | +28.361314 | +28.290146 | +28.218978 | +... | +28.646606 | +28.698558 | +28.750511 | +28.802464 | +28.802190 | +28.792062 | +28.791971 | +28.791971 | +28.791971 | +28.791971 | +
| 2 | +Algeria | +0.699908 | +0.696214 | +0.692519 | +0.688824 | +0.685129 | +0.681435 | +0.677740 | +0.674045 | +0.670350 | +... | +0.811675 | +0.814866 | +0.818057 | +0.821248 | +0.821248 | +0.815790 | +0.810332 | +0.814110 | +0.818309 | +0.822228 | +
| 3 | +American Samoa | +90.350000 | +90.180000 | +90.010000 | +89.840000 | +89.670000 | +89.500000 | +89.330000 | +89.160000 | +88.990000 | +... | +86.850000 | +86.700000 | +86.550000 | +86.400000 | +86.250000 | +86.100000 | +85.950000 | +85.800000 | +85.650000 | +85.500000 | +
| 4 | +Andorra | +34.042553 | +34.042553 | +34.042553 | +34.042553 | +34.042553 | +34.042553 | +34.042553 | +34.042553 | +34.042553 | +... | +34.042553 | +34.042553 | +34.042553 | +34.042553 | +34.042553 | +34.042553 | +34.042553 | +34.042553 | +34.042553 | +34.042553 | +
5 rows × 33 columns
+Data Preparation: Prepare data frames for Analysis¶
+# Filter world forest data for timeline visualization
+world_df = forest_df[forest_df["Country Name"] == "World"].set_index("Country Name").transpose().dropna()
+
+# Country vise grouping of data for visualization of Forest area ratio change per country
+countries = []
+for country in pycountry.countries:
+ countries.append(country.name)
+country_df = forest_df[forest_df["Country Name"].isin(countries)]
+c_diff_df = country_df.set_index("Country Name").transpose().diff().sum().reset_index(name="change")
+
+# Top 10 countries with expanding forest area data for visualization
+top_p_change_df = c_diff_df.sort_values("change", ascending=False).head(10)
+
+# Top 20 Countries with Most Forest Area Ratio (2021)
+top_df = country_df.sort_values("2021", ascending=False).head(20)
+
+# Combined data for heatmap visulization
+df_2021 = country_df[["Country Name", "2021"]].rename(columns={"2021": "Forest Area"})
+air_country_df = air_quality_df.groupby("Country")[
+ ["AQI Value", "CO AQI Value", "Ozone AQI Value", "NO2 AQI Value", "PM2.5 AQI Value"]].mean().reset_index()
+comb_df = pd.merge(df_2021, air_country_df, left_on="Country Name", right_on="Country").drop(columns="Country")
++
fig, ax = plt.subplots(figsize=(12, 5))
+world_df.plot(colormap="Accent", ax=ax)
+ax.yaxis.set_major_formatter(percent_d_formatter)
+plt.title("World's Forest Area Ratio Change since 1992")
+plt.legend("")
+plt.show()
+
+Forest Area Ratio Change per Country¶
The graph depicts the change in forest area ratio per country, highlighting the varying degrees of deforestation or reforestation across different nations. Notable trends and significant changes in forest coverage over the observed period are clearly visible.
+ax = c_diff_df.sort_values("change").plot.barh(x="Country Name", y="change", figsize=(5, 32), colormap="Accent")
+plt.title("Forest Area Ratio Change per Country (1980 - 2021)")
+plt.legend("")
+ax.xaxis.set_major_formatter(percent_formatter)
+plt.show()
+
+Top 10 Countries with Expanding Forest Area Ratio¶
The graph showcases the top 10 countries with the most significant increases in forest area ratio, emphasizing those nations that have successfully expanded their forest coverage. These positive trends highlight effective reforestation and conservation efforts.
+fig = px.pie(values=top_p_change_df["change"], names=top_p_change_df["Country Name"],
+ title="Top 10 Countries with Expanding Forest Area Ratio")
+fig.update_traces(textposition='outside',
+ textinfo='percent+label',
+ marker=dict(line=dict(color='#FFFFFF', width=2)),
+ textfont_size=10)
+fig.show()
+Top 20 Countries with Most Forest Area Ratio (2021)¶
The graph presents the top 20 countries with the highest forest area ratio in 2021, highlighting those with the largest proportion of land covered by forests. These countries play a crucial role in global biodiversity conservation and climate regulation due to their extensive forested areas.
+fig, ax = plt.subplots(figsize=(10, 5))
+top_df[["Country Name", "2021"]].plot.bar(x="Country Name", y="2021", colormap='Accent', ax=ax)
+plt.title("Countries with Most Forest Area Ratio (2021)")
+plt.legend("")
+ax.yaxis.set_major_formatter(percent_formatter)
+plt.show()
+
+Comparision of contries Air Quality with Forest cover (2021)¶
we are comparing with contries Air Quality using the World Air Quality Index by City and Coordinates dataset. Here we do see somewhat a significant correlation, especially with AQI, Ozone AQI, and PM2.5 AQI values.
+plt.figure(figsize=(5, 5))
+sns.heatmap(comb_df.select_dtypes(include=[np.number]).corr()[["Forest Area"]], annot=True, cmap='Greens')
+plt.show()
+
++
Conclusion¶
The analysis of 2021 datasets reveals a significant correlation between forest cover and air quality across various countries. Nations with higher forest area ratios tend to exhibit better air quality metrics, underscoring the critical role of forests in enhancing urban air quality and overall environmental health. This finding highlights the importance of forest conservation and reforestation efforts in mitigating air pollution and promoting sustainable urban environments.
+