Abstract:
Particulate matter, particularly PM2.5, plays a significant role in global air pollution and is a major contributor to serious health issues in both humans and animals. In Rwanda, rapid economic growth in recent years has adversely affected air quality, with PM2.5 becoming a key factor influencing urban pollution, particularly in Kigali. As a result, PM2.5 pollution has emerged as a critical environmental concern.
This study focuses on mapping PM2.5 concentrations in Kigali using nonlinear regression models, with the goal of assessing hourly PM2.5 levels, confirming the Air Quality Index (AQI), and identifying areas with high pollution levels over a five-year study period. Remote sensing data, alongside ground-based meteorological measurements, were utilized to estimate PM2.5 concentrations across various regions in Kigali. Both nonlinear and linear regression models were applied to predict PM2.5 levels and assess the associated health risks.
Although linear regression models are more commonly used, nonlinear models were chosen for their higher correlation with ground-level observations, resulting in more accurate PM2.5 concentration estimates. Additionally, the use of MISR AOD data in the nonlinear regressions provided more reliable results compared to MODIS AOD data.
In summary, the nonlinear regression model, which integrates remote sensing and meteorological data, offers a robust method for estimating PM2.5 concentrations in Kigali. This approach is crucial for addressing the city’s air pollution challenges. The findings revealed that PM2.5 pollution is primarily driven by industrial emissions, vehicle exhaust, and agricultural waste. To mitigate the growth of air pollution, strategies such as improving land use planning, enhancing industrial zoning, promoting healthier lifestyles, and expanding the use of renewable energy sources are recommended.
