Introduction

Rapid urbanization in Sub-Saharan Africa has intensified air quality challenges in cities such as Kampala and Fort Portal. Increasing traffic volumes, informal industrial activities, household biomass use, and population growth contribute to elevated and spatially heterogeneous PM2.5 concentrations. Despite growing deployment of low-cost sensors, understanding how meteorological conditions and temporal persistence shape observed pollution levels remains limited.

Data and Methods

This study uses calibrated hourly PM2.5 measurements from established monitoring networks in Kampala and Fort Portal. Meteorological data, including temperature, relative humidity, wind speed, and rainfall, were integrated with air quality observations. Multilinear regression models were developed separately for each city, incorporating both contemporaneous meteorological variables and lagged PM2.5 terms to account for temporal autocorrelation.

Results

The regression results show that lagged PM2.5 is a significant predictor of current concentrations in both cities, indicating strong pollution persistence. Relative humidity exhibits a consistent positive association with PM2.5, while wind speed and rainfall demonstrate pollutant dispersion and washout effects that vary by city. Differences between Kampala and Fort Portal highlight the role of urban form, emission intensity, and local meteorological regimes.

Discussion

The inclusion of lagged pollution terms substantially improves model performance and provides insight into short-term pollution carryover. City-specific responses to meteorological variables emphasize the need for localized air quality models rather than uniform regional assumptions. These findings are particularly relevant for developing early-warning systems and supporting data-driven urban air quality management.

Conclusion

Meteorological conditions and temporal dependence play a critical role in shaping PM2.5 dynamics in Kampala and Fort Portal. Multilinear regression offers an interpretable and policy-relevant framework for disentangling these effects, supporting improved forecasting and targeted mitigation strategies in African cities with expanding monitoring networks.

Multilinear Regression Analysis of PM2.5 in Kampala and Fort Portal Cities: Effects of Meteorological Factors and Lagged Pollution

Abstract

Content

Keywords

Resources