No abstract

Air pollution is prevalent in cities and urban centers in developing countries including sub-Saharan Africa, but ground monitoring data on local pollution remain inadequate, hindering effective mitigation. We employed low-cost sensing and measurement technologies to quantify pollution levels based on particulate matter (PM 2.5 ), NO 2 , and O 3 over a 6 month period for selected urban centers in three of the four macroregions in Uganda. PM 2.5 diurnal profiles exhibited consistent patterns across all monitoring locations with higher pollution levels manifesting from 18:00 to 00:00 and from 06:00 to 09:00; while the periods from 00:00 to 05:00 and from 09:00 to 17:00 had the lowest levels. Daily PM 2.5 varied widely between 34 and 107 μg/m 3 over a 7 day period, well within unhealthy levels (55.5−150.4 μg/m 3 ) for short-term exposure. The inconsistent daily trend are instructive for multiple pollutant assessment to aid specific policy initiatives. The results also show inverse relations between seasonal particulate levels and precipitation, that is, R (correlation coefficient) = −0.93 and −0.62 for Kampala and Wakiso, R = −0.49 and −0.44 for the Eastern region, and R = −0.65 and −0.96 for the Western region. NO 2 monthly concentrations replicated PM 2.5 spatial levels, whereas O 3 exhibited inverse relations probably due to a higher retention time in lessurbanized environments. Both PM 2.5 and NO 2 correlated positively with the resident population. Our findings show significant spatiotemporal variations and exceedances of health guidelines by about 4−6 times across most study locations (with two exceptions) for longer-term exposure. This paper demonstrably highlights the practicability and potential of low-cost approaches for air quality monitoring, with strong prospects for citizen science. This paper also provides novel information regarding air pollution that is needed to improve control strategies for reducing exposures.

No abstract

With rapid urbanization, hazardous environmental exposures such as air, noise, plastic, soil and water pollution have emerged as a major threat to urban health. Recent studies show that 9 out of 10 people worldwide breathe contaminated air contributing to over 7 million premature deaths annually. Internet of Things (IoT) and Artificial Intelligence (AI)-based environmental sensing and modelling systems have potential for contributing low-cost and effective solutions by providing timely data and insights to inform mitigation and management actions. While low and middleincome countries are among those most affected by environmental health risks, the appropriateness and deployment of IoT and AI systems in low-resource settings is least understood. Motivated by this knowledge gap, this paper presents a design space for a custom environmental sensing and management system designed and developed to fill the data gaps in low-resource urban settings with a particular focus on African cities. The paper presents the AirQo system, which is the first instance of the design space requirements. The AirQo system includes: (1) autonomous AirQo sensors designed and customised to be deployed in resource constrained environments (2) a distributed sensor network that includes over 120 static and mobile nodes for air quality sensing (3) AirQo network manager tool for tracking and management of installation and maintenance of nodes, (4) AirQo platform that provides calibration, data access and analytics tools to support usage among policy makers and citizens. Case studies from African cities that are using the data and insights for education, awareness and policy are presented. The paper provides a template for designing and deploying a technology-driven solution for cities in low resource settings.

Background: There are major air pollution monitoring gaps in sub-Saharan Africa. Developing capacity in the region to conduct air monitoring in the region can help estimate exposure to air pollution for epidemiology research. The purpose of our study is to develop a land use regression (LUR) model using low-cost air quality sensors developed by a research group in Uganda (AirQo). Methods: Using these low-cost sensors, we collected continuous measurements of fine particulate matter (PM2.5) between May 1, 2019 and February 29, 2020 at 22 monitoring sites across urban municipalities of Uganda. We compared average monthly PM2.5 concentrations from the AirQo sensors with measurements from a BAM-1020 reference monitor operated at the US Embassy in Kampala. Monthly PM2.5 concentrations were used for LUR modeling. We used eight Machine Learning (ML) algorithms and ensemble modeling; using 10-fold cross validation and root mean squared error (RMSE) to evaluate model performance. Results: Monthly PM2.5 concentration was 60.2 µg/m3 (IQR: 45.4-73.0 µg/m3; median= 57.5 µg/m3). For the ML LUR models, RMSE values ranged between 5.43 µg/m3 - 15.43 µg/m3 and explained between 28% and 92% of monthly PM2.5 variability. Generalized additive models explained the largest amount of PM2.5 variability (R2=0.92) and produced the lowest RMSE (5.43 µg/m3) in the held-out test set. The most important predictors of monthly PM2.5 concentrations included monthly precipitation, major roadway density, population density, latitude, greenness, and percentage of households using solid fuels. Conclusion: To our knowledge, ours is the first study to model the spatial distribution of urban air pollution in sub-Saharan Africa using air monitors developed from the region itself. Non-parametric ML for LUR modeling performed with high accuracy for prediction of monthly PM2.5 levels. Our analysis suggests that locally produced low-cost air quality sensors can help build capacity to conduct air pollution epidemiology research in the region.