The study was conducted during a three-day dry-season campaign from 10 to 12 December 2025, a period chosen deliberately because dry road surfaces maximize the availability of loose dust that vehicles and wind can throw into the air. The team, led by Yemisi Aramide Tijani of Redeemer’s University with colleagues from Olabisi Onabanjo University, Redeemer’s University, and the Federal University of Technology, Ikot Abasi, selected seven paved and seven unpaved roads spanning major traffic arteries and quieter residential streets. At each site, measurements were taken at roughly two meters above the ground—the height of the human breathing zone—during both rush-hour and off-rush-hour periods, with seven replicate readings per period. In total, the campaign produced 296 valid observations, recorded using a calibrated Aeroqual Series 300 monitor operating on optical light-scattering principles, alongside a handheld WindMate weather station tracking temperature, relative humidity, and wind speed.

The numbers that emerged are striking. On paved roads, mean PM2.5 concentrations ranged from about 26 to 215 micrograms per cubic meter, while on unpaved roads they climbed from roughly 66 to nearly 248 micrograms per cubic meter. Coarse particles told an even more dramatic story: PM10 on paved roads spanned 79 to about 1,405 micrograms per cubic meter, but on unpaved roads reached as high as 1,871 micrograms per cubic meter, with a single extreme short-term reading of 2,742 micrograms per cubic meter recorded at one site. For context, the World Health Organization’s 24-hour guideline values sit at 15 micrograms per cubic meter for PM2.5 and 45 for PM10—orders of magnitude below what was measured. The highest concentrations clustered at busy junctions such as Adatan/Carwash and Elega/Bode-Olude, while the lowest values appeared on Okeero, a minor residential street with little commercial activity.

Statistical analysis confirmed that this spatial variation was not random. One-way analysis of variance revealed highly significant differences in both PM2.5 (F = 13.248, p < 0.001) and PM10 (F = 75.473, p < 0.001) across the fourteen sites, with the coarse fraction showing the strongest heterogeneity. Geospatial interpolation of the measurements produced concentration maps revealing distinct high-pollution zones in the central and southern portions of the metropolis, aligned with high-traffic corridors and mechanically disturbed surfaces, while the northeastern sector registered comparatively lower burdens. The authors caution that these interpolated hotspots are visual estimates based on only fourteen monitoring points rather than statistically confirmed pollution zones, but the broad pattern is clear: particulate pollution in Abeokuta is spatially clustered around key activity nodes.

One of the study’s most intriguing findings concerns the relationship between traffic timing and particle behavior. On paved roads, mean PM2.5 fell from 97.73 micrograms per cubic meter during rush hours to 61.04 during off-rush periods—the intuitive expectation that fewer vehicles mean cleaner air. But on unpaved roads the pattern reversed, with PM2.5 rising from 97.14 to 132.90 micrograms per cubic meter between the same periods. This opposing behavior produced a traffic-period by road-surface interaction that approached statistical significance with a substantial effect size (p = 0.069, partial eta squared = 0.249), meaning roughly a quarter of the explainable variance was tied to this interaction. The result suggests that on dusty roads, vehicle movement itself—through wheel-induced turbulence and mechanical resuspension of loose surface material—can dominate the particulate signal in ways that do not simply track the number of vehicles present.

To probe the origin of the particles, the researchers turned to the PM2.5/PM10 ratio, a diagnostic tool that distinguishes coarse-mode dominance from fine-particle contributions. Ratios below 0.5 indicate that coarse particles—typically generated by mechanical processes such as road-dust resuspension—predominate, while higher ratios point toward combustion-related fine particles from exhaust, biomass burning, or secondary aerosol formation. Several of the highest-concentration zones in Abeokuta corresponded to low-to-moderate ratios, supporting the inference that mechanical dust generation along unpaved and heavily trafficked corridors is a major driver of the city’s particulate burden. The authors are careful to note, however, that the ratio is not source-specific: without chemical tracers such as black carbon or elemental composition data, specific emission sources cannot be conclusively identified.

Meteorology emerged as a secondary but meaningful influence on the city’s aerosol field. Wind speed varied significantly across sites (F = 4.517, p < 0.001), ranging from 0.27 to 2.10 meters per second, and showed positive—though not statistically significant—associations with both particle fractions on unpaved roads, consistent with wind-assisted resuspension of loose material. Relative humidity also differed significantly between locations, with values at many unpaved sites exceeding 70 percent, a level at which hygroscopic particle growth and enhanced deposition can alter atmospheric residence times. Temperature, by contrast, remained relatively uniform across the study area and showed no significant spatial variation, ruling it out as a primary driver of the observed particulate differences. On unpaved roads, PM10 was significantly and negatively correlated with temperature (r = −0.788, p = 0.035), while fine and coarse particles were strongly co-varying on both road types, with correlation coefficients of 0.967 on paved and 0.927 on unpaved roads.

When the measured concentrations were converted into Air Quality Index values using the United States Environmental Protection Agency’s standardized method, the results painted a sobering picture of public health risk. PM2.5 AQI values ranged from 81 to 265 on paved roads and 156 to 298 on unpaved roads, spanning categories from moderate to very unhealthy for all population groups. PM10 fared worse still, with AQI values reaching beyond 500—the top of the scale—at multiple locations, corresponding to hazardous conditions. Most of the sampled roads, both paved and unpaved, fell within the very unhealthy to hazardous range, implying significant risks for children, the elderly, people with respiratory or cardiovascular conditions, and roadside vendors who spend extended hours exposed to traffic corridors. The authors emphasize that these are short-term dry-season roadside measurements rather than formal 24-hour guideline exceedances, but the magnitude of the values leaves little room for comfort.

The findings carry direct implications for how rapidly urbanizing African cities manage their air. The study’s practical recommendations center on progressively paving or chemically stabilizing heavily trafficked unpaved roads, deploying dust suppression during prolonged dry periods, enforcing lower vehicle speeds on dust-prone routes, and improving roadside cleaning practices in ways that avoid dry sweeping, which itself resuspends dust. Strengthened vehicle-emission inspection along major corridors is also urged. More broadly, the work demonstrates that road-surface condition, traffic activity, and meteorology must be considered together rather than in isolation when assessing urban particulate pollution—a lesson with resonance well beyond Nigeria, given that unpaved-road dust emissions are a recognized problem across the developing world. The authors acknowledge the limitations of their short campaign: no collocation with reference-grade instruments, no direct measurement of road-surface silt loading or moisture, and no continuous multi-day monitoring. They call for future work incorporating wet- and dry-season sampling, 24-hour measurements, and vehicle classification to build a fuller picture. For now, the message from Abeokuta is unambiguous: what a road is made of may matter as much as what drives on it.

Subject of Research: Spatial variability of PM2.5 and PM10 particulate pollution across paved and unpaved urban roads and its meteorological controls in Abeokuta, Nigeria

Article Title: Spatial variability and meteorological associations of pm₂.₅ and pm₁₀ across paved and unpaved roads in Abeokuta, Nigeria

Article References: Tijani, Y. A., Olukayode, O. O., Ojo, O. T., & Agbasi, O. E. (2026). Spatial variability and meteorological associations of pm₂.₅ and pm₁₀ across paved and unpaved roads in Abeokuta, Nigeria. Discover Cities, 3(1), Article 191. https://doi.org/10.1007/s44327-026-00371-4

Image Credits: AI Generated

DOI: 10.1007/s44327-026-00371-4

Keywords: particulate matter, PM2.5, PM10, unpaved roads, road dust resuspension, air quality index, Abeokuta, Nigeria, urban air pollution, meteorology, spatial interpolation, dry season