The research team, led by Francisco Fortuné Olou of the University of Chinese Academy of Sciences and the Université d’Abomey-Calavi, together with Kpèdétin Aklounontin Karen Cintia Ahouandogbo and Kodjo Apelete Raoul Kpegli, divided the entire corridor into 2,589 grid cells of one square kilometer each. For every cell, they compiled annual fine particulate matter (PM2.5) concentrations for the period 2001 through 2022, drawing on established global satellite-derived PM2.5 products that combine aerosol optical depth measurements from sensors such as MODIS, MISR, SeaWiFS, and VIIRS with chemical transport modeling. A second, more focused analysis aligned air and water data for the shorter window of 2018 through 2022, when both high-quality PM2.5 estimates and Sentinel-2 optical imagery were available.

The headline numbers are sobering. Across the full 22-year record, the corridor’s mean PM2.5 concentration was 32.17 micrograms per cubic meter, rising slightly to 32.70 micrograms per cubic meter during the 2018-2022 assessment window. That is more than six times the annual guideline value of 5 micrograms per cubic meter set by the World Health Organization in its 2021 global air quality guidelines. Perhaps more striking, however, is what the researchers did not find: no corridor-wide linear trend in PM2.5 was statistically supported over the study period, and after applying a false-discovery-rate correction to account for the thousands of cells tested simultaneously, not a single grid cell retained a statistically significant trend. In other words, the air pollution burden in this corridor is high, persistent, and remarkably flat rather than clearly worsening or improving, which itself is a critical finding for policy.

On the water side, the team narrowed its attention to 205 grid cells that satisfied strict criteria for mapped surface water and sufficient valid satellite pixels. Lake Nokoué, a shallow, urbanized brackish lagoon sandwiched between Cotonou and the Atlantic coast, is notoriously turbid and nutrient-rich, shaped by domestic wastewater inputs, aquaculture enclosures known locally as acadjas, and seasonal exchange with the sea through the Cotonou Channel. To gauge optical water quality remotely, the researchers evaluated Sentinel-2-derived indicators, most notably the Normalized Difference Turbidity Index, a band-ratio metric that exploits how suspended sediments change the reflectance of red and near-infrared light, and a red-edge or red band ratio associated with chlorophyll-a, the photosynthetic pigment that signals algal biomass and eutrophication.

Crucially, the satellite indices were not taken on faith. The team validated them against field measurements of turbidity and chlorophyll-a collected at 19 monitoring stations across the corridor’s waters. The Normalized Difference Turbidity Index showed a positive association with measured turbidity, with a Spearman rank correlation coefficient of 0.488 and a station-cluster 95 percent confidence interval running from 0.369 to 0.590. The red-edge or red ratio tracked chlorophyll-a with a Spearman coefficient of 0.432 and a confidence interval of 0.295 to 0.544. These are moderate but meaningful correlations, in line with what remote sensing studies of optically complex inland and coastal waters typically achieve, and they give the screening framework an empirical anchor that purely satellite-driven exercises often lack.

With both environmental dimensions quantified on the same grid, the researchers applied a co-occurrence criterion to flag cells where elevated PM2.5 concentrations and elevated water-optical signals appeared together. Using a 80th-percentile threshold for each indicator, meaning cells ranked in the top 20 percent on both air and water dimensions, the analysis identified six priority cells: four in the commune of Adjara and two in Porto-Novo. These are places where residents potentially face compounded exposure, breathing comparatively polluted air while living beside waters whose optical signature suggests elevated sediment loads or algal activity. The framework deliberately stops short of calling these cells polluted in an absolute sense. As the authors emphasize, the output identifies relative monitoring priorities rather than confirmed pollution or complete environmental vulnerability.

That humility is built into the method through sensitivity testing. When the co-occurrence threshold was relaxed to the 75th percentile, the number of flagged cells rose to 16; when it was tightened to the 90th percentile, the count fell to zero. This swing illustrates how sensitive such classifications are to threshold choices, a reality that less careful hotspot analyses often gloss over. It also explains a second design decision: unlike the composite vulnerability indices that dominate much of the environmental ranking literature, this framework deliberately separates the environmental classification of each cell from contextual layers such as population density, built-up land cover, and hydrological setting. Those layers matter enormously for interpreting results, the authors argue, but folding them into a single score obscures what is actually being measured and why a cell was flagged.

The technical architecture behind the study is as noteworthy as its findings. The entire workflow runs on Google Earth Engine, the cloud-based planetary-scale geospatial platform that has transformed what resource-constrained research groups can accomplish without local computing infrastructure. Surface water extents were defined using the global surface water dataset developed by Pekel and colleagues, land cover came from ESA WorldCover at 10-meter resolution, population counts from the GHS-POP multitemporal grid, administrative boundaries from FAO’s Global Administrative Unit Layers, and elevation from the Shuttle Radar Topography Mission. Because all of these datasets are openly and freely available, the framework is reproducible by any government agency, university lab, or NGO in the region, and the processed grid-level and commune-level outputs are available from the corresponding author on reasonable request.

For a corridor that concentrates a large share of Benin’s population, commerce, and fishing economy onto a narrow strip of land between a lagoon and the ocean, the practical implications are immediate. Air quality monitoring networks in West Africa remain sparse, and the region’s cities are repeatedly identified in global reviews as among the least adequately measured in the world despite bearing substantial air pollution health burdens. Field campaigns in Cotonou have previously documented the physical and chemical character of local particulate pollution, and nutrient budget studies have quantified eutrophication pressures on Lake Nokoué, but until now there has been no common spatial unit in which air and water pressures could be compared. The 1-kilometer grid provides exactly that, giving municipal authorities in Cotonou, Adjara, Sèmè-Podji, and Porto-Novo a defensible, data-driven shortlist of locations where ground-truthing instruments and enforcement attention would yield the greatest return.

The study’s limitations are candid and instructive. Satellite-derived PM2.5 estimates inherit uncertainties from aerosol optical depth retrievals and chemical transport modeling, particularly in regions with complex emission mixtures of biomass burning, traffic, and dust. Water-optical indices can be confounded by atmospheric effects, sun glint, and the extreme optical complexity of shallow lagoons, which is why field validation remained essential. The moderate correlation coefficients, the absence of significant temporal trends after multiple-testing correction, and the threshold sensitivity of the hotspot counts all reinforce the authors’ central message: this is a screening tool, a way of triaging limited monitoring resources across a complex urban-coastal landscape, not a definitive verdict on any square kilometer of Benin. In a world where low-cost sensors, open satellite data, and cloud computing are converging, the Cotonou-Lake Nokoué-Porto-Novo corridor may well become a template for how fast-growing coastal cities across West Africa and beyond can finally see their air and water problems on the same map.

Subject of Research: Grid-based screening of fine particulate matter concentrations and satellite-derived water-optical conditions to identify spatial co-occurrence hotspots in southern Benin

Article Title: Grid-based screening of fine particulate matter and water-optical conditions in Southern Benin: the Cotonou, Lake Nokoué, and Porto-Novo Corridor

Article References: Grid-based screening of fine particulate matter and water-optical conditions in Southern Benin: the Cotonou, Lake Nokoué, and Porto-Novo Corridor. (n.d.). https://doi.org/10.1007/s10661-026-15908-8

Image Credits: AI Generated

DOI: 10.1007/s10661-026-15908-8

Keywords: PM2.5, remote sensing, water quality, Lake Nokoué, Cotonou, Porto-Novo, Benin, Sentinel-2, turbidity, chlorophyll-a, environmental monitoring, Google Earth Engine