evidoria

← Back to browse

Good practice Imported

A Calibrated Low-Cost PM2.5 Sensor Network Across Burkina Faso's Three Largest Cities

Burkina Faso · Ouagadougou · See the Burkina Faso profile · See the Ouagadougou profile

Evidence: Observational / pre–post Top 54% 67/100 · Ask Evidence Copilot about this practice

Burkinabé and Italian research teams installed 19 calibrated Clarity PM2.5 sensors across Ouagadougou, Bobo-Dioulasso and Koudougou, recording a one-year average of 46.7 µg/m3 — about nine times the WHO annual guideline — as a replicable low-cost air-monitoring model for Sahelian

19 sensors
Sensors installed (Nov 2021 – Nov 2022)
46.7 µg/m3
Overall average PM2.5 concentration (Nov 2021 – Nov 2022)
48.5 µg/m3
Ouagadougou average PM2.5 concentration (Nov 2021 – Nov 2022)
17–68 µg/m3
Range of corrected daily PM2.5 across sites

Details

Maturity
Pilot
Promoter
Université Joseph Ki-Zerbo, ENS & Université Nazi Boni environmental physics-chemistry labs and IRSAT-CNRST (Burkina Faso), with CNR-IBE (Italy)
Period
November 2021 – November 2022
Keywords
air quality, environmental monitoring, urban sensing, open data

Context

Most cities in the Sahel, including Burkina Faso's, have no reference-grade air-quality monitoring, leaving residents and city governments without data to act on pollution levels.

Objectives

Burkinabé university teams and Italy's CNR-IBE set out to test whether a network of low-cost, locally calibrated sensors could reliably fill this monitoring gap.

Activities

Nineteen Clarity optical PM2.5 sensors were installed at 13 sites in Ouagadougou and three each in Bobo-Dioulasso and Koudougou, with readings corrected using a Gaussian Mixture Regression model trained on a two-month co-location against a TEOM reference monitor; the network ran continuously from November 2021 to November 2022, alongside companion deployments in Niger and Guinea to compare performance across contexts.

Results

Corrected mean daily PM2.5 concentrations across Burkina Faso sites ranged from 17 to 68 µg/m3, averaging 46.7 µg/m3 overall and 48.5 µg/m3 in Ouagadougou — roughly nine to ten times the WHO daily guideline, with estimated annual levels 11-14 times the WHO annual guideline.

Conclusions

A companion 'lessons learned' paper is candid that the network required ongoing effort to manage dust fouling, power interruptions and calibration drift, framing the approach as practical guidance for future low-cost deployments rather than a finished, maintenance-free system.

Implementation

Indicative cost
Low (< €50k)
Time to results
Short (< 1 year)
Staffing & skills
Burkinabé university research teams (Université Joseph Ki-Zerbo, École Normale Supérieure, Université Nazi Boni), IRSAT-CNRST (Burkina Faso), CNR-IBE (Italy)

Conditions for success

  • Access to at least one reference-grade instrument (TEOM) for local calibration
  • An academic partner willing to run ongoing calibration and upkeep
  • A locally trained correction model (Gaussian Mixture Regression) to adjust low-cost sensor readings

Common failure modes

  • Dust fouling the optical sensors
  • Power interruptions
  • Calibration drift over time

Where it fits

Governance type
academic/research partnership
Scale
3 cities in Burkina Faso, plus comparison sites in Niger and Guinea
Income level
low-income

Commonly funded by

National / regional programmes

Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Similar practices you may find useful