electricsheepafrica/africa-air-quality-all

---
title: African Air Pollution Source Apportionment
language: en
license: cc-by-4.0
task_categories:
- time-series-forecasting
- tabular-classification
- tabular-regression
pretty_name: African Air Pollution Source Apportionment
tags:
- air-quality
- africa
- pm25
- pm10
- source-apportionment
- health-impacts
- who-guidelines
- environmental-health
- urban-pollution
configs:
- config_name: daily_measurements
data_files: daily_measurements.csv
description: "Daily air quality measurements for 20 African cities across 3 scenarios (2020-2024). Contains per-station PM2.5/PM10, source contributions (traffic, industry, cooking/biomass, dust, waste burning, power generation), AQI, WHO exceedance flags, health impact estimates, and policy compliance rates."
- config_name: monthly_summary
data_files: monthly_summary.csv
description: "Monthly aggregated statistics per city/scenario/year. Includes mean, median, std, min, max, P95 for PM2.5/PM10, AQI statistics, exceedance days, source contribution percentages, and health metrics."
- config_name: annual_summary
data_files: annual_summary.csv
description: "Annual aggregated statistics per city/scenario/year. Comprehensive summary with PM2.5/PM10 annual means, AQI statistics, WHO guideline exceedance, source apportionment percentages, policy compliance, monitoring coverage, DALYs, and premature death estimates."
---

# African Air Pollution Source Apportionment Dataset

## Overview

A comprehensive air pollution source apportionment dataset covering **20 major African cities** across **3 policy scenarios** from **2020-2024**. Parameters are calibrated to peer-reviewed literature and WHO guidelines.

**Total records:** 22,000+ (daily + monthly + annual)

## Cities Covered

| City | Country | Region | Population (M) | Baseline PM2.5 (μg/m³) |
|------|---------|--------|----------------|------------------------|
| Lagos | Nigeria | West Africa | 15.4 | 52.0 |
| Cairo | Egypt | North Africa | 10.1 | 72.0 |
| Johannesburg | South Africa | Southern Africa | 5.6 | 28.0 |
| Nairobi | Kenya | East Africa | 4.7 | 22.0 |
| Addis Ababa | Ethiopia | East Africa | 5.2 | 32.0 |
| Accra | Ghana | West Africa | 2.5 | 38.0 |
| Dakar | Senegal | West Africa | 3.1 | 35.0 |
| Kinshasa | DR Congo | Central Africa | 14.3 | 42.0 |
| Dar es Salaam | Tanzania | East Africa | 7.0 | 30.0 |
| Kampala | Uganda | East Africa | 1.7 | 36.0 |
| Casablanca | Morocco | North Africa | 3.7 | 32.0 |
| Tunis | Tunisia | North Africa | 2.9 | 30.0 |
| Luanda | Angola | Southern Africa | 8.3 | 38.0 |
| Maputo | Mozambique | Southern Africa | 1.1 | 26.0 |
| Kigali | Rwanda | East Africa | 1.2 | 18.0 |
| Bamako | Mali | West Africa | 2.5 | 58.0 |
| Abidjan | Cote d'Ivoire | West Africa | 5.5 | 40.0 |
| Douala | Cameroon | Central Africa | 3.6 | 34.0 |
| Algiers | Algeria | North Africa | 3.5 | 36.0 |
| Khartoum | Sudan | North Africa | 5.3 | 68.0 |

## Scenarios

### Baseline
Current conditions reflecting 2020-2024 measurements. Parameters calibrated to observed PM2.5/PM10 concentrations from monitoring networks and satellite-derived estimates.

### Clean Air Action
Aggressive policy intervention scenario modeling:
- Transition to clean cooking fuels (LPG, electric)
- Euro 5/6 vehicle emission standards
- Industrial emission controls and scrubbers
- Improved waste management (reduced open burning)
- Power sector decarbonization
- **Result:** ~32% reduction in PM2.5, ~28% in PM10

### Industrialization Pressure
Rapid industrial growth without environmental safeguards:
- Increased vehicle fleet and older vehicle imports
- Uncontrolled industrial expansion
- Continued reliance on biomass cooking
- Increased construction dust
- Growing waste burning due to inadequate management
- Coal/diesel power generation expansion
- **Result:** ~32% increase in PM2.5, ~28% in PM10

## Source Apportionment Categories

| Source | Description | Key Literature |
|--------|-------------|----------------|
| Traffic | Vehicle exhaust, brake/tire wear, road dust resuspension | African vehicle emission studies |
| Industry | Manufacturing, construction, mining emissions | Industrial monitoring in Africa |
| Cooking/Biomass | Household solid fuel use (wood, charcoal, dung) | WHO household air pollution studies |
| Dust | Saharan/sahelian dust, soil erosion, construction | Saharan dust transport studies |
| Waste Burning | Open burning of municipal solid waste | African waste management studies |
| Power Generation | Diesel generators, coal plants, grid emissions | African energy sector studies |

## Data Files

### `daily_measurements.csv`
Daily per-station measurements with ~20,000+ records.

| Column | Type | Description |
|--------|------|-------------|
| city | string | City name |
| country | string | Country name |
| region | string | African region |
| date | string | Date (YYYY-MM-DD) |
| year | int | Year (2020-2024) |
| month | int | Month (1-12) |
| day_of_year | int | Day of year (1-365) |
| day_of_week | int | Day of week (0=Mon, 6=Sun) |
| scenario | string | Scenario name |
| scenario_description | string | Scenario description |
| latitude | float | City latitude |
| longitude | float | City longitude |
| elevation_m | int | City elevation (meters) |
| climate_zone | string | Climate classification |
| population_millions | float | City population (millions) |
| station_id | string | Monitoring station ID |
| station_type | string | Station type (urban_traffic, urban_background, industrial, suburban, rural) |
| pm25_ugm3 | float | PM2.5 concentration (μg/m³) |
| pm10_ugm3 | float | PM10 concentration (μg/m³) |
| pm25_pm10_ratio | float | PM2.5/PM10 ratio |
| aqi_pm25 | float | AQI based on PM2.5 |
| aqi_pm10 | float | AQI based on PM10 |
| aqi | float | Overall AQI (max of PM2.5/PM10) |
| aqi_category | string | AQI category |
| source_traffic_ugm3 | float | Traffic contribution to PM2.5 (μg/m³) |
| source_industry_ugm3 | float | Industrial contribution to PM2.5 (μg/m³) |
| source_cooking_biomass_ugm3 | float | Cooking/biomass contribution to PM2.5 (μg/m³) |
| source_dust_ugm3 | float | Dust contribution to PM2.5 (μg/m³) |
| source_waste_burning_ugm3 | float | Waste burning contribution to PM2.5 (μg/m³) |
| source_power_generation_ugm3 | float | Power generation contribution to PM2.5 (μg/m³) |
| seasonal_factor | float | Seasonal multiplier |
| day_of_week_factor | float | Weekday/weekend multiplier |
| who_pm25_24hr_exceedance | int | WHO 24-hr PM2.5 guideline exceedance (1/0) |
| who_pm10_24hr_exceedance | int | WHO 24-hr PM10 guideline exceedance (1/0) |
| who_annual_pm25_exceedance | int | WHO annual PM2.5 guideline exceedance (1/0) |
| who_annual_pm10_exceedance | int | WHO annual PM10 guideline exceedance (1/0) |
| policy_compliance_rate | float | Policy compliance rate (0-1) |
| monitoring_stations_active | int | Number of active monitoring stations |
| data_completeness_pct | float | Data completeness percentage |
| population_exposed_pct | float | Population exposed percentage |
| grid_electricity_access_pct | float | Grid electricity access percentage |
| clean_cooking_access_pct | float | Clean cooking fuel access percentage |
| vehicle_fleet_millions | float | City vehicle fleet (millions) |
| industrial_zones | int | Number of industrial zones |
| dalys_per_100k | float | DALYs per 100,000 population |
| premature_deaths_estimated_annual | float | Estimated annual premature deaths |
| exceedance_days_who_pm25_annual | int | Annual exceedance days vs WHO PM2.5 |
| exceedance_days_who_pm10_annual | int | Annual exceedance days vs WHO PM10 |

### `monthly_summary.csv`
Monthly aggregated statistics (~1,440 records).

| Column | Type | Description |
|--------|------|-------------|
| city | string | City name |
| country | string | Country name |
| region | string | African region |
| year | int | Year |
| month | int | Month |
| scenario | string | Scenario name |
| pm25_monthly_mean | float | Monthly mean PM2.5 |
| pm25_monthly_median | float | Monthly median PM2.5 |
| pm25_monthly_std | float | Monthly std dev PM2.5 |
| pm25_monthly_max | float | Monthly max PM2.5 |
| pm25_monthly_min | float | Monthly min PM2.5 |
| pm25_monthly_p95 | float | Monthly 95th percentile PM2.5 |
| pm10_monthly_mean | float | Monthly mean PM10 |
| pm10_monthly_median | float | Monthly median PM10 |
| pm10_monthly_std | float | Monthly std dev PM10 |
| pm10_monthly_max | float | Monthly max PM10 |
| pm10_monthly_min | float | Monthly min PM10 |
| pm10_monthly_p95 | float | Monthly 95th percentile PM10 |
| aqi_monthly_mean | float | Monthly mean AQI |
| aqi_monthly_max | float | Monthly max AQI |
| who_pm25_24hr_exceedance_days | int | Monthly exceedance days PM2.5 |
| who_pm10_24hr_exceedance_days | int | Monthly exceedance days PM10 |
| source_traffic_pct | float | Traffic % of PM2.5 |
| source_industry_pct | float | Industry % of PM2.5 |
| source_cooking_biomass_pct | float | Cooking/biomass % of PM2.5 |
| source_dust_pct | float | Dust % of PM2.5 |
| source_waste_burning_pct | float | Waste burning % of PM2.5 |
| source_power_generation_pct | float | Power generation % of PM2.5 |
| policy_compliance_rate | float | Policy compliance rate |
| monitoring_stations_active | int | Active monitoring stations |
| data_completeness_pct | float | Data completeness % |
| dalys_per_100k | float | DALYs per 100k |
| premature_deaths_estimated_annual | float | Estimated annual premature deaths |

### `annual_summary.csv`
Annual aggregated statistics (~300 records).

| Column | Type | Description |
|--------|------|-------------|
| city | string | City name |
| country | string | Country name |
| region | string | African region |
| year | int | Year |
| scenario | string | Scenario name |
| pm25_annual_mean | float | Annual mean PM2.5 |
| pm25_annual_median | float | Annual median PM2.5 |
| pm25_annual_std | float | Annual std dev PM2.5 |
| pm25_annual_max | float | Annual max PM2.5 |
| pm25_annual_min | float | Annual min PM2.5 |
| pm25_annual_p95 | float | Annual 95th percentile PM2.5 |
| pm10_annual_mean | float | Annual mean PM10 |
| pm10_annual_median | float | Annual median PM10 |
| pm10_annual_std | float | Annual std dev PM10 |
| pm10_annual_max | float | Annual max PM10 |
| pm10_annual_min | float | Annual min PM10 |
| pm10_annual_p95 | float | Annual 95th percentile PM10 |
| aqi_annual_mean | float | Annual mean AQI |
| aqi_annual_max | float | Annual max AQI |
| who_pm25_24hr_exceedance_days | int | Annual exceedance days PM2.5 |
| who_pm10_24hr_exceedance_days | int | Annual exceedance days PM10 |
| who_annual_pm25_exceedance | int | WHO annual PM2.5 exceedance flag |
| who_annual_pm10_exceedance | int | WHO annual PM10 exceedance flag |
| source_traffic_pct | float | Traffic % of PM2.5 |
| source_industry_pct | float | Industry % of PM2.5 |
| source_cooking_biomass_pct | float | Cooking/biomass % of PM2.5 |
| source_dust_pct | float | Dust % of PM2.5 |
| source_waste_burning_pct | float | Waste burning % of PM2.5 |
| source_power_generation_pct | float | Power generation % of PM2.5 |
| policy_compliance_rate | float | Policy compliance rate |
| monitoring_stations_active | int | Active monitoring stations |
| data_completeness_pct | float | Data completeness % |
| population_exposed_pct | float | Population exposed % |
| grid_electricity_access_pct | float | Grid electricity access % |
| clean_cooking_access_pct | float | Clean cooking access % |
| vehicle_fleet_millions | float | Vehicle fleet (millions) |
| industrial_zones | int | Industrial zones |
| dalys_per_100k | float | DALYs per 100k |
| premature_deaths_estimated_annual | float | Estimated annual premature deaths |
| exceedance_days_who_pm25_annual | int | Annual exceedance days PM2.5 |
| exceedance_days_who_pm10_annual | int | Annual exceedance days PM10 |

## WHO Air Quality Guidelines (2021)

This dataset uses the WHO 2021 AQG thresholds:

| Pollutant | Annual Guideline | 24-hour Guideline |
|-----------|-----------------|-------------------|
| PM2.5 | 5 μg/m³ | 15 μg/m³ |
| PM10 | 15 μg/m³ | 45 μg/m³ |

## Usage

```python
import pandas as pd

# Load datasets
daily = pd.read_csv("daily_measurements.csv")
monthly = pd.read_csv("monthly_summary.csv")
annual = pd.read_csv("annual_summary.csv")

# Compare scenarios for a city
lagos = annual[annual["city"] == "Lagos"]
lagos.groupby("scenario")["pm25_annual_mean"].mean()

# Find cities exceeding WHO guidelines by >300 days/year
high_exceed = daily[daily["who_pm25_24hr_exceedance"] == 1].groupby(
["city", "year", "scenario"]
).size().reset_index(name="days")
high_exceed[high_exceed["days"] > 300]
```

## References

- WHO Global Air Quality Guidelines 2021
- State of Global Air 2024 Report
- PM2.5 source apportionment studies for African cities
- Household air pollution from cooking studies
- Vehicle emissions in African cities research
- Industrial pollution monitoring in Africa

## License

CC BY 4.0