Predicting Particulate Matter (PM ₁₀) Levels in Morocco: A 5-Day Forecast Using the Analog Ensemble Method

Area of Science:

• Environmental Science
• Atmospheric Science
• Public Health

Background:

• Accurate Particulate Matter (PM10) prediction is vital for public health and environmental management.
• PM10 levels, influenced by natural pollutants like dust storms, pose significant risks.
• Existing forecasts require improvement for effective air quality management.

Purpose of the Study:

• To enhance the accuracy of five-day PM10 forecasts over Morocco.
• To evaluate the effectiveness of Analog Ensemble (AnEn) and Bias Correction (AnEnBc) techniques for PM10 prediction.
• To establish a benchmark for PM10 forecasting in the Middle East and North Africa region.

Main Methods:

• Post-processing of Copernicus Atmosphere Monitoring Service (CAMS) global forecasts using AnEn and AnEnBc.
• Utilizing CAMS reanalysis data as a reference for model calibration.
• Quantitative assessment of forecast accuracy using RMSE, MAE, R², and Pearson correlation coefficient.

Main Results:

• Root Mean Square Error (RMSE) decreased from 63.83 to 44.73 μg/m³.
• Mean Absolute Error (MAE) reduced from 36.70 to 24.30 μg/m³.
• Coefficient of determination (R²) increased from 29.11% to 65.18%, and Pearson correlation rose from 0.61 to 0.82.

Conclusions:

• AnEn and AnEnBc significantly improve PM10 forecast accuracy over Morocco.
• This approach offers potential for early warnings of PM10 pollution events.
• The findings support the integration of advanced forecasting into environmental policies and public health strategies.