Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Using Deep Ensemble Forest for High Resolution Mapping of PM2.5 from MODIS MAIAC AOD in Tehran, Iran

Hossein Bagheri

TL;DR

The study tackles the challenge of high-resolution PM2.5 mapping in Tehran with limited ground stations by leveraging MAIAC 1 km AOD and ERA5 meteorology. It introduces a cascade-based deep ensemble forest that combines Random Forest and Extremely Randomized Trees to learn a PM2.5–AOD relationship without backpropagation, validated against baseline regressors and deep networks. On test data, the approach achieves $R^2$ ≈ 0.74 with RMSE ≈ 8.86 μg/m^3 and MAE ≈ 6.86 μg/m^3, outperforming both deep learning and traditional ensemble methods. The method yields reliable 1 km PM2.5 maps, including daily and annual views, highlighting pollution patterns in Tehran and offering a practical tool for urban air quality management.

Abstract

High resolution mapping of PM2.5 concentration over Tehran city is challenging because of the complicated behavior of numerous sources of pollution and the insufficient number of ground air quality monitoring stations. Alternatively, high resolution satellite Aerosol Optical Depth (AOD) data can be employed for high resolution mapping of PM2.5. For this purpose, different data-driven methods have been used in the literature. Recently, deep learning methods have demonstrated their ability to estimate PM2.5 from AOD data. However, these methods have several weaknesses in solving the problem of estimating PM2.5 from satellite AOD data. In this paper, the potential of the deep ensemble forest method for estimating the PM2.5 concentration from AOD data was evaluated. The results showed that the deep ensemble forest method with R2 = 0.74 gives a higher accuracy of PM2.5 estimation than deep learning methods (R2 = 0.67) as well as classic data-driven methods such as random forest (R2 = 0.68). Additionally, the estimated values of PM2.5 using the deep ensemble forest algorithm were used along with ground data to generate a high resolution map of PM2.5. Evaluation of the produced PM2.5 map revealed the good performance of the deep ensemble forest for modeling the variation of PM2.5 in the city of Tehran.

Using Deep Ensemble Forest for High Resolution Mapping of PM2.5 from MODIS MAIAC AOD in Tehran, Iran

TL;DR

The study tackles the challenge of high-resolution PM2.5 mapping in Tehran with limited ground stations by leveraging MAIAC 1 km AOD and ERA5 meteorology. It introduces a cascade-based deep ensemble forest that combines Random Forest and Extremely Randomized Trees to learn a PM2.5–AOD relationship without backpropagation, validated against baseline regressors and deep networks. On test data, the approach achieves ≈ 0.74 with RMSE ≈ 8.86 μg/m^3 and MAE ≈ 6.86 μg/m^3, outperforming both deep learning and traditional ensemble methods. The method yields reliable 1 km PM2.5 maps, including daily and annual views, highlighting pollution patterns in Tehran and offering a practical tool for urban air quality management.

Abstract

High resolution mapping of PM2.5 concentration over Tehran city is challenging because of the complicated behavior of numerous sources of pollution and the insufficient number of ground air quality monitoring stations. Alternatively, high resolution satellite Aerosol Optical Depth (AOD) data can be employed for high resolution mapping of PM2.5. For this purpose, different data-driven methods have been used in the literature. Recently, deep learning methods have demonstrated their ability to estimate PM2.5 from AOD data. However, these methods have several weaknesses in solving the problem of estimating PM2.5 from satellite AOD data. In this paper, the potential of the deep ensemble forest method for estimating the PM2.5 concentration from AOD data was evaluated. The results showed that the deep ensemble forest method with R2 = 0.74 gives a higher accuracy of PM2.5 estimation than deep learning methods (R2 = 0.67) as well as classic data-driven methods such as random forest (R2 = 0.68). Additionally, the estimated values of PM2.5 using the deep ensemble forest algorithm were used along with ground data to generate a high resolution map of PM2.5. Evaluation of the produced PM2.5 map revealed the good performance of the deep ensemble forest for modeling the variation of PM2.5 in the city of Tehran.
Paper Structure (18 sections, 11 equations, 12 figures, 3 tables)

This paper contains 18 sections, 11 equations, 12 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Study Area and Datasets
  3. PM2.5 In-situ Measurements
  4. Satellite AOD Measurements
  5. Meteorological Data
  6. Methods
  7. Structure of Deep Ensemble Forest
  8. Cascade Layers
  9. Estimators
  10. Comparative Study
  11. Experimental Setup
  12. Data Preparation for PM2.5 Estimation by Deep Ensemble Forest
  13. Deep Ensemble Forest Setup
  14. Setup Algorithms to Compare with Deep Ensemble Forest
  15. Results and Discussion
  16. ...and 3 more sections

Figures (12)

  • Figure 1: A display of the study area. Numbers have been assigned by the author for further analysis in Section \ref{['result']}.
  • Figure 2: Deep ensemble forest structure (with four cascade layers including two estimators per layer) for PM2.5 estimation
  • Figure 3: Random Forest estimator for PM2.5 prediction
  • Figure 4: Extra Trees estimator for PM2.5 prediction
  • Figure 5: The framework designed for estimation of PM2.5 over Tehran at 1 $km$ resolution
  • ...and 7 more figures