A Novel Evolutionary Deep Learning Approach for PM2.5 Prediction Using Remote Sensing and Spatial–Temporal Data: A Case Study of Tehran

Loading...
Thumbnail Image

Access rights

openAccess
CC BY
publishedVersion

URL

Journal Title

Journal ISSN

Volume Title

A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä

Date

2025-02

Major/Subject

Mcode

Degree programme

Language

en

Pages

42

Series

ISPRS International Journal of Geo-Information, Volume 14, issue 2

Abstract

Forecasting particulate matter with a diameter of 2.5 μm (PM2.5) is critical due to its significant effects on both human health and the environment. While ground-based pollution measurement stations provide highly accurate PM2.5 data, their limited number and geographic coverage present significant challenges. Recently, the use of aerosol optical depth (AOD) has emerged as a viable alternative for estimating PM2.5 levels, offering a broader spatial coverage and higher resolution. Concurrently, long short-term memory (LSTM) models have shown considerable promise in enhancing air quality predictions, often outperforming other prediction techniques. To address these challenges, this study leverages geographic information systems (GIS), remote sensing (RS), and a hybrid LSTM architecture to predict PM2.5 concentrations. Training LSTM models, however, is an NP-hard problem, with gradient-based methods facing limitations such as getting trapped in local minima, high computational costs, and the need for continuous objective functions. To overcome these issues, we propose integrating the novel orchard algorithm (OA) with LSTM to optimize air pollution forecasting. This paper utilizes meteorological data, topographical features, PM2.5 pollution levels, and satellite imagery from the city of Tehran. Data preparation processes include noise reduction, spatial interpolation, and addressing missing data. The performance of the proposed OA-LSTM model is compared to five advanced machine learning (ML) algorithms. The proposed OA-LSTM model achieved the lowest root mean square error (RMSE) value of 3.01 µg/m3 and the highest coefficient of determination (R2) value of 0.88, underscoring its effectiveness compared to other models. This paper employs a binary OA method for sensitivity analysis, optimizing feature selection by minimizing prediction error while retaining critical predictors through a penalty-based objective function. The generated maps reveal higher PM2.5 concentrations in autumn and winter compared to spring and summer, with northern and central areas showing the highest pollution levels.

Description

Publisher Copyright: © 2025 by the authors.

Keywords

aerosol optical depth, geographic information systems, long short-term memory, orchard algorithm, PM, remote sensing

Other note

Citation

Kaveh, M, Mesgari, M S & Kaveh, M 2025, ' A Novel Evolutionary Deep Learning Approach for PM 2.5 Prediction Using Remote Sensing and Spatial–Temporal Data: A Case Study of Tehran ', ISPRS International Journal of Geo-Information, vol. 14, no. 2, 42 . https://doi.org/10.3390/ijgi14020042