Adaptive learning rate optimization in deep recurrent architectures for precision PM2.5 forecasting under climate variability
Chemosphere · 2026-03 · 397 · 144875
DOI: 10.1016/j.chemosphere.2026.144875 ↗Kurz gefasst
Fachartikel in der Zeitschrift Chemosphere, 2026-03. Verfasst von Caner Erden. Beteiligte Einrichtung: Department of Computer Engineering, Faculty of Technology, Sakarya University of Applied Sciences, Sakarya, Turkey. Electronic address: cerden@subu.e…. Hauptthemen laut MEDLINE: Particulate Matter, Environmental Monitoring, Deep Learning, Air Pollutants. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 34 Literaturangaben. Gefördert von: Scientific and Technological Research Council of Turkey.
Abstract (Originalsprache)
Accurate forecasting of fine particulate matter (PM_{2.5}) is critical for safeguarding public health; however, conventional deep learning models frequently encounter convergence instability when exposed to highly volatile meteorological inputs. This study evaluates advanced learning rate scheduling strategies to improve long short-term memory (LSTM) and gated recurrent unit (GRU) performance using Atlanta's 2023-2025 air quality and meteorological data. The research compares polynomial, piecewise constant, and cosine decay schedules, revealing architecture-specific sensitivities: the analysis identifies that LSTM achieved optimal stability with piecewise constant decay (MAE = 0.9255 μg/m3), whereas GRU performed best under polynomial decay (MAE = 1.1092 μg/m3). Although Cosine Decay reduces peak errors in isolated instances, it demonstrates insufficient robustness against stochastic gradient noise. The optimized framework yields exceptional predictive fidelity, with the GRU and LSTM models attaining R2 values of 0.9495 and 0.9443, respectively. Beyond methodological advances, this study establishes a reliable computational baseline for site-specific temporal air quality monitoring, offering actionable insights into the interplay between anthropogenic emissions and climate-driven volatility. By supporting industrial emissions management and evidence-based policy, this study contributes to the broader goals of mitigating air pollution in a changing climate and advancing ecological co-governance.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- Department of Computer Engineering, Faculty of Technology, Sakarya University of Applied Sciences, Sakarya, Turkey. Electronic address: cerden@subu.edu.tr.
Förderung
- Scientific and Technological Research Council of Turkey
Zitieren
APA
Erden, C. (2026). Adaptive learning rate optimization in deep recurrent architectures for precision PM2.5 forecasting under climate variability. Chemosphere, 397, 144875. https://doi.org/10.1016/j.chemosphere.2026.144875
Vancouver
Erden C. Adaptive learning rate optimization in deep recurrent architectures for precision PM2.5 forecasting under climate variability. Chemosphere. 2026;397:144875. doi:10.1016/j.chemosphere.2026.144875
BibTeX
@article{Erden2026,
author = {Erden, Caner},
title = {Adaptive learning rate optimization in deep recurrent architectures for precision PM2.5 forecasting under climate variability},
journal = {Chemosphere},
year = {2026},
volume = {397},
pages = {144875},
doi = {10.1016/j.chemosphere.2026.144875}
}
EUVIMED