Estimating black carbon levels using machine learning models in high-concentration regions - Scientific abstract and introduction

Aug 5, 2026

Upload by Oliver

Black carbon (BC) is produced during combustion and co-occurs with pollutants such as NOx and O3, driving aerosol and PM concentrations that influence the atmospheric radiative budget. BC monitoring is limited by high instrumentation and maintenance costs, motivating cost-minimized estimation using machine learning. Models predict BC mass concentrations from readily available air pollution and meteorological inputs (NOx, O3, PM2.5, RH, and solar radiation) and are evaluated across Northern Indian cities and seasons. Results show strong agreement, with the multilayer perceptron (MLP) performing best.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages11

Reading Time28 min

AccessFree to download

Abstract

Introduction

Science of the Total Environment 948 (2024) 174804 Contents lists available at ScienceDirect Science of the Total Environment journal [homepage:](homepage: www.elsevier.com/locate/scitotenv)[ www.elsevier.com/locate/scitotenv](homepage: www.elsevier.com/locate/scitotenv) Estimating black carbon levels using machine learning models in high-concentration regions Pratima Gupta a,1, Pau Ferrer-Cid b,1, Jose M. Barcelo-Ordinasb, Jorge Garcia-Vidalb, Vijay Kumar Sonic, Mira L. P¨ohlkerd, Ajit Ahlawatd,*, Mar Vianae a Centre for Atmospheric Sciences, Indian Institute of Technology (IIT) Delhi, India b Department of Computer Architecture, Universitat Polit`ecnica de Catalunya (UPC), Barcelona, Spain c India Meteorological Department, Delhi, India d Atmospheric Microphysics Department, Leibniz Institute for Tropospheric Research, Leipzig, Germany e Institute of Environmental Assessment and Water Research, Spanish Research Council, IDAEA-CSIC, Barcelona, Spain H I G H L I G H T S G R A P H I C A L A B S T R A C T A R T I C L E I N F O Editor: Jianmin Chen Keywords: Air quality Air pollution Black carbon Monitoring Aethalometer Modelling Prediction | | | --- | | A B S T R A C T | Black carbon (BC) is emitted into the atmosphere during combustion processes, often in conjunction with emissions such as nitrogen oxides (NOx) and ozone (O3), which are also by-products of combustion. In highly polluted regions, combustion processes are one of the main sources of aerosols and particulate matter (PM) concentrations, which affect the radiative budget. Despite the high relevance of this air pollution metric, BC monitoring is quite expensive in terms of instrumentation and of maintenance and servicing. With the aim to provide tools to estimate BC while minimising instrumentation costs, we use machine learning approaches to estimate BC from air pollution and meteorological parameters (NOx, O3, PM2.5, relative humidity (RH), and solar radiation (SR)) from currently available networks. We assess the effectiveness of various machine learning models, such as random forest (RF), support vector regression (SVR), and multilayer perceptron (MLP) artificial neural network, for predicting black carbon (BC) mass concentrations in areas with high BC levels such as Northern Indian cities (Delhi and Agra), across different seasons. The results demonstrate comparable effectiveness among the models, with the multilayer perceptron (MLP) showing the most promising results. In * Corresponding author. E-mail address: [ahlawat@tropos.de](ahlawat@tropos.de) (A. Ahlawat). 1 Equal authorship. [https://doi.org/10.1016/j.scitotenv.2024.174804](https://doi.org/10.1016/j.scitotenv.2024.174804) Received 25 April 2024; Received in revised form 25 June 2024; Accepted 12 July 2024 Available online 15 July 2024 0048-9697/© 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license ([http://creativecommons.org/licenses/by/4.0/](http://creativecommons.org/licenses/by/4.0/)). P. Gupta et al. Science of the Total Environment 948 (2024) 174804 addition, the comparability between estimated and monitored BC concentrations was high. In Delhi, the MLP shows high correlations between measured and modelled concentrations during winter (R2: 0.85) and postmonsoon (R2: 0.83) seasons, and notable metrics in the pre-monsoon (R2: 0.72). The results from Agra are consistent with those from Delhi, highlighting the consistency of the neural network's performance. These results highlight the usefulness of machine learning, particularly MLP, as a valuable tool for predicting BC concentrations. This approach provides critical new opportunities for urban air quality management and mitigation strategies and may be especially valuable for megacities in medium- and low-income regions. 1. Introduction The increase in urbanization and industrialization in recent decades has brought about a concerning rise in air p

FAQs

Why is estimating black carbon (BC) important in highly polluted regions?

BC is closely linked to combustion emissions and affects air quality and climate-relevant radiative processes. Accurate concentration estimation supports effective air quality management and mitigation of health risks.

Which machine learning models are evaluated for predicting BC levels?

The study assesses random forest (RF), support vector regression (SVR), and a multilayer perceptron (MLP) neural network to predict BC mass concentrations.

What input variables are used to estimate BC in this approach?

The models use air pollution and meteorological parameters including NOx, O3, PM2.5, relative humidity (RH), and solar radiation (SR) obtained from existing monitoring networks.

Aug 5, 2026

Upload by Oliver