Characterization and prediction of PM2.5 levels in Afghanistan using machine learning techniques

IF 2.8 4区 地球科学 Q3 METEOROLOGY & ATMOSPHERIC SCIENCES
Obaidullah Salehie, Mohamad Hidayat Bin Jamal, Shamsuddin Shahid
{"title":"Characterization and prediction of PM2.5 levels in Afghanistan using machine learning techniques","authors":"Obaidullah Salehie, Mohamad Hidayat Bin Jamal, Shamsuddin Shahid","doi":"10.1007/s00704-024-05172-6","DOIUrl":null,"url":null,"abstract":"<p>Afghanistan faces severe air quality issues in major cities due to various sources like transportation, domestic energy use, and industrial activity. This study investigates PM2.5 spatiotemporal variability and its future relationship with six meteorological variables: precipitation, temperature, dewpoint temperature, wind speed, boundary layer height and surface pressure. This study aims to assess the spatiotemporal variability of PM2.5 concentrations in Afghanistan and derive models for predicting PM2.5 from the six variables. Satellite-measured PM2.5 and six reanalyses (ERA5) meteorological datasets for 1998–2020 were used as predictors. Three machine learning models, AdaBoost, Random Forest (RF), and Support Vector Machine (SVM), were used to develop the annual and seasonal PM2.5 concentration prediction model. Results suggest PM2.5 levels ranging from 60–80 µg/m<sup>3</sup> in northern, southern, and western regions, while other areas experience lower levels (12–50 µg/m<sup>3</sup>). The lowest PM2.5 concentrations are in the Hindu Kush mountain range. Summer exhibited the highest PM2.5 concentrations, reaching a maximum of 137.4 µg/m<sup>3</sup> and an average of 48.5 µg/m<sup>3</sup>. Among the prediction models, RF performed best in predicting PM2.5 across Afghanistan, as evidenced by the evaluation metrics: NRMSE (59.2), RSR (0.59), rSD (0.75), and higher values of NSE (0.65), R<sup>2</sup> (0.65), and KGE (0.68). The geographical and seasonal distribution of observed PM2.5 distribution was very similar to the PM2.5 estimated using RF compared to the other two models. The analysis showed that air temperature, precipitation, wind speeds, and boundary layer heights play significant roles in PM2.5 distribution. However, the relationship between precipitation and PM2.5 was more pronounced than other meteorological variables.</p>","PeriodicalId":22945,"journal":{"name":"Theoretical and Applied Climatology","volume":null,"pages":null},"PeriodicalIF":2.8000,"publicationDate":"2024-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Theoretical and Applied Climatology","FirstCategoryId":"89","ListUrlMain":"https://doi.org/10.1007/s00704-024-05172-6","RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"METEOROLOGY & ATMOSPHERIC SCIENCES","Score":null,"Total":0}
引用次数: 0

Abstract

Afghanistan faces severe air quality issues in major cities due to various sources like transportation, domestic energy use, and industrial activity. This study investigates PM2.5 spatiotemporal variability and its future relationship with six meteorological variables: precipitation, temperature, dewpoint temperature, wind speed, boundary layer height and surface pressure. This study aims to assess the spatiotemporal variability of PM2.5 concentrations in Afghanistan and derive models for predicting PM2.5 from the six variables. Satellite-measured PM2.5 and six reanalyses (ERA5) meteorological datasets for 1998–2020 were used as predictors. Three machine learning models, AdaBoost, Random Forest (RF), and Support Vector Machine (SVM), were used to develop the annual and seasonal PM2.5 concentration prediction model. Results suggest PM2.5 levels ranging from 60–80 µg/m3 in northern, southern, and western regions, while other areas experience lower levels (12–50 µg/m3). The lowest PM2.5 concentrations are in the Hindu Kush mountain range. Summer exhibited the highest PM2.5 concentrations, reaching a maximum of 137.4 µg/m3 and an average of 48.5 µg/m3. Among the prediction models, RF performed best in predicting PM2.5 across Afghanistan, as evidenced by the evaluation metrics: NRMSE (59.2), RSR (0.59), rSD (0.75), and higher values of NSE (0.65), R2 (0.65), and KGE (0.68). The geographical and seasonal distribution of observed PM2.5 distribution was very similar to the PM2.5 estimated using RF compared to the other two models. The analysis showed that air temperature, precipitation, wind speeds, and boundary layer heights play significant roles in PM2.5 distribution. However, the relationship between precipitation and PM2.5 was more pronounced than other meteorological variables.

Abstract Image

利用机器学习技术描述和预测阿富汗 PM2.5 水平
由于交通、家庭能源使用和工业活动等各种原因,阿富汗主要城市面临着严重的空气质量问题。本研究调查了 PM2.5 的时空变异性及其未来与六个气象变量的关系:降水、温度、露点温度、风速、边界层高度和表面气压。这项研究旨在评估阿富汗 PM2.5 浓度的时空变异性,并根据这六个变量推导出预测 PM2.5 的模型。卫星测量的 PM2.5 和 1998-2020 年的六个再分析(ERA5)气象数据集被用作预测因子。三种机器学习模型,即 AdaBoost、随机森林(RF)和支持向量机(SVM),被用来开发年度和季节 PM2.5 浓度预测模型。结果表明,北部、南部和西部地区的 PM2.5 浓度在 60-80 µg/m3 之间,而其他地区的浓度较低(12-50 µg/m3)。兴都库什山脉的 PM2.5 浓度最低。夏季的 PM2.5 浓度最高,最高达 137.4 微克/立方米,平均为 48.5 微克/立方米。在各种预测模型中,RF 在预测阿富汗各地 PM2.5 方面表现最佳,这一点可以从评估指标中得到证明:NRMSE (59.2)、RSR (0.59)、rSD (0.75),以及较高的 NSE (0.65)、R2 (0.65) 和 KGE (0.68)。与其他两个模型相比,观测到的 PM2.5 分布的地理分布和季节分布与使用 RF 估算的 PM2.5 分布非常相似。分析表明,气温、降水、风速和边界层高度对 PM2.5 的分布有重要影响。然而,降水与 PM2.5 之间的关系比其他气象变量更为明显。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Theoretical and Applied Climatology
Theoretical and Applied Climatology 地学-气象与大气科学
CiteScore
6.00
自引率
11.80%
发文量
376
审稿时长
4.3 months
期刊介绍: Theoretical and Applied Climatology covers the following topics: - climate modeling, climatic changes and climate forecasting, micro- to mesoclimate, applied meteorology as in agro- and forestmeteorology, biometeorology, building meteorology and atmospheric radiation problems as they relate to the biosphere - effects of anthropogenic and natural aerosols or gaseous trace constituents - hardware and software elements of meteorological measurements, including techniques of remote sensing
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信