[Analysis of Carbon Emission Impact Factors and Peak Scenario Simulation for Resource-based Cities in China Based on RF-RFECV Feature Selection and BO-CNN-BiLSTM-attention].

Q2 Environmental Science
Yi Han, Zhen-Mei Hou
{"title":"[Analysis of Carbon Emission Impact Factors and Peak Scenario Simulation for Resource-based Cities in China Based on RF-RFECV Feature Selection and BO-CNN-BiLSTM-attention].","authors":"Yi Han, Zhen-Mei Hou","doi":"10.13227/j.hjkx.202501278","DOIUrl":null,"url":null,"abstract":"<p><p>As China's 2030 carbon peak target approaches, carbon emission reduction efforts have become increasingly urgent and crucial. Resource-based cities, characterized by their reliance on high-carbon industries, play a pivotal role in the nation's carbon peak progress. This study focuses on 108 resource-based cities from 2000 to 2022, employing the RF-RFECV algorithm to identify key factors influencing carbon emissions in these cities and utilizing the SHAP algorithm to evaluate feature importance. Furthermore, a BO-CNN-BiLSTM-attention prediction model is constructed, combined with scenario analysis to simulate the dynamic pathways of carbon peaking in resource-based cities under low-carbon, baseline, and high-speed scenarios. The results indicate the following: ① From the perspective of influencing factors, energy consumption was the most critical driver of carbon emissions in resource-based cities, reflecting their dependence on energy-intensive industries. The GDP of the primary industry and population density had a negative impact on carbon emissions, while the other six variables exerted a positive influence. ② In terms of city types, the impact of energy consumption on regenerative cities gradually declined, the development of secondary industries varied in its influence across different city types, and urbanization levels had the most significant impact on growing resource-based cities. ③ According to the peak scenario simulations, under the baseline and high-speed scenarios, carbon emissions in resource-based cities will continue to rise before 2040, whereas under the low-carbon scenario, emissions are projected to peak by 2034. Based on these findings, resource-based cities should achieve low-carbon transformation and sustainable development by improving energy efficiency, developing renewable energy, advancing green finance, adjusting industrial structures, and establishing carbon emission trading markets.</p>","PeriodicalId":35937,"journal":{"name":"环境科学","volume":"47 3","pages":"1433-1448"},"PeriodicalIF":0.0000,"publicationDate":"2026-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"环境科学","FirstCategoryId":"1087","ListUrlMain":"https://doi.org/10.13227/j.hjkx.202501278","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"Environmental Science","Score":null,"Total":0}
引用次数: 0

Abstract

As China's 2030 carbon peak target approaches, carbon emission reduction efforts have become increasingly urgent and crucial. Resource-based cities, characterized by their reliance on high-carbon industries, play a pivotal role in the nation's carbon peak progress. This study focuses on 108 resource-based cities from 2000 to 2022, employing the RF-RFECV algorithm to identify key factors influencing carbon emissions in these cities and utilizing the SHAP algorithm to evaluate feature importance. Furthermore, a BO-CNN-BiLSTM-attention prediction model is constructed, combined with scenario analysis to simulate the dynamic pathways of carbon peaking in resource-based cities under low-carbon, baseline, and high-speed scenarios. The results indicate the following: ① From the perspective of influencing factors, energy consumption was the most critical driver of carbon emissions in resource-based cities, reflecting their dependence on energy-intensive industries. The GDP of the primary industry and population density had a negative impact on carbon emissions, while the other six variables exerted a positive influence. ② In terms of city types, the impact of energy consumption on regenerative cities gradually declined, the development of secondary industries varied in its influence across different city types, and urbanization levels had the most significant impact on growing resource-based cities. ③ According to the peak scenario simulations, under the baseline and high-speed scenarios, carbon emissions in resource-based cities will continue to rise before 2040, whereas under the low-carbon scenario, emissions are projected to peak by 2034. Based on these findings, resource-based cities should achieve low-carbon transformation and sustainable development by improving energy efficiency, developing renewable energy, advancing green finance, adjusting industrial structures, and establishing carbon emission trading markets.

[基于RF-RFECV特征选择和BO-CNN-BiLSTM-attention的中国资源型城市碳排放影响因子分析及峰值情景模拟]。
随着中国2030年碳峰值目标的临近,碳减排工作变得越来越紧迫和重要。资源型城市以依赖高碳产业为特征,在国家碳峰值进程中发挥着关键作用。本研究以2000 - 2022年108个资源型城市为研究对象,采用RF-RFECV算法识别影响城市碳排放的关键因素,并利用SHAP算法评价特征重要性。构建bo - cnn - bilstm -注意力预测模型,结合情景分析,模拟低碳、基线和高速情景下资源型城市碳峰值的动态路径。结果表明:①从影响因素看,能源消费是资源型城市碳排放的最主要驱动因素,反映了资源型城市对能源密集型产业的依赖程度;第一产业GDP和人口密度对碳排放有负向影响,其他6个变量对碳排放有正向影响。②从城市类型看,能源消费对再生型城市的影响逐渐减弱,第二产业发展对不同城市类型的影响存在差异,城市化水平对增长型资源型城市的影响最为显著。③峰值情景模拟表明,在基线情景和高速情景下,资源型城市碳排放在2040年前将继续上升,而在低碳情景下,碳排放将在2034年达到峰值。据此,资源型城市应从提高能效、发展可再生能源、推进绿色金融、调整产业结构、建立碳排放权交易市场等方面实现低碳转型和可持续发展。
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
环境科学
环境科学 Environmental Science-Environmental Science (all)
CiteScore
4.40
自引率
0.00%
发文量
15329
期刊介绍:
×
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:604180095
Book学术官方微信
小红书