基于机器学习模型的海绵城市建设政策对黄河流域鸟类生物多样性的影响

doi:10.12118/j.issn.1000-6060.2025.705

摘要/Abstract

摘要：

探讨环保政策对生物多样性的影响对于推进生态文明建设至关重要。选取2010—2023年黄河流域9个省（区）县域鸟类丰富度及社会经济数据，使用双重机器学习方法实证检验海绵城市建设政策对黄河流域鸟类生物多样性的影响。结果表明：（1）相较于非试点地区，海绵城市建设政策有效提升了黄河流域试点地区鸟类生物多样性。（2）作用机制结果显示，海绵城市建设政策主要通过污染减排效应、温室缓解效应及绿色创新效应提升黄河流域鸟类生物多样性。（3）异质性分析显示，在城市行政级别和资源依赖方面，海绵城市建设政策对黄河流域鸟类生物多样性的影响在黄河流域中心城市和非资源型城市更显著；在地理空间方面，海绵城市建设政策对黄河流域中游地区鸟类生物多样性产生的显著促进作用大于下游地区，但对上游地区并未产生显著影响。

关键词: 海绵城市建设政策, 鸟类生物多样性, 空气污染, CO₂排放, 绿色技术创新, 机器学习模型

Abstract:

Biodiversity, which is vital for human survival and development, is an essential component of the healthy dynamic cycles of environmental systems, ecosystems, atmospheric systems, and socioeconomic systems. The expansion of biodiversity is also necessary for advancing ecological civilization construction. In addition, a dequate water quality is critical for the survival and existence of all biological populations. The sponge city construction policy (SCCP), which aims to restore aquatic ecosystems, conserve water resources, and enhance flood prevention capabilities, may be a significant institutional factor influencing regional biodiversity. This study focused on the rich biological communities of the Yellow River Basin over the period from 2010 to 2023. Bird species richness data and socioeconomic indicators were studied in counties across nine provinces in the region. A double ML approach was employed to empirically examine the impact of the SCCP on avian biodiversity within the Yellow River Basin. The results show that: (1) Compared with nonpilot areas, the SCCP effectively improved avian biodiversity in the pilot areas of the Yellow River Basin. (2) The mechanistic analysis revealed that the SCCP enhanced avian biodiversity in the Yellow River Basin, primarily through the effects of pollution reduction, greenhouse gas mitigation, and green innovation. (3) Heterogeneity analysis indicated that, in terms of urban hierarchy and resource dependence, the impact of the SCCP on avian biodiversity in the Yellow River Basin was more pronounced in central cities and nonresource-based cities within the basin. Geographically, the SCCP has significantly promoted avian biodiversity in the middle and lower reaches of the Yellow River Basin, with a stronger impact on the middle reaches, and no significant effect in the upper reaches. This study provides theoretical and empirical evidence, based on institutional and policy perspectives, for achieving the goal of biodiversity conservation in the Yellow River Basin.

Key words: sponge city construction policy, avian biodiversity, air pollution, carbon dioxide emissions, green technological innovation, machine learning model

李雷, 薛新帅, 刘家民. 基于机器学习模型的海绵城市建设政策对黄河流域鸟类生物多样性的影响[J]. 干旱区地理, 2026, 49(10): 2205-2214.

LI Lei, XUE Xinshuai, LIU Jiamin. Impact of sponge city construction policy on bird biodiversity in the Yellow River Basin based on machine learning model[J]. Arid Land Geography, 2026, 49(10): 2205-2214.

图/表 12

图1

表1

图2

图3

图4

图5

表2

表3

表4

图6

图7

图8

参考文献 25

[1]	Pimm S L, Jenkins C N, Abell R, et al. The biodiversity of species and their rates of extinction, distribution, and protection[J]. Science, 2014, 344(6187): 1246752, doi: 10.1126/science.1246752.
[2]	欧阳志云, 孔令桥, 黄斌斌, 等. 科技支撑西部生态系统保护修复, 保障国家生态安全[J]. 中国科学院院刊, 2025, 40(6): 991-999.
	[Ouyang Zhiyun, Kong Lingqiao, Huang Binbin, et al. Science and technology support ecosystem protection and restoration in western China to ensure ecological security of the country[J]. Bulletin of Chinese Academy of Sciences, 2025, 40(6): 991-999. ] doi: 10.3724/j.issn.1000-3045.20250421003
[3]	孟望生, 刘玮霖, 刘华桢. 水循环视角下黄河流域城市包容性绿色发展效率测度与分析[J]. 干旱区地理, 2025, 48(10): 1841-1854. doi: 10.12118/j.issn.1000-6060.2025.098
	[Meng Wangsheng, Liu Weilin, Liu Huazhen. Measurement and analysis of inclusive green development efficiency in the Yellow River Basin cities from the perspective of water cycle[J]. Arid Land Geography, 2025, 48(10): 1841-1854. ] doi: 10.12118/j.issn.1000-6060.2025.098
[4]	Gorenflo J L, Warner B D. Integrating biodiversity conservation and water development: In search of long-term solutions[J]. Wiley Interdisciplinary Reviews: Water, 2016, 3(3): 301-311. doi: 10.1002/wat2.2016.3.issue-3
[5]	田成诗, 戚琳. 生物多样性核算的国际经验及中国探索[J]. 自然资源学报, 2025, 40(5): 1141-1156. doi: 10.31497/zrzyxb.20250501
	[Tian Chengshi, Qi Lin. International experience and China’s exploration of biodiversity accounting[J]. Journal of Natural Resources, 2025, 40(5): 1141-1156. ] doi: 10.31497/zrzyxb.20250501
[6]	徐冬梅, 彭建. 生物多样性无净损失的概念、途径与成效[J]. 地理学报, 2025, 80(7): 1872-1887. doi: 10.11821/dlxb202507010
	[Xu Dongmei, Peng Jian. The concept, approach, and effectiveness of no net loss of biodiversity[J]. Acta Geographica Sinica, 2025, 80(7): 1872-1887. ] doi: 10.11821/dlxb202507010
[7]	方冬莉, 麻慧云, 于航, 等. 绿色金融助力生物多样性保护--保险的角色和趋势[J]. 南方金融, 2025(7): 64-74.
	[Fang Dongli, Ma Huiyun, Yu Hang, et al. Green finance for biodiversity conservation: The role and trends of insurance[J]. South China Finance, 2025(7): 64-74. ]
[8]	刘钊, 孔令乾, 张俊飚. 绿色金融与生物多样性--基于中国观鸟数据的考察[J]. 经济学(季刊), 2025, 25(3): 565-580.
	[Liu Zhao, Kong Lingqian, Zhang Junbiao. Green finance and biodiversity: Empirical research based on Chinese bird watching data[J]. China Economic Quarterly, 2025, 25(3): 565-580. ]
[9]	Dawson P T, Jackson T S, House J I, et al. Beyond predictions: Biodiversity conservation in a changing climate[J]. Science, 2011, 332(6025): 53-58. doi: 10.1126/science.1200303 pmid: 21454781
[10]	Agathokleous E, Sicard P, Feng Z Z, et al. Ozone pollution threatens bird populations to collapse: An imminent ecological threat?[J]. Journal of Forestry Research, 2023, 34(6): 1653-1656. doi: 10.1007/s11676-023-01645-y
[11]	Verones F, Saner D, Pfister S, et al. Effects of consumptive water use on biodiversity in wetlands of international importance[J]. Environmental Science & Technology, 2013, 47(21): 12248-12257. doi: 10.1021/es403635j
[12]	Dulsat-Masvidal M, Ciudad C, Infante O, et al. Water pollution threats in important bird and biodiversity areas from Spain[J]. Journal of Hazardous Materials, 2023, 448: 130938, doi: 10.1016/j.jhazmat.2023.130938.
[13]	张恒, 常江. 国家海绵城市建设对城市水生态环境质量的影响[J]. 自然资源学报, 2024, 39(11): 2721-2734. doi: 10.31497/zrzyxb.20241113
	[Zhang Heng, Chang Jiang. The impact of national sponge city construction on urban water ecological environment quality[J]. Journal of Natural Resources, 2024, 39(11): 2721-2734. ] doi: 10.31497/zrzyxb.20241113
[14]	Ma Y C, Jiang Y, Swallow S. China’s sponge city development for urban water resilience and sustainability: A policy discussion[J]. Science of the Total Environment, 2020, 729: 139078, doi: 10.1016/j.scitotenv.2020.139078.
[15]	姚鸣奇, 张卓群, 郑艳, 等. 海绵城市试点建设提高了生态韧性吗?--一项准自然实验[J]. 城市发展研究, 2023, 30(4): 25-33.
	[Yao Mingqi, Zhang Zhuoqun, Zheng Yan, et al. Does sponge city construction improve ecological resilience? : A quasi-natural experiment[J]. Urban Development Studies, 2023, 30(4): 25-33. ]
[16]	王丹, 杨超, 郑永宏, 等. 海绵城市改造对热岛效应减缓效果的估算[J]. 长江流域资源与环境, 2021, 30(4): 968-975.
	[Wang Dan, Yang Chao, Zheng Yonghong, et al. Estimation of mitigation effect of sponge city reconstruction on heat island effect[J]. Resources and Environment in the Yangtze Basin, 2021, 30(4): 968-975. ]
[17]	魏建飞, 袁悠燃, 李强, 等. 黄河流域国土空间效率时空变化及障碍因子分析[J]. 干旱区地理, 2025, 48(6): 1055-1066. doi: 10.12118/j.issn.1000-6060.2024.654
	[Wei Jianfei, Yuan Youran, Li Qiang, et al. Spatio-temporal variation and obstacle factors of territorial spatial efficiency in the Yellow River Basin[J]. Arid Land Geography, 2025, 48(6): 1055-1066. ] doi: 10.12118/j.issn.1000-6060.2024.654
[18]	李珂, 何静, 孟阳阳, 等. 黄河流域乡村旅游重点村空间分布及影响因素[J]. 干旱区地理, 2025, 48(11): 2031-2041. doi: 10.12118/j.issn.1000-6060.2025.181
	[Li Ke, He Jing, Meng Yangyang, et al. Spatial distribution and influencing factors of key rural tourism villages in the Yellow River Basin[J]. Arid Land Geography, 2025, 48(11): 2031-2041. ] doi: 10.12118/j.issn.1000-6060.2025.181
[19]	赵银兵, 蔡婷婷, 孙然好, 等. 海绵城市研究进展综述: 从水文过程到生态恢复[J]. 生态学报, 2019, 39(13): 4638-4646.
	[Zhao Yinbing, Cai Tingting, Sun Ranhao, et al. Review on sponge city research: From hydrological process to ecological restoration[J]. Acta Ecologica Sinica, 2019, 39(13): 4638-4646. ]
[20]	Chang W C, Liu G Y, Chen Y, et al. Assessing inter-industrial ecosystem service flows and economic benefits of sponge city: A comprehensive input-output analysis[J]. Environmental Impact Assessment Review, 2025, 114: 107955, doi: 10.1016/j.eiar.2025.107955.
[21]	Zhang Q Y, Zhu J X, Mulder J, et al. High environmental costs behind rapid economic development: Evidence from economic loss caused by atmospheric acid deposition[J]. Journal of Environmental Management, 2023, 334: 117511, doi: 10.1016/j.jenvman.2023.117511.
[22]	Lovett G M, Tear T H, Evers D C, et al. Effects of air pollution on ecosystems and biological diversity in the eastern United States[J]. Annals of the New York Academy of Sciences, 2009, 1162(1): 99-135. doi: 10.1111/nyas.2009.1162.issue-1
[23]	Mannina G, Butler D, Benedetti L, et al. Greenhouse gas emissions from integrated urban drainage systems: Where do we stand?[J]. Journal of Hydrology, 2018, 559: 307-314. doi: 10.1016/j.jhydrol.2018.02.058
[24]	魏辅文, 孙航, 朱教君, 等. 科技支撑我国西部生物多样性保护的现状、挑战与建议[J]. 中国科学院院刊, 2025, 40(6): 1009-1015.
	[Wei Fuwen, Sun Hang, Zhu Jiaojun, et al. Biodiversity conservation in western China with scientific and technological support: Current status, challenges, and suggestions[J]. Bulletin of Chinese Academy of Sciences, 2025, 40(6): 1009-1015. ] doi: 10.3724/j.issn.1000-3045.20250402002
[25]	蒋金荷, 黄珊. 贸易新业态对绿色技术创新的影响研究--来自跨境电商综合试验区政策的证据[J]. 数量经济技术经济研究, 2024, 41(12): 133-154.
	[Jiang Jinhe, Huang Shan. Influence of new trade forms on green technology innovation: Based on the evidence of cross-border e-commerce comprehensive pilot area policy[J]. Journal of Quantitative & Technological Economics, 2024, 41(12): 133-154. ]

变量类别	变量名称	变量符号	观测值	平均值	标准差	最小值	最大值
被解释变量	黄河流域鸟类生物多样性	BID	2292	2.2488	0.7234	-0.2650	4.3384
核心解释变量	海绵城市建设政策	SCP	2292	0.1130	0.3167	0.0000	1.0000
机制变量	CO₂排放	CO₂	2292	13.0616	1.5931	6.5769	16.5057
	空气污染	PM_2.5	2292	10.9419	1.2957	6.4728	15.4352
	绿色技术创新	GTI	2292	2.3522	1.9177	0.0000	7.8766
控制变量	经济发展水平	ECD	2292	11.0341	0.8123	8.5271	17.5154
	人口密度	POI	2292	4.9810	1.9547	0.1609	10.0371
	政府干预	GOI	2292	0.3143	0.4219	0.0109	5.4318
	财政收入	FRL	2292	11.3467	1.4774	7.0975	15.3350
	产业结构	INS	2292	0.3798	0.1702	0.0182	0.9549

变量	BID
变量	（1）	（2）	（3）
SCP	0.4485^***	0.4634^***	0.4069^***
SCP	（0.0517）	（0.0624）	（0.0655）
常数项	0.0076	0.1200^***	-0.0640^***
常数项	（0.0147）	（0.0157）	（0.0147）
控制变量	是	否	是
县域固定效应	否	是	是
年份固定效应	否	是	是
样本量	2292	2292	2292

变量	BID
变量	高维固定	增加控制变量二次项	缩尾检验	1∶2分割比例	1∶6分割比例
SCP	0.4932^***	0.4042^***	0.3631^***	0.4112^***	0.3802^***
SCP	（0.0828）	（0.0661）	（0.0665）	（0.0580）	（0.0661）
常数项	-0.0822^***	-0.0639^***	-0.0568^***	-0.0688^***	-0.0622^***
常数项	（0.0150）	（0.0147）	（0.0146）	（0.0149）	（0.0147）
控制变量	是	是	是	是	是
控制变量二次项	否	是	否	否	否
县域固定效应	是	是	是	是	是
年份固定效应	是	是	是	是	是
城市×年份	是	否	否	否	否
样本量	2292	2292	2064	2292	2292

变量	PM_2.5	CO₂	GTI
变量	（1）	（2）	（3）
SCP	-0.2709^***	-0.2862^***	0.2146^**
SCP	（0.0749）	（0.0814）	（0.1084）
常数项	-0.0051	-0.0126	0.0079
常数项	（0.0091）	（0.0127）	（0.0155）
控制变量	是	是	是
县域固定效应	是	是	是
年份固定效应	是	是	是
样本量	2292	2292	2292