基于CMIP6气候模式的黄河内蒙古段未来冰情特征

Future river-ice characteristics of the Inner Mongolia reach of the Yellow River based on CMIP6 climate models

摘要

为探究气候变化背景下黄河内蒙古段未来冰情变化特征，基于多年实测冰厚资料对不同冰厚模型进行对比分析，并结合CMIP6中8个全球气候模式的逐日气温数据，模拟未来冰厚变化，分析最大冰厚日期及封冻时长演变特征，探讨黄河内蒙古段未来冰情的演变特征。结果表明：统一度日法模型精度最高（E_MA=2.81 cm，E_RMS=3.44 cm），优于Stefan模型和Zubov模型；未来各站最大冰厚与平均冰厚整体呈下降趋势，最大冰厚下降速率为0.043～0.297 cm/a，未来（2026−2100年）情景下凌汛期平均气温每升高1 ℃，最大冰厚减少5.0～6.1 cm，平均冰厚减少2.1～3.1 cm；冰厚增长速率表现为“年际波动明显、长期缓慢下降”；封冻时长平均缩短速率约为0.097、0.232和0.651 d/a，未来（2026−2100年）情景下凌汛期平均气温每升高1 ℃，封冻时长减少4.02～9.73 d。

Abstract

The Inner Mongolia reach of Yellow River is typical of a high-latitude, cold-region river section. This area is crucial for flood control and disaster reduction in the Yellow River basin because winter freeze-up and spring breakup are complex, and ice-flood disasters occur frequently. Under global climate change, rising air temperatures and intensified cold-warm fluctuations have continuously affected the formation, development, and decay of river ice, thereby altering ice thickness, freeze duration, breakup timing, and ice-flood risk. Previous studies, based mainly on historical observations, revealed the general patterns of river-ice variation, but they were insufficient to support projections of future river-ice evolution under changing climate scenarios. This study selected an ice-thickness model with high simulation accuracy and combined it with daily air temperature data from eight global climate models in the Coupled Model Intercomparison Project Phase 6 (CMIP6) to simulate future ice-thickness changes, analyze the maximum ice-thickness date and freeze duration, and investigate future river-ice conditions in the Inner Mongolia reach of the Yellow River. CMIP6 outputs generally have coarse spatial resolution and exhibit systematic bias. Direct use of the original data would underestimate accumulated negative air temperature and bias the determination of freezing thresholds, lowering the accuracy of river-ice simulation. As a result, these climate data needed to be downscaled and corrected for bias before being used in river-ice studies. The quantile delta mapping approach was chosen because it successfully reduced bias in the daily temperature series produced by the climate models. Model parameters were calibrated using observed ice-thickness data, and simulation and validation were conducted using multi-year records. Long-term observations were used to compare various ice-thickness models, and future variations in ice thickness and freeze duration were simulated using the corrected daily air temperature data from eight global climate models to examine future river-ice conditions within the study reach. The results showed that the unified degree-day method had the highest simulation accuracy, with a mean absolute error of 2.81 cm and a root mean square error of 3.44 cm, both lower than those of the Stefan et ai model and the Zubov model. Under future climate scenarios, both maximum ice thickness and mean ice thickness at all stations showed an overall decreasing trend. The decline rate of maximum ice thickness ranged from 0.043 to 0.297 cm/a. During the ice-flood season, each increase of 1 ℃ in mean air temperature corresponded to a decrease of 5.0 to 6.1 cm in maximum ice thickness and a decrease of 2.1 to 3.1 cm in mean ice thickness. The ice growth rate fluctuated from year to year, but it declined slowly over time. Freeze duration also exhibited a shortening trend, with average rates of approximately 0.097, 0.232, and 0.651 d/a under different future scenarios. From 2026 to 2100, each increase of 1 ℃ in mean air temperature during the ice-flood season corresponded to a reduction of 4.02 to 9.73 d in freeze duration. The study further suggested that breakup in the Inner Mongolia reach of the Yellow River would occur earlier in the future, and the frequency of thermal breakup would increase. Meanwhile, the river channel would tend to shift from continuous freeze-up toward segmented freeze-up. The number and area of open-water leads were also expected to expand, and the frequency and scale of ice-jam events might increase, leading to greater uncertainty in ice-related disasters. Overall, future warming would substantially reshape the river-ice regime of the Inner Mongolia reach of the Yellow River, with implications for ice-flood processes and river management. These findings offer a scientific and theoretical foundation for future ice-flood prevention, disaster reduction, and water-management decisions in the Yellow River basin.

关键词

CMIP6 / 凌汛期 / 未来冰情 / 气候变化 / 黄河内蒙古段 / 冰厚

Key words

CMIP6 / ice-flood season / future river-ice condition / climate change / Inner Mongolia reach of the Yellow River / ice thickness

参考文献

[1] Burrell B C, Beltaos S, Turcotte B. Effects of climate change on river-ice processes and ice jams[J]. International Journal of River Basin Management, 2023, 21(3): 421-441. DOI: 10.1080/15715124.2021.2007936.

[2] Beltaos S. Assessing the frequency of floods in ice-covered rivers under a changing climate: Review of methodology[J]. Geosciences, 2021, 11(12): 514. DOI: 10.3390/geosciences11120514.

[3] 孙亚翡, 王涛, 周中元. 黄河内蒙古河段凌汛期首封影响参数辨识和预报模型研究[J]. 中国水利水电科学研究院学报（中英文）, 2024, 22(2): 149-158. DOI: 10.13244/j.cnki.jiwhr.20230162.

[4] 郭春华, 朱秀芳, 张世喆, 等. 基于CMIP6的中国未来高温危险性变化评估[J]. 地球信息科学学报, 2022, 24(7): 1391-1405. DOI: 10.12082/dqxxkx.2022.210491.

[5] 王魁, 王军, 张防修. 黄河内蒙古段开河流量预测[J]. 南水北调与水利科技（中英文）, 2021, 19(6): 1167-1174. DOI: 10.13476/j.cnki.nsbdqk.2021.0119.

[6] 周中元, 王涛, 孙泽朝, 等. 黄河宁蒙段冰情变化规律及其驱动因素解析[J]. 南水北调与水利科技（中英文）, 2025, 23(3): 619-629. DOI: 10.13476/j.cnki.nsbdqk.2025.0067.

[7] Wang Tao, Guo Xinlei, Liu Jifeng, et al. Ice process simulation on hydraulic characteristics in the Yellow River[J]. Journal of Hydraulic Engineering (ASCE), 2024, 150(3): 567-578. DOI: 10.1061/JHEND8.HYENG-13679.

[8] Zhao Shuixia, Wang Wenjun, Shi Xiaohong, et al. Freeze-up ice jam formation in the river bend: A case study on the Inner Mongolia reach of Yellow River[J]. Crystals, 2021, 11(6): 631. DOI: 10.3390/cryst11060631.

[9] Liu Bin, Ji Honglan, Luo Hongchun, et al. Study on the spatiotemporal distribution characteristics and influencing factors of open water during the freezing period in the Inner Mongolia section of the Yellow River from 1986 to 2023[J]. Journal of Hydrology: Regional Studies, 2025, 59: 102423. DOI: 10.1016/j.ejrh.2025.102423.

[10] Stefan J. Über die Theorie der Eisbildung, insbesondere über die Eisbildung im Polarmeere[J]. Annalen der Physik, 1891, 278(2): 269-286. DOI: 10.1002/andp.18912780206.

[11] Zubov N N. Arctic ice[M]. San Diego: U. S. Navy Electronics Laboratory, 1963. DOI: 10.5962/bhl.title.39101.

[12] Shen H T, YAPA P D. A unified degree-day method for river ice cover thickness simulation[J]. Canadian Journal of Civil Engineering, 1985, 12(1): 54-62. DOI: 10.1139/l85-006.

[13] 练继建, 赵新. 静动水冰厚生长消融全过程的辐射冰冻度-日法预测研究[J]. 水利学报, 2011, 42(11): 1261-1267. DOI: 10.13243/j.cnki.slxb.2011.11.006.

[14] 王军, 侯智星. 黄河内蒙古段冰盖厚度模拟[J]. 南水北调与水利科技（中英文）, 2021, 19(5): 921-929. DOI: 10.13476/j.cnki.nsbdqk.2021.0096.

[15] Fukś M, Jasik A, Kasperski J. Changes in river ice cover in the context of climate change: A review[J]. Aquatic Sciences, 2023, 85: 74. DOI: 10.1007/s00027-023-01011-4.

[16] 范雪薇, 缪驰远, 苟娇娇, 等. 国际耦合模式比较计划及其模拟能力研究进展[J]. 地理科学进展, 2023, 42(6): 1204-1215. DOI: 10.18306/dlkxjz.2023.06.015.

[17] Lin Yu, Lü Haishen, Lindenschmidt K E, et al. Future global river ice in CMIP6 models under climate change[J]. Journal of Applied Meteorology and Climatology, 2024, 63(10): 1191-1206. DOI: 10.1175/JAMC-D-23-0208.1.

[18] Yang X, Pavelsky T M, Allen G H. The past and future of global river ice[J]. Nature, 2020, 577(7788): 69-73. DOI: 10.1038/s41586-019-1848-1.

[19] Liu Mingwen, Lindenschmidt K E, Lü Haishen, et al. Changes in future ice-jam flood severity of a regulated cascade reservoir-river system[J]. Journal of Hydrology, 2025, 659: 133307. DOI: 10.1016/j.jhydrol.2025.133307.

[20] 王佳琦, 王涛, 孙宇飞, 等. 基于CMIP6的黄河流域三级水资源区未来气候变化预估[J]. 南水北调与水利科技（中英文）, 2025, 23(2): 242-250. DOI: 10.13476/j.cnki.nsbdqk.2025.0031.

[21] 周中元, 王涛, 孙亚翡, 等. 黄河未来气候背景下冰情特征演变分析[J]. 水利学报, 2024, 55(3): 355-366. DOI: 10.13243/j.cnki.slxb.20230388.

[22] 孙令东. 内蒙古境内黄河封（开）河预测分析[J]. 中国防汛抗旱, 2025, 35(6): 99-102. DOI: 10.16867/j.issn.1673-9264.2024211.

[23] 陈银太, 张末, 杨会颖, 等. 2019−2020年度黄河凌情及防御措施[J]. 中国防汛抗旱, 2020, 30(5): 13-17. DOI: 10.16867/j.issn.1673-9264.2020104.

[24] 杨瑞敏, 周茅先. 1850−2100年北极阿拉斯加地区最大河冰厚度时空变化特征[J]. 冰川冻土, 2023, 45(2): 468-479. DOI: 10.7522/j.issn.1000-0240.2023.0036.

[25] Hou Zhixin, Wang Jun, Sui Jueyi, et al. Discriminant analysis of the freeze-up and break-up conditions in the Inner Mongolia reach of the Yellow River[J]. Journal of Water and Climate Change, 2023, 14(9): 3166-3177. DOI: 10.2166/wcc.2023.195.

[26] 赵水霞, 陈晓俊, 王文君, 等. 黄河内蒙古段封河期水面-大气非线性热力学模型探究[J]. 水资源与水工程学报, 2021, 32(3): 130-136. DOI: 10.11705/j.issn.1672-643X.2021.03.18.

[27] 张璐, 芦璐, 刘溯宇, 等. 黄河内蒙古段封河日动态加权集成预报方法[J]. 南水北调与水利科技（中英文）, 2025, 23(6): 1349-1357. DOI: 10.13476/j.cnki.nsbdqk.2025.0134.

[28] Zhang Lu, Liu Suyu, Fan Minhao, et al. Machine learning prediction of river freeze-up dates under human interventions: Insights from the Ningxia-Inner Mongolia reach of the Yellow River[J]. Water, 2025, 17(23): 3357. DOI: 10.3390/w17233357.

[29] 田福昌, 苑希民, 何立新, 等. 寒区河道-堤防-泛区凌汛灾害风险评估防控研究进展[J]. 水利学报, 2022, 53(5): 549-559. DOI: 10.13243/j.cnki.slxb.20210862.

[30] Deng Yu, Duan Kaidi, Zhu Yong. Study on the evolution law of ice-water transport during the ice flood period in the Shisifen section of the Yellow River in Inner Mongolia[J]. Applied Sciences, 2025, 15(18): 10270. DOI: 10.3390/app151810270.

[31] 毛宇鑫, 冀鸿兰, 杨震, 等. 基于InSAR技术的黄河内蒙古段河岸形变监测与影响因素分析[J]. 农业工程学报, 2025, 41(5): 87-95. DOI: 10.11975/j.issn.1002-6819.202409005.

[32] Ma Lan, Li Chao, Yao Zhanguan, et al. Numerical simulation of typical river closure process and sensitivity analysis of influencing factors[J]. Hydrology, 2026, 13(1): 29. DOI: 10.3390/hydrology13010029.

[33] 王军, 李建昆, 苗平. 黄河凌汛和伏汛引水对库布齐沙漠北缘生态格局及其耗水变化影响研究[J]. 水利学报, 2023, 54(5): 541-553. DOI: 10.13243/j.cnki.slxb.20220956.

[34] 李超, 冀雪飞, 赵水霞, 等. 黄河头道拐段冰输移及堆积过程数值模拟[J]. 水利学报, 2023, 54(4): 474-485. DOI: 10.13243/j.cnki.slxb.20220629.

[35] Wu Heng, Lei Huimin, Lu Weiwei, et al. Future changes in precipitation over the upper Yangtze River basin based on bias correction spatial downscaling of models from CMIP6[J]. Environmental Research Communications, 2022, 4(4): 045002. DOI: 10.1088/2515-7620/ac620e.

[36] 李艺航, 张洪波, 刘芳, 等. 黄河流域秦岭北麓河川基流变化解析及对气候变化的响应[J]. 南水北调与水利科技（中英文）, 2025, 23(5): 1234-1248. DOI: 10.13476/j.cnki.nsbdqk.2025.0123.

[37] 谢文强, 王双双, 延晓冬. CMIP6全球气候模式对中国年平均日最高气温和最低气温模拟的评估[J]. 气候与环境研究, 2022, 27(1): 63-78. DOI: 10.3878/j.issn.1006-9585.2021.21027.

[38] Wang Tian, Tu Xinjun, Singh V P, et al. A composite index coupling five key elements of water cycle for drought analysis in Pearl River basin, China[J]. Journal of Environmental Management, 2022, 320: 115813. DOI: 10.1016/j.jenvman.2022.115813.

[39] 刘逸龙, 郭斌, 孔冬贤, 等. 基于CMIP6多模式的黄河流域降雨侵蚀力未来变化预估[J]. 中国生态农业学报（中英文）, 2025, 33(10): 1967-1976. DOI: 10.12357/cjea.20250295.

[40] Wang Fang, Tian Di. On deep learning-based bias correction and downscaling of multiple climate models simulations[J]. Climate Dynamics, 2022, 59(11-12): 3451-3468. DOI: 10.1007/s00382-022-06277-2.

[41] Jose D M, Dwarakish G S. Bias correction and trend analysis of temperature data by a high-resolution CMIP6 model over a tropical river basin[J]. Asia-Pacific Journal of Atmospheric Sciences, 2022, 58(1): 97-115. DOI: 10.1007/s13143-021-00240-7.

基金资助

国家自然科学基金项目(52379014)

国家自然科学基金联合基金项目(U23A2012)