| 24 | 0 | 25 |
| 下载次数 | 被引频次 | 阅读次数 |
【目的】针对变化环境下传统静态风险模型难以刻画多水源系统联合干旱响应非平稳特征的不足，提出一套耦合非平稳边缘分布与时变相依结构的“双重非平稳”动态分析框架，并以青海省引大济湟工程“大通河—湟水”双水源系统为例，揭示其联合干旱风险的动态演化规律。【方法】基于1956—2010年逐月径流序列，首先采用广义加性位置尺度形状模型（GAMLSS）构建双水源月径流最优边缘分布模型，并识别其非平稳结构；随后基于最优边缘分布函数对双水源径流序列进行概率积分变换（PIT），获得标准均匀概率变量，在此基础上构建静态与时变Copula模型，并引入气候指数（NAO、PDO、NINO3.4）及区域干旱指数（SPEI-3）等协变量刻画双水源相依结构动态变化；最后计算双水源同步偏枯（双枯）联合概率，量化系统联合干旱风险。【结果】结果表明：24组月尺度序列中非平稳模型占54.2%；双水源相依关系存在显著季节差异，最优Copula族以Gaussian Copula为主，8个月份优选为时变模型，占全年月份的66.7%，表明双水源相依结构具有明显动态性；SPEI-3在部分关键月份对双枯风险识别具有较强解释作用，其中5月模型识别效果最佳（AUC=0.78）；双枯风险整体呈现夏秋季较高、冬春季较低的季节分异特征。【结论】提出的双重非平稳耦合框架突破了传统Copula静态相依假设的局限，能够有效捕捉多水源系统在变化环境下的动态联合风险响应，为跨流域调水工程的适应性调度和供水安全预警提供了新的方法论支撑和定量依据。
Abstract：[Objective] To address the limitation of traditional static risk models in characterizing the non-stationary joint drought response of multi-source water systems under changing environments， a “dual non-stationary” dynamic analysis framework coupling non-stationary marginal distributions with time-varying dependence structures is proposed. The Datong River–Huangshui River dual-source system of the Yinda-Jihuang Project in Qinghai Province is used as a case study to reveal the dynamic evolution patterns of its joint drought risk. [Methods] Based on monthly runoff series from 1956 to 2010， the generalized additive models for location， scale， and shape (GAMLSS) were first used to construct optimal marginal distribution models for monthly runoff series of the two water sources and to identify their non-stationary structures. Probability integral transform (PIT) was then applied to the runoff series of the two water sources using the optimal marginal distribution functions to obtain standard uniform probability variables. Static and time-varying Copula models were further developed， and climatic indices (NAO， PDO， and NINO3.4) and the regional drought index SPEI-3 were introduced as covariates to characterize dynamic changes in the dependence structure between the two water sources. Finally， the joint probability of synchronous low-flow events of the two water sources， referred to as dual-drought， was calculated to quantify the joint drought risk of the system. [Results] The results showed that non-stationary models accounted for 54.2% of the 24 monthly runoff series. The dependence between the two water sources exhibited significant seasonal variability， with Gaussian Copula being the dominant Copula family. Time-varying models were preferred in eight months， accounting for 66.7% of the year， indicating pronounced dynamics in the dependence structure between the two water sources. SPEI-3 showed strong explanatory power for dual-drought risk identification in several key months， with the best performance occurring in May (AUC = 0.78). Overall， dual-drought risk was higher in summer and autumn and lower in winter and spring. [Conclusion] The proposed dual non-stationary coupling framework overcomes the limitation of static dependence assumptions in traditional Copula models and effectively captures the dynamic joint risk response of multi-source water systems under changing environments， providing methodological support and quantitative evidence for adaptive operation and water supply security early warning in inter-basin water transfer projects.
[1] XIONG J H， YANG Y T. Climate change and hydrological extremes[J]. Current Climate Change Reports， 2024， 11(1): 1.
[2] 夏琳蓉， 唐颖， 隆院男， 等. 基于GAMLSS模型的资水流域最大日径流研究[J]. 水利水电技术(中英文)， 2026， 57(1): 183-191.
[3] RAHMAN G， JUNG M K， KIM T W， et al. Drought impact， vulnerability， risk assessment， management and mitigation under climate change: A comprehensive review[J]. KSCE Journal of Civil Engineering， 2025， 29(1): 100120.
[4] VICENTE-SERRANO S M， BEGUERÍA S， LÓPEZ-MORENO J I. A multiscalar drought index sensitive to global warming: The standardized precipitation evapotranspiration index[J]. Journal of Climate， 2010， 23(7): 1696-1718.
[5] EDWARDS E， SANCHEZ L， SEKHRI S. The economics of drought[J]. Annual Review of Resource Economics， 2024， 16: 105-124.
[6] 朱晓钰， 穆振侠， 宋志林， 等. 基于GAMLSS模型的吐鲁番阿拉沟河流域非平稳水文干旱分析[J]. 水利水电技术(中英文)， 2026， 57(7): 17-32.
[7] FISCHER S， SCHUMANN A H. Temporal changes in the frequency of flood types and their impact on flood statistics[J]. Journal of Hydrology X， 2024， 22: 100171.
[8] MILLY P C D， BETANCOURT J， FALKENMARK M， et al. Stationarity is dead: Whither water management?[J]. Science， 2008， 319(5863): 573-574.
[9] TWINOMUHANGI M B， BAMUTAZE Y， KABENGE I， et al. Analysis of stationary and non-stationary hydrological extremes under a changing environment: A systematic review[J]. HydroResearch， 2025， 8: 332-350.
[10] CHEN C， PENG T， SINGH V P， et al. Assessment of dynamic hydrological drought risk from a non-stationary perspective[J]. Hydrological Processes， 2024， 38(8): e15267.
[11] SERINALDI F， KILSBY C G. Stationarity is undead: Uncertainty dominates the distribution of extremes[J]. Advances in Water Resources， 2015， 77: 17-36.
[12] KUNDZEWICZ Z W， ROBSON A J. Change detection in hydrological records: A review of the methodology/Revue méthodologique de la détection de changements dans les chroniques hydrologiques[J]. Hydrological Sciences Journal， 2004， 49(1): 7-19.
[13] VILLARINI G， SERINALDI F， SMITH J A， et al. On the stationarity of annual flood peaks in the continental United States during the 20th century[J]. Water Resources Research， 2009， 45(8): 2008WR007645.
[14] ZHANG D D， YAN D H， WANG Y C， et al. GAMLSS-based nonstationary modeling of extreme precipitation in Beijing–Tianjin–Hebei region of China[J]. Natural Hazards， 2015， 77(2): 1037-1053.
[15] YU J， KIM T W， PARK D H. Future hydrological drought risk assessment based on nonstationary joint drought management index[J]. Water， 2019， 11(3): 532.
[16] RIGBY R A， STASINOPOULOS D M. Generalized additive models for location， scale and shape[J]. Journal of the Royal Statistical Society Series C: Applied Statistics， 2005， 54(3): 507-554.
[17] 冷吉卫， 马凯， 李运刚， 等. 基于GAMLSS模型的元江—红河上游径流非一致性研究[J]. 应用基础与工程科学学报， 2025， 33(2): 324-337.
[18] NIU B， LI Y， LIU D L， et al. Improved identification and monitoring of meteorological， agricultural， and hydrological droughts using the modified nonstationary drought indices in the Yellow River Basin of China[J]. Journal of Hydrology， 2024， 643: 131788.
[19] ZHONG H Y， LIAO T， FANG G H， et al. Exploring optimal joint operating rules for large-scale inter-basin water transfer projects with multiple water sources， diversion routes， and water demand areas[J]. Journal of Hydrology: Regional Studies， 2023， 49: 101504.
[20] NAZEMI A， WHEATER H S. On inclusion of water resource management in Earth system models–Part 1: Problem definition and representation of water demand[J]. Hydrology and Earth System Sciences， 2015， 19(1): 33-61.
[21] NAZEMI A， WHEATER H S. On inclusion of water resource management in Earth system models–Part 2: Representation of water supply and allocation and opportunities for improved modeling[J]. Hydrology and Earth System Sciences， 2015， 19(1): 63-90.
[22] 刘章君， 郭生练， 许新发， 等. Copula函数在水文水资源中的研究进展与述评[J]. 水科学进展， 2021， 32(1): 148-159.
[23] 张润， 陈伏龙， 何朝飞， 等. 基于Vine Copula的非一致性水文干旱动态风险评估模型[J]. 农业工程学报， 2026， 42(5): 147-157.
[24] MIRABBASI R， FAKHERI-FARD A， DINPASHOH Y. Bivariate drought frequency analysis using the copula method[J]. Theoretical and Applied Climatology， 2012， 108(1): 191-206.
[25] SHIAU J T. Fitting drought duration and severity with two-dimensional copulas[J]. Water Resources Management， 2006， 20(5): 795-815.
[26] KIM W， KANG S， WOO S， et al. Bivariate copula framework for drought risk assessment under climate change and practical water supply systems[J]. Natural Hazards， 2025， 121(18): 22067-22094.
[27] HAO Z C， AGHAKOUCHAK A. Multivariate Standardized Drought Index: A parametric multi-index model[J]. Advances in Water Resources， 2013， 57: 12-18.
[28] HAO Z C， AGHAKOUCHAK A. A nonparametric multivariate multi-index drought monitoring framework[J]. Journal of Hydrometeorology， 2014， 15(1): 89-101.
[29] SARHADI A， BURN D H， CONCEPCIÓN AUSÍN M， et al. Time-varying nonstationary multivariate risk analysis using a dynamic Bayesian copula[J]. Water Resources Research， 2016， 52(3): 2327-2349.
[30] LIU W L， ZHANG J R， ZHOU Z X， et al. Hydrological drought dynamic using copula functions and drought center migration in the Ganjiang river basin[J]. Scientific Reports， 2025， 15: 39209.
[31] ZHONG H Y， WU S Y， YAN M， et al. Derivation of optimal scheduling strategies for large-scale inter-basin water transfer projects considering streamflow wet-dry encounter uncertainty of multiple reservoirs[J]. Journal of Hydrology， 2025， 662: 133784.
[32] MING B， ZHONG H Y， ZHANG W， et al. Deriving operating rules for inter-basin water transfer projects incorporating a scenario reduction strategy[J]. Journal of Hydrology， 2023， 624: 129854.
[33] 董军， 胡进宝， 魏国孝. 大通河流域径流变化及特征分析[J]. 水资源与水工程学报， 2018， 29(6): 75-80.
[34] 杨青， 马世震， 高晓鹏. 青海省湟水流域水环境现状评价[J]. 青海环境， 1994， 4(2): 63-68.
[35] WANG Y F， LIU G X， GUO E L. Spatial distribution and temporal variation of drought in Inner Mongolia during 1901–2014 using Standardized Precipitation Evapotranspiration Index[J]. Science of the Total Environment， 2019， 654: 850-862.
[36] MARIANI R V， LYRA G B， ABREU M C， et al. Validação da SPEIbase para a região sudeste brasileira[J]. Revista Brasileira de Meteorologia， 2024， 39: e339230056.
[37] PASCALE S， RAGONE F. Widespread multi-year droughts in Italy: Identification and causes of development[J]. International Journal of Climatology， 2025， 45(8): e8827.
[38] 张巧凤， 于红博， 黄方. 蒙古高原干旱时空特征及对植被物候的累积影响[J]. 干旱区研究， 2024， 41(9): 1548-1559.
[39] AKAIKE H. A new look at the statistical model identification[J]. IEEE Transactions on Automatic Control， 1974， 19(6): 716-723.
[40] CAMMALLERI C， DE MICHELE C， TORETI A. Exploring the joint probability of precipitation and soil moisture over Europe using copulas[J]. Hydrology and Earth System Sciences， 2024， 28(1): 103-115.
[41] FAWCETT T. An introduction to ROC analysis[J]. Pattern Recognition Letters， 2006， 27(8): 861-874.
[42] AL MOTERI M， ALROWAIS F， MTOUAA W， et al. An enhanced drought forecasting in coastal arid regions using deep learning approach with evaporation index[J]. Environmental Research， 2024， 246: 118171.
[43] SUJON K M， HASSAN R， CHOI K， et al. Accuracy， precision， recall， f1-score， or MCC? Empirical evidence from advanced statistics， ML， and XAI for evaluating business predictive models[J]. Journal of Big Data， 2025， 12(1): 268.
[44] THARWAT A. Classification assessment methods[J]. Applied Computing and Informatics， 2021， 17(1): 168-192.
[45] CUO L， ZHANG Y X， WANG Q C， et al. Climate change on the northern Tibetan Plateau during 1957–2009: Spatial patterns and possible mechanisms[J]. Journal of Climate， 2013， 26(1): 85-109.
[46] XIANG Y Y， WANG T， WANG H J， et al. Influence of the Pacific decadal oscillation on winter temperatures and precipitation over the southern Tibetan Plateau[J]. Journal of Geophysical Research: Atmospheres， 2024， 129(6): e2023JD038653.
[47] ZHANG Z X， ZHANG L， LIU Y F， et al. Responses of annual streamflow variability to annual precipitation， extreme climate events and large-scale climate phenomena in the Qinghai-Tibet Plateau[J]. Journal of Hydrology， 2024， 632: 130969.
[48] JIANG X W， ZHANG T T， TAM C Y， et al. Impacts of ENSO and IOD on snow depth over the Tibetan Plateau: Roles of convections over the western north Pacific and Indian Ocean[J]. Journal of Geophysical Research: Atmospheres， 2019， 124(22): 11961-11975.
[49] 米子甲， 李想， 沈延青， 等. 基于Copula函数和蒙特卡洛法的黄河上游多站径流相关分析与随机模拟[J]. 水利水电技术(中英文)， 2025， 56(12): 67-86.
[50] 胡江， 李星， 朱清帅， 等. 国家水网背景下调水工程运行风险评估技术研究进展[J]. 水科学进展， 2025， 36(6): 1042-1057.
[51] LIU R T， YIN J B， SLATER L， et al. Machine-learning-constrained projection of bivariate hydrological drought magnitudes and socioeconomic risks over China[J]. Hydrology and Earth System Sciences， 2024， 28(14): 3305-3326.
基本信息:
中图分类号:TV68;P426.616
引用信息:
[1]王钰升,游进军,马真臻,等.双重非平稳框架下跨流域调水双水源联合干旱风险评估[J].水利水电技术(中英文)().
基金信息:
国家自然科学基金项目(52079143); 青海省自然科学基金项目(2025-ZJ-908Q); 国家重点研发计划项目(2022YFC3204601，2024YFC3210901)
2026-10-14
2026-10-14
2026-10-14