基于干旱和水位特征构建水库人工增雨需求指数

doi:10.11755/j.issn.1006-7639(2023)-02-0341

摘要/Abstract

摘要：

为定量化描述水库人工增雨需求，以浙江兰溪芝堰水库为研究对象，基于降水、径流和水位历史观测数据，采用实际概率分布阈值法确定逐月不同等级水库水位指标（Water Level Index， WLI）百分位阈值，通过熵权法联合表征气象干旱的标准化降水指数（Standardized Precipitation Index， SPI）和表征水文干旱的标准化径流指数（Standardized Streamflow Index， SSI）构建水库整体干旱指数（Drought Index， DI），综合WLI和DI构建水库人工增雨需求指数（Demand Level Index， DLI）等级，研究WLI、DI和DLI时间特征，结合水库历史记录分析DLI的适用性，结果表明：（1）确定的逐月不同等级WLI百分位阈值能够精细化表征水库在一年内不同时期的缺水程度。（2）1990—2019年气象干旱无明显变化，水文干旱程度加大，其中春季变旱趋势最明显。（3）气象（水文）干旱在夏秋季发生总频率为33.9%（35.0%），高于冬春两季的30.0%（28.3%）。重、特旱在春季的发生频率最高，气象和水文干旱发生频率分别为11.2%和10.0%。水文干旱与气象干旱的时滞性不明显，水文干旱平均历时和平均烈度均高于气象干旱，具有更高的危害性。（4）不同等级的DLI年际分布与WLI较为相近，季节分布与DI分布相似。与2004年之前相比，2004年之后增雨需求出现更为频繁，持续存在时间更长。增雨需求在夏季占比最高，为40.0%，但高度、强烈增雨需求在春季占比最高，为14.4%。（5）构建的DLI等级可以较好地表征水库实际需求，当DLI等级大于等于4持续多个月时，水库可能会出现用水紧张并应采取相关紧急措施。

关键词: 水库增蓄, 人工增雨, 需求指数, 水文干旱, 气象干旱

Abstract:

In order to provide a quantitative method to describe artificial precipitation demand, taking Zhiyan reservoir in Lanxi as the research object, based on precipitation, runoff and water level data, the percentile threshold of monthly water level index (WLI) with different grades of the reservoir was derived from real sample probability distribution of water level, the drought index (DI) was constructed by using the entropy weight method combined with the standardized precipitation index (SPI) and the standardized streamflow index (SSI), then WLI was integrated with DI to generate the demand level index (DLI) to describe artificial precipitation demand of reservoir objectively. The temporal characteristics of WLI, DI and DLI were studied, the applicability of DLI was analyzed based on the reservoir history records, the main conclusions are as follows: (1) The constructed percentile threshold of monthly WLI with different grades was able to reflect the water shortage of the reservoir precisely in different periods of a year. (2) There was no significant change in meteorological drought from 1990 to 2019, meanwhile hydrological drought showed an increasing trend, and the increasing trend was most obvious in spring. (3) The total occurrence frequency of meteorological (hydrological) drought in summer and autumn was 33.9% (35.0%), it is higher than that （30.0% (28.3%)） in winter and spring. The occurrence frequency of severe and extreme drought in spring was the highest, and meteorological and hydrological drought accounted for 11.2% and 10.0% respectively in spring. Hydrological drought did not lag behind and had a more serious effect than meteorological drought. (4) The inter-annual distribution of DLI was similar to that of WLI, and the seasonal distribution of DLI was similar to that of DI. Artificial precipitation demand appeared more frequently and last longer in the years after 2004 than before 2004. Demand occurred most frequently in summer, accounting for 40.0%, however the demand of high and very high level occurred most frequently in spring, accounting for 14.4%. (5) The integrated DLI grades could well reflect the actual demand of the reservoir, and when DLI grade was greater than or equal to 4 for several months, the reservoir might be short of water and emergency measures required to be taken.

Key words: reservoir storage increasing, artificial precipitation, demand level index, hydrological drought, meteorological drought

中图分类号:

P481

姜舒婕, 程莹, 方楠, 周毓荃, 单中华, 张磊. 基于干旱和水位特征构建水库人工增雨需求指数[J]. 干旱气象, 2023, 41(2): 341-349.

JIANG Shujie, CHENG Ying, FANG Nan, ZHOU Yuquan, SHAN Zhonghua, ZHANG Lei. Construction of artificial precipitation demand level index of the reservoir based on drought and water level characteristics[J]. Journal of Arid Meteorology, 2023, 41(2): 341-349.

图/表 11

表1 WLI等级划分

Tab.1 Classification of WLI grades

WLI等级	百分位值
1	（60%，100%]
2	（30%，60%]
3	（15%，30%]
4	（5%，15%]
5	（0，5%]

表2 SPI和SSI等级分类

Tab.2 Classification of standardized precipitation index and standardized streamflow index grades

干旱等级	干旱分类	$S P I$ 值	$S S I$ 值
1	无旱	-0.5<SPI	-0.5<SSI
2	轻旱	-1<SPI≤-0.5	-1< SSI≤-0.5
3	中旱	-1.5<SPI≤-1	-1.5< SSI≤-1
4	重旱	-2.0<SPI≤-1.5	-2.0< SSI≤-1.5
5	特旱	SPI≤-2.0	SSI≤-2.0

表2 SPI和SSI等级分类

Tab.2 Classification of standardized precipitation index and standardized streamflow index grades

干旱等级	干旱分类	$S P I$ 值	$S S I$ 值
1	无旱	-0.5<SPI	-0.5<SSI
2	轻旱	-1<SPI≤-0.5	-1< SSI≤-0.5
3	中旱	-1.5<SPI≤-1	-1.5< SSI≤-1
4	重旱	-2.0<SPI≤-1.5	-2.0< SSI≤-1.5
5	特旱	SPI≤-2.0	SSI≤-2.0

图1 兰溪芝堰水库不同等级WLI的百分位阈值和平均水位月际变化

Fig.1 Monthly variation of average water level and percentile threshold of WLI with different grades in Zhiyan reservoir in Lanxi

图2 1990—2019年不同等级WLI频率年际分布和标准化年平均水位

Fig.2 The inter-annual distribution of frequency of WLI with different grades and normalized annual mean water level from 1990 to 2019

图3 1990—2019年12个月时间尺度SPI（a）和SSI（b）年际变化及M-K检验

Fig.3 The inter-annual variation of SPI （a）， SSI （b） at 12 month time scale and M-K test from 1990 to 2019

图4 1990—2019年不同等级气象干旱发生频率年际分布和标准化年降水（a），不同等级水文干旱发生频率年际分布和标准化年入流（b）

Fig.4 The inter-annual distribution of occurrence frequency of meteorological drought with different grades and normalized annual precipitation (a), inter-annual distribution of occurrence frequency of hydrological drought with different grades and normalized annual inflow (b) from 1990 to 2019

图5 1990—2019年不同等级气象（a、c）、水文（b、d）干旱发生频率季节（a、b）及月际（c、d）分布

Fig.5 Seasonal （a，b） and monthly （c，d） distribution of occurrence frequency of meteorological （a，c） and hydrological （b，d） drought with different grades from 1990 to 2019

表3 DI等级分类

Tab.3 Classification of DI grades

干旱等级	干旱分类	$D I$ 值
1	无旱	-0.5<DI
2	轻旱	-1< DI≤-0.5
3	中旱	-1.5< DI≤-1
4	重旱	-2.0< DI≤-1.5
5	特旱	DI≤-2.0

表3 DI等级分类

Tab.3 Classification of DI grades

干旱等级	干旱分类	$D I$ 值
1	无旱	-0.5<DI
2	轻旱	-1< DI≤-0.5
3	中旱	-1.5< DI≤-1
4	重旱	-2.0< DI≤-1.5
5	特旱	DI≤-2.0

表4 DLI等级与增雨需求程度对应表

Tab.4 The demand level index grades and corresponding demand degree of artificial precipitation enhancement

DLI等级	WLI等级	DI等级	增雨需求程度
1	1~2	1~2	无需求
2	1	3~5	轻度需求
	2	3
	3	1~2
3	2	4~5	中度需求
3	3	3~4	中度需求
4	3	5	高度需求
	4	1~3
	5	1
5	4	4~5	强烈需求
5	5	2~5	强烈需求

图6 1990年1月至2019年12月逐月不同等级DI（a）和DLI（b）分布

Fig.6 Monthly distribution of DI （a） and DLI （b） with different grades from January 1990 to December 2019

图7 1990—2019年不同等级DI（a）和DLI（b）频率季节分布

Fig.7 Seasonal distribution of frequency of DI （a） and DLI （b） with different grades from 1990 to 2019

参考文献 34

[1]	白先达, 2013. 桂林干旱风险评估及人工增雨抗旱研究[J]. 气象, 39(10):1 369-1 373.
[2]	曹欣荣, 2013. 浙江省气象干旱规律分析研究[D]. 杭州: 浙江大学.
[3]	曹永强, 苏阳, 张兰霞, 等, 2012. 基于AWTP指数的浙江省干旱规律时空分析[J]. 自然资源学报, 27(7):1 233-1 240.
[4]	陈再清, 侯威, 左冬冬, 等, 2016. 基于修订Copula函数的中国干旱特征研究[J]. 干旱气象, 34(2):213-222. DOI
[5]	樊高峰, 苗长明, 毛燕军, 等, 2008. 浙江干旱特征及其与区域气候变化关系[J]. 气象科技, 36(2):180-184.
[6]	樊高峰, 苗长明, 毛裕定, 2006. 干旱指标及其在浙江省干旱监测分析中的应用[J]. 气象, 32(2):70-74.
[7]	郭广芬, 周月华, 史瑞琴, 等, 2009. 湖北省暴雨洪涝致灾指标研究[J]. 暴雨灾害, 28(4):71-75.
[8]	金新芽, 邵学强, 劳国民, 2012. 浙江省历史旱情评估及成因分析[J]. 水文, 32(6):75-79.
[9]	雷江群, 黄强, 王义民, 等, 2014. 基于可变模糊评价法的渭河流域综合干旱分区研究[J]. 水利学报, 45(5):574-584.
[10]	李庆祥, 黄嘉佑, 2011. 对我国极端高温事件阈值的探讨[J]. 应用气象学报, 22(2):138-144.
[11]	李忆平, 李耀辉, 2017. 气象干旱指数在中国的适应性研究进展[J]. 干旱气象, 35(5):709-723. DOI
[12]	刘永佳, 黄生志, 方伟, 等, 2021. 不同季节气象干旱向水文干旱的传播及其动态变化[J]. 水利学报, 52(1):93-102.
[13]	马鹏里, 韩兰英, 张旭东, 等, 2019. 气候变暖背景下中国干旱变化的区域特征[J]. 中国沙漠, 39(6):209-215. DOI
[14]	倪深海, 顾颖, 彭岳津, 等, 2019. 近七十年中国干旱灾害时空格局及演变[J]. 自然灾害学报, 28(6):176-181.
[15]	国家气候中心, 2017. 气象干旱等级:GB/T 20481—2017[S]. 北京: 中国标准出版社.
[16]	孙洪泉, 吕娟, 苏志诚, 等, 2017. 分位数法对多指标干旱等级划分一致性的作用[J]. 灾害学, 32(2):13-17,53.
[17]	王本德, 周惠成, 卢迪, 2016. 我国水库(群)调度理论方法研究应用现状与展望[J]. 水利学报, 47(3),337-345.
[18]	王健疆, 马浩, 余丽萍, 等, 2021. 2019年浙江省秋旱大气环流特征分析[J]. 干旱气象, 39(1):1-7.
[19]	王可法, 张旭晖, 郭松平, 等, 2008. 改善农业干旱的人工增雨作业需求指数[C]// 中国气象学会.长三角气象科技创新论坛论文集. 上海: 上海市气象学会:628-631.
[20]	王晓峰, 张园, 冯晓明, 等, 2017. 基于游程理论和Copula函数的干旱特征分析及应用[J]. 农业工程学报, 33(10):206-214.
[21]	王义民, 屠子倩, 畅建霞, 等, 2020. 水库抗旱调度分期特征确定方法研究[J]. 自然灾害学报, 29(1),112-120.
[22]	文佐, 尹义星, 邹睿, 等, 2021. 淮河流域1960—2014年水文、气象干旱指数及其与历史旱灾的关系[J]. 中国农村水利水电,(2):71-77.
[23]	吴杰峰, 陈兴伟, 高路, 2017. 水文干旱对气象干旱的响应及其临界条件[J]. 灾害学, 32(1):199-204.
[24]	杨国庆, 王佳真, 孙萌萌, 2019. 基于标准化降水指数的沧州干旱时空特征[J]. 干旱气象, 37(2):218-225.
[25]	张超, 罗伯良, 2021. 湖南夏秋季持续性区域气象干旱的时空特征[J]. 干旱气象, 39(2):193-202.
[26]	张中平, 刘高平, 刘洪民, 等, 2007. 水库蓄水人工增雨业务系统[J]. 气象科技, 35(5):687-691.
[27]	赵福年, 王润元, 王莺, 等, 2018. 干旱过程、时空尺度及干旱指数构建机制的探讨[J]. 灾害学, 33(4):32-39.
[28]	FALKENMARK M, LUNDQUIST J, WIDSTRAND C, 2010. Macro-scale water scarcity requires micro-scale approaches: aspects of vulnerability in semi-arid development[J]. Natural Resources Forum, 13(4):258-267. DOI URL
[29]	KONG M, WANG G, WU Y, et al, 2021. A nationwide analysis of water scarcity and cloud seeding demand levels from analyzing water utilization data, agricultural drought maps, and local conditions in China mainland[J]. Earth and Space Science, 8, e2020EA001477. DOI:10.1029/2020EA001477. DOI
[30]	SHAO D, LI X, GU W, 2015. A method for temporary water scarcity analysis in humid region under droughts condition[J]. Water Resources Management, 29(10): 3 823-3 839. DOI URL
[31]	SULLIVAN C A, 2001. The potential for calculating a meaningful water poverty index[J]. Water International, 26(4):471-480. DOI URL
[32]	WILHITE D A, 2000. Drought: a global assessment[M]. New York: Routledge.
[33]	YEVJEVICH V, 1967. An objective approach to definition and investigations of continential hydrologic drought[D]. Fort Collins: Colorado State University:10-40.
[34]	ZHAI P M, ZHANG X B, WAN H, et al, 2005. Trends in total precipitation and frequency of daily precipitation extremes over China[J]. Journal of Climate, 18(7): 1 096-1 108. DOI URL