山西龙子祠岩溶泉流量动态特征与影响因素分析

Dynamic characteristics and influence factors of discharge of the Longzici karst spring in Shanxi Province

选取山西南部典型岩溶泉龙子祠泉域为研究对象,利用1954-2018年65年长时序的泉水流量、降水监测数据及1990-2018年的开采数据,分析其动态特征及响应规律。结果表明:该泉域泉流量在2.94~8.39 m^(3)·s^(-1)之间波动,多年平均流量为4.74 m^(3)·s^(-1);年内泉流量对降水补给响应有4个月的时间滞后性;开采量保持较高水平时,泉流量总体呈明显下降趋势。利用多元回归方法建立两个阶段(1954-1989年和1990-2018年)的数学模型,并对其进行模拟和误差分析,结果显示:第一阶段泉流量影响因素主要为前一年降水和前两年降水;1990年开始存在开采活动,第二阶段受降水和开采共同制约,与当年开采量负相关且受其影响最为强烈,其次受前一年和前两年降水影响;整个时间序列上流量误差为11.62%,模拟效果较好。

The Longzici spring is located in front of the Xishan mountains,13 km southwest of Linfen City,Shanxi Province.The spring area is 2,250 km2.The exposed strata are dominated by Cambrian and Ordovician,Carboniferous Permian,Triassic and Cenozoic Quaternary.Overall,the geological structure of the spring area is featured by uplift on the north and south sides,and a depression in the central part.The north is the Longzici syncline,which forms syn⁃cline mountains in topography,and the south is an underground structure sloping to the north or northeast.The re⁃charge in this area is dominated by precipitation infiltration in the exposed carbonate rock area,followed by surface runoff infiltration recharge from the Carboniferous-Permian sand shale area.The aquifer is dominated by the Cambrian-Ordovician carbonate water-bearing rock group.In order to explore the spatial-temporal variation characteristics and response laws of spring dynamics in karst water-bearing systems,this paper uses a 65-year time series of spring water flow and precipitation monitoring data from 1954 to 2018 and mining data from 1990 to 2018,and employs the stepwise analysis of multiple regression meth⁃ods to establish the response relationship models of spring discharge,precipitation and mining in stages.Then using the models we simulate and predict spring discharge over a long time sequence.The research results of this paper can provide data support for the prevention and control of the flow attenuation in the Longzici spring area.The conclusions of this article are as follows,(1)The precipitation ranges 287.2-868.7 mm with a multi-year average 541.1 mm in the spring area.More pre⁃cipitation occurred after 1990 than before,and there was a four-month time lag in the response of spring flow to precipi⁃tation replenishment during the year.The average extraction volume of karst water for many years is 3.233 million cu⁃bic meters,and the average extraction volume of pore water in the piedmont area is 8.243 million cubic meters.When the extraction volume remains at a relatively high level,the spring flow generally shows a significant downward trend.(2)The spring flow fluctuates between 2.94-8.39 m^(3)·s^(-1),and the multi-year average flow is 4.74 m^(3)·s^(-1).Before 1990,the average flow was 5.61m^(3)·s^(-1),while after 1990 it was 3.68 m^(3)·s^(-1).In addition to factors such as mining and precipitation that affect spring flow,local coal mining activities also affect spring flow dynamics,which need to be ana⁃lyzed in subsequent studies.(3)The results of multiple regression show that the main influencing factors of spring flow in the first stage(1954-1989)were the precipitations of the previous year and the previous two years.In the second stage(1990-2018),the spring flow was restricted by both precipitation and mining and was negatively correlated with the mining volume of the year,most strongly affected by it;followed by the precipitation of the previous year and the previous two years.Er⁃ror analysis shows that the flow error in the entire time series is 11.62%,implying a good simulation effect.