黄龙带水库入库水量与降水量关系研究
Study on the Relationship Between the Inflow Water Volume of Huanglongdai Reservoir and Precipitation
中文摘要：
入库水量是科学制定水资源调度方案、确保水利设施安全和发挥其经济效益的重要依据。利用黄龙带水库水源区三个观测站点1977—2020年逐月降水量和入库水量数据，研究了黄龙带水库流域降水量和入库水量的分布特征及其相关性，利用线性回归建立了逐月入库水量诊断模型。结果表明，黄龙带水库流域月降水量和入库水量存在显著的月变化，降水量4—6月最大，入库水量5—7月最大；逐月的入库水量与降水量存在一定的滞后关系，除1月入库水量只与当月降水量显著相关外，其余各月的入库水量均与当月和前1～2个月的降水量存在显著相关。建立入库水量与同期和前期降水量的线性回归模型，各月解释方差在63.3%～91.0%，具有较高的拟合能力。同时，拟合程度与降水集中度密切相关，降水集中度高的月份，诊断结果易偏低，而降水集中度低的月份，诊断结果易偏高。研究为利用降水量预测入库水量奠定了基础。
中文关键词：
黄龙带水库，降水量，入库水量，回归分析，诊断误差
KeyWords：
Huanglongdai Reservoir, precipitation, inflow water volume, regression analysis, diagnostic error
Abstract：
Inflow volume of a reservoir acts as a key reference for formulating scientific water resource regulation plans, ensuring the safety of water conservancy facilities and exerting their economic benefits. By using the monthly precipitation and inflow data from 1977 to 2020 at three observation stations in the water source area of Huanglongdai Reservoir, the distribution characteristics and correlations between the precipitation and inflow water volume are analyzed. In addition, the diagnosis model of monthly inflow volume is established with linear regression method. The results show that there are significant monthly changes in the monthly precipitation and inflow water volume of Huanglongdai Reservoir. The highest precipitation is identified during April－June, and the inflow water volume peaks from May to July. Meanwhile, there is a certain lag relationship between the monthly inflow of water volume and precipitation. Except that the inflow in January is only significantly correlated with the precipitation of the same month, the inflow of water in other months is highly related not only to the precipitation of the same month, but also to the precipitation of the previous one to two months. According to the linear regression model between the inflow water volume and the precipitation in the same period as well as the previous period, the explained variance of each month ranges between 63.3% and 90.98%, showing a high fitting ability. Moreover, the fitting degree is closely related to the precipitation concentration degree. Months with high precipitation concentration degrees are prone to lower diagnostic results, while months with low precipitation concentration degrees are accompanied with higher diagnostic results. This lays the foundation for using precipitation to predict the inflow water volume of a reservoir.