中文摘要：

针对分层水库垂向水温分布，提出一种估算水温的新方法——Logistic曲线法。利用新安江水库为期1a的原型观测资料，采用SPSS软件率定并优选模型参数，结果表明拟合精度较高。在此基础上，利用收集到的其他水库实测资料验证了模型的适用性，发现该模型对不同类型（混合型、稳定分层型）、不同时期（升温期、稳定分层期）的水库垂向水温分布有较好的拟合效果。此外，探讨了该模型应用于预测水库水温时的模型参数确定方法及步骤。

英文摘要：

Based on the layered vertical distribution of water temperature in reservoirs, a new method, which is called the logistic curve method, is proposed to estimate the water temperature. In this study, an S-type logistic curve equation was used to fit one year of observed data from the Xin＇ anjiang Reservoir. The results show a high accuracy of fitting by determining the parameters in the empirical equation （s） using the nonlinear regression method in the SPSS software. Meanwhile, the applicability of the empirical equation （s） was analyzed using observed data from other reservoirs. The results show that the empirical equation （s） can well fit the vertical distribution of water temperature in the reservoir for different types （the mixed type and stable layered type） in different periods （the temperature raising period and stable layered period）. In addition, several approaches to obtaining the parameters are proposed for water temperature forecasting with the empirical equation（s）.