基于EFDC的滇池藻类模型参数敏感性时空特征  

Spatial-temporal parameter sensitivities of the Dianchi algae model based on EFDC

作　　者：李国辉 李莉杰 李根保[1] LI Guohui;LI Lijie;LI Genbao(State Key Laboratory of Freshwater Ecology and Biotechnology,Institute of Hydrobiology,Chinese Academy of Sciences,Wuhan 430072,China;University of Chinese Academy of Sciences,Beijing 100049,China)

机构地区：[1]中国科学院水生生物研究所淡水生态与生物技术国家重点实验室,武汉430072 [2]中国科学院大学,北京100049

出　　处：《应用与环境生物学报》2021年第4期1047-1054,共8页Chinese Journal of Applied and Environmental Biology

基　　金：水专项滇池项目(2013ZX07102-005)资助。

摘　　要：分析大型浅水湖泊藻类模型参数敏感性时空特征,可以提高模型率定效率和模拟精度,还可通过模型参数时空敏感性认识藻类生理特点的时空变化.基于环境流体动力学模型(Environmental Fluid Dynamic Code,EFDC),构建滇池藻类模型,选取13个藻类模型参数对计算网格进行参数敏感性分析,并基于K-means聚类法对网格聚类,分析参数在4个时期的空间敏感性特征及其时间差异性.结果表明:(1)1月份参数敏感性在空间上被聚为2类,呈现南北分布,其差异主要由藻类最大生长速率(PMc)及低温限制参数(KTG1c)敏感性不同导致;3月份参数敏感性的空间差异消失,叶绿素消光系数(KeChl)最敏感,其次为藻类最大生长速率参数(PMc);7月份由于磷营养盐吸收(KHPc)和水动力条件相关参数(WSc)敏感性不同,参数敏感性在空间上被聚为3类,呈北部、湖心、湖岸分布;10月份水华转向衰退,温度相关参数(TMc1、KTG1c)敏感性提高,水温的限制作用最早在湖心区显现.(2)参数敏感性的时间差异大于空间差异,其时间差异主要由温度变化引起,而空间差异主要由湖体水动力条件和磷营养盐浓度的空间分布不同引起,其中水动力条件为主导因素.以上结果表明,滇池藻类模型参数敏感性在水温、水动力和营养盐的作用下呈现时空异质性,在进行敏感性分析时,要注意这些因子的变化对结果的影响.Analysis of the spatial-temporal sensitivities of key parameters in algal models of large shallow lakes can not only improve calibration efficiency and simulation accuracy,but also support a better understanding of the spatial-temporal divergence of algae physiological characteristics.The Dianchi algae model was established using the Environmental Fluid Dynamic Code(EFDC).Thirteen parameters were selected based on algal model equations for sensitivity analysis of each grid cell,and K-means clustering was performed to analyze the spatialtemporal features of parameter sensitivities in four different periods.The results showed that(1)in January,the parameter spatial sensitivities were divided into two regions that appeared in a north-south distribution pattern,which was mainly caused by the effect of algal growth rate(PMc)and temperature limitation(KTG1 c).In March,owing to the increase in wind speed,the spatial sensitivities of the parameters disappeared.The extinction coefficient of chlorophyll(KeChl)was the most sensitive parameter,followed by the algae maximum growth rate parameter(PMc).Growth was the dominant biological process of algae during this period.In July,the simulation area was separated into three regions(north,middle,and shore)of the lake along the parameter sensitivities related to phosphorus nutrient uptake and hydrodynamic conditions.The main metabolic process of algae throughout the lake was respiration.In October,the bloom declined,and the sensitivity of temperaturerelated parameters(TMc1 and KTG1 c)increased.The effect of temperature limitation appeared at the center of the lake.(2)The temporal difference in the parameter sensitivities of the algal model was greater than that of the spatial difference.The former was mainly caused by the change in temperature,whereas the latter was mainly caused by the hydrodynamic condition and spatial distribution of phosphorus nutrients,in which the hydrodynamic condition was the dominant factor.These results indicate that the sensitivity of the algal model p

关 键 词：时空敏感性 K-MEANS聚类 藻类模型 滇池 

分 类 号：X524[环境科学与工程—环境工程]