数值模型结构不确定性对模拟结果的影响分析

以内蒙古鄂托克旗为例,基于GMS中的MODFLOW模块构建了地下水流数值模拟模型,分析了模型结构(含水层厚度、参数分区)与模型参数不确定性因素对模拟结果的影响.研究结果表明:含水层不确定情景(含水层下边界概化为隔水底板平均值870 m)与实际情况水头差值绝对值的累计和最大为701 m,对模拟结果起了主控作用;当含水层下边界概化为910,940 m时,累计和分别增加为1 013,1 593 m;与仅考虑单个不确定性因素相比,同时考虑模型参数与含水层不确定情景累计和最大为738 m,同时考虑参数分区与含水层不确定情景累计和最大为791 m.因此,在构建地下水流数值模拟模型时,应优先考虑含水层空间结构概化的合理程度,同时考虑多个不确定性因素对模拟结果的综合影响,使地下水数值模拟模型能更精确地反映真实的地下水流状况.

Analysis of groundwater numerical model structural uncertainty on simulation results

Based on the MODFLOW module in GMS,a numerical simulation model of underground water flow was developed,which focuses on the structural uncertainty of groundwater numerical model and takes Etuoke Banner,Inner Mongolia as an example.The influence of model structure(aquifer thickness,parameter partition)and uncertainty factors of model parameters on the simulation results was analyzed.The results show that the accumulative and maximum value of the absolute difference between the uncertain aquifer scenario(the lower boundary of the aquifer is set to the average of 870 m of water-proof floor)and the actual situation is 701 m,which plays a main role in controlling the simulation results.When the lower boundary of the aquifer is set to 910 m and 940 m,the accumulative sum increases to 1013 m and 1593 m respectively.Compared with only considering a single uncertainty factor,the accumulative sum of model parameters and aquifer uncertainty scenarios is 738 m at the same time,and the accumulative sum of parameter zoning and aquifer uncertainty scenarios is 791 m at the same time.Therefore,when constructing the numerical simulation model of groundwater flow,the reasonable degree of spatial structure generalization of aquifer should be given priority,and the comprehensive influence of multiple uncertain factors on the simulation results should be considered,so that the numerical simulation model of groundwater flow can reflect the real groundwater flow situation more accurately.