RUI: Predicting Transport Through Heterogeneous Facies Assemblages: Geostatistical Anatomy of Buried-Valley Aquifers III

这些研究验证了由相似沉积环境形成的颗粒含水层在物理非均质性上存在相似性的假设。含水层的物理非均质性可以用三维地质统计模型来表示，该模型既表示其结构方面，也表示其随机方面。由于环境站点的地下环境不可能得到详尽的描述，因此必须在不确定性和风险限定模型的情况下做出决策。有用的含水层非均质性统计模型只能来自于许多数据丰富的地点，在各种沉积环境中得到充分研究的例子。我们的长期研究目标是为特定沉积环境下颗粒含水层系统非均质性的最相关特征开发统计模型，并评估模拟宏观分散输送所需的详细程度。我们开始研究北美大陆中部冰川时期的隐谷含水层。这些非均质性跨越多个尺度，从相内渗透率的单峰分布，到建筑元素内渗透率的弱多峰分布，再到相组合尺度上的强多峰非均质性。该项目的目标是建立数据库并计算相比例、平移比例方差、平均长度、长度方差、嵌入系数和转移概率。此外，我们还计算了相内渗透率的几何平均值、方差和相关范围。当与沉积环境特征相联系时，我们将这些统计数据统称为定量相模型。我们将比较在中大陆广泛地区发生的四种不同的隐谷含水层开发的定量相模型。流动和运输模型将用于评估所观察到的差异是否与环境决策相关。这是一个RUI提案，与之前的资助阶段一样，该项目将有大量本科生参与。

9903067RitziThese research tests the hypothesis that similarities exist in the physical heterogeneity among granular aquifers formed by similar depositional environments. The physical heterogeneity of aquifers can be represented within a three-dimensional geostatistical model that represents both its structured and random aspects. Because the subsurface at environmental sites is never exhaustively characterized, decisions must be made under uncertainty and risk-qualified models are needed. Useful statistical models of aquifer heterogeneity will only come from many well-studied examples at data-rich sites, in a variety of depositional environments. Our long-term research goal is to develop statistical models for the most relevant features of the heterogeneity in granular aquifer systems in specific depositional setting and to evaluate the level of detail necessary in simulating macrodispersive transport. We have begun by looking at glacigenic buried-valley aquifers in the North American mid-continent. These have heterogeneity ranging across several scales, from unimodal distributions of permeability within facies, to weakly multimodal distributions of permeability within architectural elements, to strongly multimodal heterogeneity at the scale of facies assemblages. The objective of the project are to build data bases and compute facies proportions, variance in proportion with translation, mean length, variance in length, coefficient of embeddedness and transition probability. Furthermore, we compute the geometric mean, variance, and correlation range of permeability within facies. We refer to these statistics collectively, when related to features of the depositional environment, as a quantitative facies model. We will compare quantitative facies models developed for four different buried-valley aquifers occurring across a broad region of the mid-continent. Flow and transport modeling will be used to assess whether or not observed differences are relevant in environmental decision making. This is an RUI proposal, and as in prior phases of funding, the project will have significant undergraduate involvement.