Sampling Strategies for Estimation of Parameters Related to Ground Water Quality

Multiple theoretical sampling designs are studied to determine whether sampling designs can be identified that will provide for characterization of ground water quality in rural regions of developing nations. Sampling design in this work includes assessing sampling frequency, analytical methods, length of sampling period, and requirements of sampling personnel. The results answer a set of questions regarding whether using innovative sampling designs can allow hydrogeologists to take advantage of a range of characterization technologies, sampling strategies, and available personnel to develop high-value, water-quality data sets. Monte Carlo studies are used to assess different sampling strategies in the estimation of three parameters related to a hypothetical chemical observed in a ground water well: mean concentration (MeanC), maximum concentration (MaxC), and total mass load (TML). Five different scenarios are simulated. These scenarios are then subsampled using multiple simulated sampling instruments, time periods (ranging from 1 to 10 years), and sampling frequencies (ranging from weekly to semiannually to parameter dependent). Results are analyzed via the statistics of the resulting estimates, including mean square error, bias, bias squared, and precision. Results suggest that developing a sampling strategy based on what may be considered lower quality instruments can represent a powerful field research approach for estimating select parameters when applied at high frequency. This result suggests the potential utility of using a combination of lower quality instrument and local populations to obtain high frequency data sets in regions where regular monitoring by technicians is not practical.     

Lattice Boltzmann Models for Flow and Transport in Saturated Karst

Flow and transport simulation in karst aquifers remains a significant challenge for the ground water modeling community. Darcy's law–based models cannot simulate the inertial flows characteristic of many karst aquifers. Eddies in these flows can strongly affect solute transport. The simple two-region conduit/matrix paradigm is inadequate for many purposes because it considers only a capacitance rather than a physical domain. Relatively new lattice Boltzmann methods (LBMs) are capable of solving inertial flows and associated solute transport in geometrically complex domains involving karst conduits and heterogeneous matrix rock. LBMs for flow and transport in heterogeneous porous media, which are needed to make the models applicable to large-scale problems, are still under development. Here we explore aspects of these future LBMs, present simple examples illustrating some of the processes that can be simulated, and compare the results with available analytical solutions. Simulations are contrived to mimic simple capacitance-based two-region models involving conduit (mobile) and matrix (immobile) regions and are compared against the analytical solution. There is a high correlation between LBM simulations and the analytical solution for two different mobile region fractions. In more realistic conduit/matrix simulation, the breakthrough curve showed classic features and the two-region model fit slightly better than the advection-dispersion equation (ADE). An LBM-based anisotropic dispersion solver is applied to simulate breakthrough curves from a heterogeneous porous medium, which fit the ADE solution. Finally, breakthrough from a karst-like system consisting of a conduit with inertial regime flow in a heterogeneous aquifer is compared with the advection-dispersion and two-region analytical solutions.     

Calibration of Parameter Fields Consisting of Multiple Statistical Populations

If a parameter field to be calibrated consists of more than one statistical population, usually not only the parameter values are uncertain, but the spatial distributions of the populations are uncertain as well. In this study, we demonstrate the potential of the multimodal calibration method we proposed recently for the calibration of such fields, as applied to real-world ground water models with several additional stochastic parameter fields. Our method enables the calibration of the spatial distribution of the statistical populations, as well as their spatially correlated parameterization, while honoring the complete prior geostatistical definition of the multimodal parameter field. We illustrate the implications of the method in terms of the reliability of the posterior model by comparing its performance to that of a "conventional" calibration approach in which the positions of the statistical populations are not allowed to change. Information from synthetic calibration runs is used to show how ignoring the uncertainty involved in the positions of the statistical populations not only denies the modeler the opportunity to use the measurement information to improve these positions but also unduly influences the posterior intrapopulation distributions, causes unjustified adjustments to the cocalibrated parameter fields, and results in poorer observation reproduction. The proposed multimodal calibration allows a more complete treatment of the relevant uncertainties, which prevents the abovementioned adverse effects and renders a more trustworthy posterior model.     

A Wet/Wet Differential Pressure Sensor for Measuring Vertical Hydraulic Gradient

Vertical hydraulic gradient is commonly measured in rivers, lakes, and streams for studies of groundwater–surface water interaction. While a number of methods with subtle differences have been applied, these methods can generally be separated into two categories; measuring surface water elevation and pressure in the subsurface separately or making direct measurements of the head difference with a manometer. Making separate head measurements allows for the use of electronic pressure sensors, providing large datasets that are particularly useful when the vertical hydraulic gradient fluctuates over time. On the other hand, using a manometer-based method provides an easier and more rapid measurement with a simpler computation to calculate the vertical hydraulic gradient. In this study, we evaluated a wet/wet differential pressure sensor for use in measuring vertical hydraulic gradient. This approach combines the advantage of high-temporal frequency measurements obtained with instrumented piezometers with the simplicity and reduced potential for human-induced error obtained with a manometer board method. Our results showed that the wet/wet differential pressure sensor provided results comparable to more traditional methods, making it an acceptable method for future use.