Search

Characterizing fractured aquifers plays a crucial role in the issues related to groundwater contamination, and geothermal and hydrocarbon resource exploitation. Due to the heterogeneity of the fractured medium, the permeability of fractured medium generally exhibits significant non-Gaussian characteristics, leading to difficulties and challenges in the estimation of hydrogeological parameters. This study used the ESMDA-DS (ensemble smoother with multiple data assimilation-direct sampling) integrating hydrogeophysical data to explore the effectiveness of the data assimilation framework in portraying the parameter field of the fractured medium and to analyze the influences of assimilating three different types of observation data, the fracture density, and the number of observation wells on the parameter estimation. The results show that the method of ESMDA-DS integrating hydrogeophysical data can estimate the spatial distribution of hydrogeological parameters in the fractured medium effectively. Comparing the estimated results from three types of observation, it finds that fusing the hydraulic head and the self-potential observational data (hydrogeophysical data) has the best effect. The fracture density in the study area and the number of observation wells also affect the data assimilation results. A reasonable number of observation wells is suggested to obtain the optimal parameter estimation scheme in practical applications. This study can provide an effective method for characterizing the parameter field of the fractured medium and a reliable theoretical basis for the development and management of fractured water resources.

The ensemble Kalman filter (EnKF) is one of the most widely used data assimilation methods. However, it exhibits limitations in handling non-Gaussian problems. To effectively address such issues and accurately describe the connectivity of aquifers, a novel approach named NS-ES-MDA is developed in this study. The proposed NS-ES-MDA synergistically combines the normal-score transformation (NST) with ensemble smoother with multiple data assimilation (ES-MDA). Through comparative experiments, the efficacy of NS-ES-MDA in estimating the hydraulic conductivity of non-Gaussian distributed aquifers is demonstrated. By assimilating the same dataset, NS-ES-MDA exhibits approximately 34% improvement in parameter estimation accuracy and about 35% enhancement in computational efficiency compared to the restart normal-score ensemble Kalman filter (rNS-EnKF). Furthermore, the NS-ES-MDA shows case robustness against the “equifinality” and displays remarkable updating capabilities, which leads to more precise parameter estimates. This study provides an effective solution for parameter estimation in non-Gaussian distributed aquifers.

In order to explore the effectiveness of salt and heat tracers in a fracture-matrix system, a fracture-matrix test model is designed. The tracer tests with different tracers are carried out. The transport process of tracer is described in the fracture-matrix with the real-time dynamic resistivity monitoring data at different measuring points, and the effectiveness of salt and heat tracers based on the resistivity method is discussed. The results show that the process of injecting three different tracers into the fracture-matrix system and the existence of fracture can be seen by the resistivity method. Under the salt-heat tracer, the difference between the volume conductivity in the fracture and the matrix is the largest, and the change rate of the volume conductivity in the fracture and the matrix decreases with the increasing depth. The fitting effect of concentration and volume conductivity is better than that of temperature and volume conductivity. These indicate that the tracer based on the resistivity method is effective in describing the positions of the fracture and matrix in the fracture-matrix system, and the salt-heat combined tracer has the best effect. The results are of certain reference for the field electrical exploration of the positions of fractures and other heterogeneous layers.

The Ensemble Kalman Filter (EnKF) has been widely applied for real-time simulation of groundwater flow and solute transport. The majority of previous studies tend to assume no bias in forecast models, therefore ignoring the model uncertainties. This assumption, however, may be invalid when a conceptual model is not accurately generalized. As a result, forecast bias will lead to incorrect estimation of the system parameters or states. In this work, a bias aware Ensemble Kalman Filter with Confirming Option (Bias-CEnKF) is proposed to take into account the forecast bias by the model during the filtering process. The proposed method is tested in a real-time groundwater simulation considering model uncertainties, by setting inaccurate boundary conditions, initial conditions and recharge items. The results show that the standard EnKF may lead to filter divergence and assimilation failure, if the forecast model is not accurately generalized. Instead, Bias-CEnKF not only achieves better performances, but also reduces the inconsistency caused by the nonlinear relationship among the parameters, variables and bias corrections. Four scenarios are investigated, with the results showing the aquifer hydraulic conductivities and heads obtained by Bias-CEnKF are close to the real values, and the prediction results are also more reliable. This study further improves the applicability of the EnKF under the uncertain condition of model generalization, and provides a reliable method for groundwater data assimilation under complex field conditions.

The migration and distribution of dense non-aqueous phase liquid (DNAPLs) in fractured media are affected by many factors, including the chemical properties of DNAPLs, geometric and chemical properties of fractures and leakage conditions. Previous studies were mostly conducted in a single fracture, while the migration of DNAPLs in stochastic network fracture were seldom examined. In this paper, stochastic network fractures are generated using the Monte Carlo method. The coordinates and apertures of fractures are identified and output by pixel scanning. The migration of Perchloroethylene (PCE) in the stochastic network fractures is simulated by PetraSim. The effects of the geometric properties of the fractures and the leakage conditions (including the leakage rate and the leakage location) on the migration of DNAPLs are discussed. The numerical simulation results show that the spatial variability of fractures also affects the migration path and spatial distribution of DNAPLs. With the increase of spatial variability of fracture width in the network fractures, dominant channels appear, the migration rate of DNAPLs accelerates, and the location of DNAPLs centroid and the spatial distribution of saturation change significantly. The leakage rate affects the migration range and spatial distribution of DNAPLs. The higher the leakage rate is, the faster the migration rate of DNAPLs will be. The more saturated the accumulated DNAPLs at the bottom of the model, the greater the spatial distribution of DNAPLs is. In the same network fracture, different leakage locations will also lead to different migration paths and distribution ranges of DNAPLs. The spatial variability of fractures in the gravity direction at different leakage locations is different, leading to different migration paths and spatial distribution of DNAPLs.

Mastering hydraulic conductivity of rock mass is an important way to precisely describe hydrogeological characteristics of a certain region. Hydraulic conductivity is a significant indicator to reflect the rock mass’ permeability. The studies of hydraulic conductivity estimation models have important implications for the development of actual engineering. In the existing estimation models of hydraulic conductivity, the single-factor model cannot take into consideration of the comprehensive influence of various factors on hydraulic conductivity in the area, and the parameters selection of the multi-factor model lacks the flexibility and its application is limited when some parameters are difficult to be obtained, etc. The classification and comparative analyses of the positive and negative correlation parameters are conducted based on public data. We propose a set of high-fitting hydraulic conductivity estimation models, which are the PNC (Positive and negative correlation) model. The research results show that the fitting result of the PNC model (R2=0.964 and R2=0.801) is superior to that of the HC model (R2=0.905 and R2=0.563) in No. 1 study area. In No. 2 study area, the fitting result of the PNC model (R2=0.959) is superior to that of the RMP model (R2=0.927). In No. 3 study area, the fitting result of the PNC model (R2=0.94 to 0.99) is also better than that of the ZRF model (R2=0.92 to 0.99). By using Nash-Sutcliffe coefficient (NSE) to carry out the error analyses of the model, it is found that the error coefficients of 5 in 7 sets of data are above 0.95. It further illustrates the convenience and reliability of the PNC model, which can provide a certain reference for estimating and verifying hydraulic conductivity in actual engineering.