Uniqueness, scale, and resolution issues in groundwater model parameter identification

Yeh, Tian Chyi Jim; Mao, D.; Zha, Yuanyuan; Wen, Jet‐Chau; Wan, Li; Hsu, Kuo Chin; Lee, Cheng haw

doi:10.1016/j.wse.2015.08.002

Cited by 57 publications

(57 citation statements)

References 68 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In other words, the entire domain with fracture systems of models B and C likely can be treated as a continuous network (EPM). This may be the reason for the good validation of the EPM parameters for the entire domain under models B and C. As an additional note, the entire model A domain, nevertheless, can be considered as an ensemble REV (Yeh, Mao, et al, ). That is, the P 32 value is representative of the entire domain, and the requirement of the elementary volume is met in the ensemble sense since the entire DFN domain is merely one possibility (element) of the specified P 32 value.…”

Section: Discussionmentioning

confidence: 99%

Equivalence of Discrete Fracture Network and Porous Media Models by Hydraulic Tomography

Dong

Yeh

et al. 2019

Water Resources Research

View full text Add to dashboard Cite

Hydraulic tomography (HT) has emerged as a potentially viable method for mapping fractures in geologic media as demonstrated by recent studies. However, most of the studies adopted equivalent porous media (EPM) models to generate and invert hydraulic interference test data for HT. While these models assign significant different hydraulic properties to fractures and matrix, they may not fully capture the discrete nature of the fractures in the rocks. As a result, HT performance may have been overrated. To explore this issue, this study employed a discrete fracture network (DFN) model to simulate hydraulic interference tests. HT with the EPM model was then applied to estimate the distributions of hydraulic conductivity (K) and specific storage (Ss) of the DFN. Afterward, the estimated fields were used to predict the observed heads from DFN models, not used in the HT analysis (i.e., validation). Additionally, this study defined the spatial representative elementary volume (REV) of the fracture connectivity probability for the entire DFN dominant. The study showed that if this spatial REV exists, the DFN is deemed equivalent to EPM and vice versa. The hydraulic properties estimated by HT with an EPM model can then predict head fields satisfactorily over the entire DFN domain with limited monitoring wells. For a sparse DFN without this spatial REV, a dense observation network is needed. Nevertheless, HT is able to capture the dominant fractures.

show abstract

Section: Discussionmentioning

confidence: 99%

Equivalence of Discrete Fracture Network and Porous Media Models by Hydraulic Tomography

Dong

Yeh

et al. 2019

Water Resources Research

View full text Add to dashboard Cite

show abstract

“…According to Sun et al (2013), the drawdown at a well at the interpolated time at which drawdown is zero in CooperJacob's method is most correlated with S s over a large portion of the aquifer and late time drawdown data at the observation well are highly correlated to K over the entire aquifer. In addition, Yeh et al (2011) and Mao et al (2013) emphasized that the head change over at least two time step is necessary to determine S s term. Similarly, Zhu and Yeh (2005) suggested that 4-5 points in a hydrograph are enough to condition the curve in the inversion because drawdown data are highly correlated in time.…”

Section: Application To the Miu Sitementioning

confidence: 99%

What does hydraulic tomography tell us about fractured geological media? A field study and synthetic experiments

Zha

Illman

Tanaka

et al. 2015

Journal of Hydrology

Self Cite

View full text Add to dashboard Cite

“…It is noteworthy that the virtual observation concept is different from the well‐known pilot point method. For reducing computational cost and making the inverse problem well‐defined [ Yeh et al ., 2015a, 2015b], the pilot point method uses only a few selected pilot points, where hydraulic parameter values are estimated by a nonlinear algorithm minimizing the simulated and observed hydraulic head differences. The entire parameter field is obtained afterward by kriging based on the parameter values at the pilot point locations and the unconditional covariance function of the parameter [ McLaughlin and Townley , ; Soueid Ahmed et al ., ].…”

Section: Methodsmentioning

confidence: 99%

“…Stochastic modeling approaches, on the other hand, adopt the stochastic conceptualization of heterogeneity [ Yeh et al ., 2015a, 2015b]. They consider the hydraulic parameter of a field site as a random field, characterized by a uniform mean and a covariance model.…”

Section: Introductionmentioning

confidence: 99%

“…[], and Yeh et al . []), which can provide high‐resolution maps of the aquifer heterogeneity with a limited number of wells, also suffer from these difficulties. Over the past years, many researchers have attempted to modify the initial starting hydraulic parameter values and the regularization term in the inverse model to incorporate the knowledge of large‐scale geologic structures.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Incorporating geologic information into hydraulic tomography: A general framework based on geostatistical approach

Zha

Yeh

Illman

et al. 2017

Water Resources Research

Self Cite

View full text Add to dashboard Cite

Hydraulic tomography (HT) has become a mature aquifer test technology over the last two decades. It collects nonredundant information of aquifer heterogeneity by sequentially stressing the aquifer at different wells and collecting aquifer responses at other wells during each stress. The collected information is then interpreted by inverse models. Among these models, the geostatistical approaches, built upon the Bayesian framework, first conceptualize hydraulic properties to be estimated as random fields, which are characterized by means and covariance functions. They then use the spatial statistics as prior information with the aquifer response data to estimate the spatial distribution of the hydraulic properties at a site. Since the spatial statistics describe the generic spatial structures of the geologic media at the site rather than site‐specific ones (e.g., known spatial distributions of facies, faults, or paleochannels), the estimates are often not optimal. To improve the estimates, we introduce a general statistical framework, which allows the inclusion of site‐specific spatial patterns of geologic features. Subsequently, we test this approach with synthetic numerical experiments. Results show that this approach, using conditional mean and covariance that reflect site‐specific large‐scale geologic features, indeed improves the HT estimates. Afterward, this approach is applied to HT surveys at a kilometer‐scale‐fractured granite field site with a distinct fault zone. We find that by including fault information from outcrops and boreholes for HT analysis, the estimated hydraulic properties are improved. The improved estimates subsequently lead to better prediction of flow during a different pumping test at the site.

show abstract

Uniqueness, scale, and resolution issues in groundwater model parameter identification

Cited by 57 publications

References 68 publications

Equivalence of Discrete Fracture Network and Porous Media Models by Hydraulic Tomography

Equivalence of Discrete Fracture Network and Porous Media Models by Hydraulic Tomography

What does hydraulic tomography tell us about fractured geological media? A field study and synthetic experiments

Incorporating geologic information into hydraulic tomography: A general framework based on geostatistical approach

Contact Info

Product

Resources

About