Hydrology [H]

H41H  MW:2020   Thursday
Hydrogeophysics: Linking Geophysical and Hydrological Data II
Presiding: G Cassiani, University of Padova; K Holliger, University of Lausanne

H41H-01 INVITED 

Representativity and Challenges in the Simulation of Subsurface Flow at Hillslope and Subcatchment Scales: Case Studies from Quebec and Nova Scotia (Canada)

* Paniconi, C (claudio.paniconi@ete.inrs.ca), INRS-ETE, Université du Québec, 490 Rue de la Couronne, Quebec City, G1K 9A9, Canada Broda, S), Université du Québec à Montréal, 201 Avenue Kennedy, Montreal, H3C 3P8, Canada Gauthier, M), INRS-ETE, Université du Québec, 490 Rue de la Couronne, Quebec City, G1K 9A9, Canada Guay, C), INRS-ETE, Université du Québec, 490 Rue de la Couronne, Quebec City, G1K 9A9, Canada Larocque, M), Université du Québec à Montréal, 201 Avenue Kennedy, Montreal, H3C 3P8, Canada Nastev, M), Geological Survey of Canada, Natural Resources Canada, 490 Rue de la Couronne, Quebec City, G1K 9A9, Canada Rivard, C), Geological Survey of Canada, Natural Resources Canada, 490 Rue de la Couronne, Quebec City, G1K 9A9, Canada

Soil on sloping bedrock is a common conceptualization in hydrological modeling at the hillslope and subcatchment scales. At these small scales morphology, vegetation, and climate can all be quite accurately represented, whereas pedology and geology are much more difficult to characterize. Thus further approximations, reasoned or ad hoc, are often introduced concerning the prevailing direction(s) of flow, the (im)permeability of the soil-bedrock interface, the presence or absence of layering and other inhomogeneities in the soil and aquifer, the influence of the unsaturated zone, the degree of interaction between the land surface and the subsurface, and so on. We will examine the impact of some of these approximations using data and simulations of sloping aquifers from two study sites, the Thomas Brook subcatchment in the Annapolis Valley, Nova Scotia and a hillslope in Havelock, part of the Chateauguay river basin in southwestern Quebec. Two models, one based on a discretization of the full (3D) Richards equation coupled to a path-based surface routing model, and the other based on the 1D Boussinesq equation for groundwater flow in unconfined aquifers, will be used to assess groundwater recharge, saturation behavior at the land surface, soil water dynamics, and other variables in response to different representations of the soil and geology. Reliable estimation of two variables in particular, leakage/recharge and soil moisture/soil water storage, is critical to integrated water resources management, and combined use of data and models is crucial to achieving this task. These examples underline the point that accurate, inexpensive, and rapid noninvasive hydrogeophysical characterization of the subsurface on a hillslope or small catchment can contribute to important advances in hydrological simulation at these scales.

H41H-02 

Hydrogeophysical inversion of time-lapse proximal ground-penetrating radar data for shallow subsurface hydraulic characterization

* Lambot, S (sebastien.lambot@uclouvain.be), Universite catholique de Louvain, Croix du Sud, 2 box 2, Louvain-la-Neuve, 1348, Belgium Jadoon, K Z (k.z.jadoon@fz-juelich.de), Forschungszentrum Juelich GmbH, Forschungszentrum Juelich GmbH, Juelich, D-52425, Germany Slob, E C (E.C.Slob@tudelft.nl), Delft University of Technology, Stevinweg 1, Delft, 2628 CN, Netherlands Vereecken, H (h.vereecken@fz-juelich.de), Forschungszentrum Juelich GmbH, Forschungszentrum Juelich GmbH, Juelich, D-52425, Germany

Sustainable and optimal agricultural and environmental management of water and land resources particularly relies on the description and understanding of soil water distribution and dynamics at the field scale. We present a non-invasive method for estimating the unsaturated soil hydraulic properties and continuous vertical water content profiles from time-lapse proximal GPR based on integrated electromagnetic and hydrodynamic inversion. The radar system is based on international standard vector network analyzer technology and a full-wave model is used to describe wave propagation in the antenna-air-soil system, including antenna-soil interactions. Hydrodynamic modeling is based on a one-dimensional solution of Richard's equation with Mualem-van Genuchten parameterization. The uniqueness of the inverse solution is analyzed using numerical experiments for different textured soils and infiltration events. The stability of the inverse solution with respect to errors in fixed key hydraulic and petrophysical parameters is quantified. Finally, the approach is tested in controlled laboratory conditions for an variable infiltration event in a homogeneous sandy soil. The proposed approach appears to be promising for characterizing the shallow subsurface hydraulic properties at the field scale with a high spatial resolution.

H41H-03 

Impact of Integrated Versus Sequential Data Fusion on Hydrologic Predictions

* Moysey, S (smoysey@clemson.edu), Clemson University, Environmental Engineering and Earth Sciences, 340 Brackett Hall, Clemson, SC 29634, United States Fowler, D E (dylanf@clemson.edu), Clemson University, Environmental Engineering and Earth Sciences, 340 Brackett Hall, Clemson, SC 29634, United States Sicilia, T (tom@coffepot.org), Clemson University, Environmental Engineering and Earth Sciences, 340 Brackett Hall, Clemson, SC 29634, United States

Fusing data from a wide variety of sources into a hydrologic model is one of the most promising approaches we have for improving hydrologic predictions. How to merge disparate types of data into a consistent hydrologic model is, however, a critical question that has not been fully resolved. One issue in particular is how different strategies for data integration may ultimately impact hydrologic predictions. We have been investigating integrated versus sequential data fusion methods as alternate strategies for integrating geophysical data into hydrologic estimation problems. The more traditional approach, sequential data fusion, relies on geophysical inversion and rock physics to produce a map of the hydrologic properties of the subsurface that can subsequently be used to constrain a hydrologic inverse problem. In contrast, integrated data fusion uses coupled hydrologic and geophysical models to directly constrain hydrologic parameters of interest with geophysical measurements. Sequential data fusion is easier to implement since it leverages existing methods and codes that often, in practice, allows geophysicists and hydrologists to work independently. Limited spatial resolution, reliance on prior knowledge of the subsurface, and the questionable existence of rock physics relationships between geophysical and hydrologic parameters at the field scale are all critical limitations of the approach. These limitations lead to bias in the hydrologic inverse problem resulting in poor hydrologic models and unreliable predictions. Integrated data fusion overcomes some of these issues by focusing on estimating parameters that control processes, rather than imaging state variables. For example, in a homogeneous soil an infiltration plume can be fully characterized by just a few variables (assuming boundary conditions are known) though the distribution of water content is highly variable through space and time and would require thousands of parameters to characterize directly. Because the parameters controlling processes are better constrained in integrated data fusion, more reliable predictions can ultimately be obtained. As a result, integrated data fusion is the most appropriate choice for data integration if the objective is to produce models capable of hydrologic predictions. We demonstrate how the choice of data fusion method impacts predictions using examples of solute transport monitored by electrical resistivity and infiltration monitored by borehole ground-penetrating radar.

H41H-04 

Quantitative Tracking of a Chloride Tracer Front Progression with 3D ERT and TDR

* Kostel, J (j.koestel@fz-juelich.de), ICG-4 (Agrosphere), Forschungszentrum Juelich GmbH, Juelich, 52425, Germany Kemna, A (a.kemna@fz-juelich.de), ICG-4 (Agrosphere), Forschungszentrum Juelich GmbH, Juelich, 52425, Germany Javaux, M (mathieu.javaux@uclouvain.be), ICG-4 (Agrosphere), Forschungszentrum Juelich GmbH, Juelich, 52425, Germany Javaux, M (mathieu.javaux@uclouvain.be), Dep. of Environmental Sciences, Universite catholique de Louvain, Louvain-la-Neuve, 1348, Belgium Binley, A (a.binley@lancaster.ac.uk), Department of Environmental Science, Lancaster University, Lancaster, LA1 4YW, United Kingdom Vereecken, H (h.vereecken@fz-juelich.de), ICG-4 (Agrosphere), Forschungszentrum Juelich GmbH, Juelich, 52425, Germany

Electrical resistivity tomography (ERT) has proved to be an valuable tool to image solute transport processes through the subsurface. However, a quantitative interpretation is aggravated by the non-uniqueness of the ERT data. In the vadose zone, additional problems arise in terms of distinguishing between contributions of water content and solute concentration to the bulk electrical conductivity. In this study, we developed an experimental setup to quantitatively monitor a chloride tracer front through a large undisturbed unsaturated soil column by means of 3D ERT and time domain reflectometry (TDR) probes. A smoothness constraint inversion algorithm was used for the ERT inversion. A comparison of bulk electrical conductivity obtained from ERT and TDR measurements showed that deviations between both methods were largely dependent of the degree of smoothness imposed on the ERT image. The deviations reached a minimum when the smoothness in ERT image was similar to values which are empirically known to yield qualitatively reasonable results. Additionally, by means of the TDR data, we were able to estimate confidence intervals for ERT derived apparent transport model parameters. We therefore conclude that TDR measurements can be used as a ground truth data for ERT measurements. As an additional validation of our approach, we compared the apparent transport model parameters for the effluent chloride flux concentrations to corresponding parameters for the flux-averaged resident concentrations derived from ERT. Here, the match between the two values was similar to and consistent with the ones found at the TDR probe locations. Furthermore, the bulk electrical conductivity changes were related to chloride concentration by calibrating a petrophysical model with spatial resolution. The validity of the petrophysical parameters was appraised by comparison with TDR data and data gathered from soil samples. A good agreement was found. Consequently, the ERT derived chloride concentrations exhibited no mass balance problems due to change of apparent petrophysical parameters with change of ERT sensitivity.

H41H-05 

Estimation of the Unsaturated Hydraulic Soil Properties From Joint Inversion of Tension Infiltrometer and ERT Measurements: Numerical Experiments

* Schneider, S (sebastien.schneider@u-psud.fr), University of Paris XI, UMR-CNRS 8148 IDES, Bât. 504, Orsay, 91405, France Vanderborght, J (j.vanderborght@fz-juelich.de), Agrosphere Institute, ICG-IV, Forschungszentrum J\"ülich GmbH, J\"ülich, 52425, Germany Kemna, A (a.kemna@fz-juelich.de), Agrosphere Institute, ICG-IV, Forschungszentrum J\"ülich GmbH, J\"ülich, 52425, Germany Pessel, M (marc.pessel@u-psud.fr), University of Paris XI, UMR-CNRS 8148 IDES, Bât. 504, Orsay, 91405, France Coquet, Y (Yves.Coquet@agroparistech.fr), UMR INRA/INAPG Environment and Arable Crops, Paris-Grignon, Thiverval-Grignon, 78850, France

An accurate and time-efficient estimation of unsaturated hydraulic soil properties in the field remains a challenge. Tension-infiltrometry is often used to determine unsaturated hydraulic soil properties and their spatial variability in the field. Due to capillary flow, a 3-D wetting bulb, which depends on the unsaturated hydraulic soil properties, the radius of the infiltrometer disk, and the applied water tension, develops under a tension infiltrometer. Electrical resistivity tomography (ERT) offers the possibility to image the spatial distribution of bulk soil electrical conductivity, which is related through a petrophysical model to the soil water content. Therefore, ERT data contain information about the 3-D structure of the wetting bulb, which may be exploited to infer hydraulic soil properties. A combination of tension-infiltrometer and ERT data for an inverse estimation of the soil hydraulic parameters was tested in a numerical experiment. Both 3-D water flow and electrical potential fields were simulated with the SWMS_3-D model using the van Genuchten hydraulic functions and the Rhoades petrophysical model. Simulated infiltration and simulated apparent electrical resistivities were subsequently inverted using the PEST software. Inversion of the combined infiltration and ERT datasets showed that the hydraulic parameters could be inverted from a single infiltration experiment, which is not possible when only infiltration data are used in the inversion. Also petrophysical parameters could be inverted simultaneously with hydraulic parameters from the combined ERT-infiltrometer data. These results demonstrate the potential of the method by considering additional information about the structure of the wetting bulb which is contained in ERT data.

H41H-06 

Quasi 3D model construction of artificial recharge through the vadose zone using time-lapse zero-offset profiling of cross borehole radar

* Kuroda, S (skuroda@affrc.go.jp), National Institute for Rural Engineering, Kannondai 2-1-6, Tsukuba, 305-8609, Japan Saito, H (hiros@cc.tuat.ac.jp), Tokyo University of Agriculture and Technology, Fuchu, Tokyo, 183-8509, Fuchu, 183-8509, Japan Okuyama, T (okuyama@affrc.go.jp), National Institute for Rural Engineering, Kannondai 2-1-6, Tsukuba, 305-8609, Japan Takeuchi, M (takeutch@affrc.go.jp

Simunek, J (Jiri.Simunek@ucr.edu), University of California, Riverside, 900 University Ave, Riverside, CA 92521, United States Van Genuchten, M T (vanGenuchten@ucr.edu), U.S. Salinity Laboratory, USDA, ARS, Riverside, CA, 900 University Ave, Riverside, CA 92521, United States

Hydrologic processes in the vadose zone were studied by using a time-laps geophysical methodology. Main focus was on using the acquired geophysical data to inversely estimate the soil water retention and unsaturated hydraulic conductivity functions. The soil hydraulic properties were estimated from cross borehole radar data obtained during a ground water recharge test in a infiltration pit. We used the HYDRUS-2D software package for the inverse parameter estimation problem. The estimated soil hydraulic parameters were subsequently employed in a forward simulation to construct a quasi three-dimensional image of the variably-saturated flow process in the vadose zone.

H41H-07 

Temporal Moments in Hydrogeophysics

Pollock, D (davina.pollock@eawag.ch), Swiss Federal Institute of Aquatic Science and Technology (Eawag), Überlandstr. 133, Dübendorf, 8600, Switzerland * Cirpka, O A (olaf.cirpka@eawag.ch), Swiss Federal Institute of Aquatic Science and Technology (Eawag), Überlandstr. 133, Dübendorf, 8600, Switzerland

Electrical Resistivity Tomography (ERT) has been tested as monitoring tool for salt-tracer experiments by various authors. So far, the analysis of such experiments has been done by a two-step procedure [Kemna et al., 2002; Vanderborght et al., 2005; Singha and Gorelick, 2005]. In the first step, classical geophysical inversion methods have been used to infer the distribution of electrical conductivity, which is transferred to an estimated concentration distribution of the tracer. Subsequently, the inferred concentration images were analyzed to estimate hydraulic quantities such as the velocity distribution. This approach has two disadvantages:

  1. The concentration distribution is reconstructed with a high spatial resolution, but the estimate is uncertain, and the estimation uncertainty is spatially correlated. These correlated uncertainties should be accounted for in the estimation of hydraulic conductivity from concentration values. The latter, unfortunately, is not practical because the reconstructed data sets are very large.
  2. The geophysical inversion is not enforced to be in agreement with basic hydromechanical constraints. E.g., Singha and Gorelick [2005] observed an apparent loss of solute mass when using ERT as monitoring tool.
We propose considering the temporal moments of potential-difference time series. These temporal moments depend on temporal moments of concentration, which have already been used in the inference of hydraulic- conductivity distributions (Cirpka and Kitanidis, 2000). In our contribution, we present the complete set of equations leading from hydraulic conductivity via hydraulic heads, velocities, temporal moments of concentrations to temporal moments of potential differences for given flow and transport boundary conditions and electrode configurations. We also present how the sensitivity of temporal moments of potential differences on the hydraulic conductivity field can be computed without the need of storing intermediate sensitivities. With these methods, we have developed the kernel of a fully coupled hydrogeophysical inversion approach, in which the direct target of inversion is the hydraulic-conductivity field. By construction, the estimated concentration distribution is prevented to show unphysical behavior.

H41H-08 

Integration of Electric Resistivity Profile and Infiltrometer Measurements to Calibrate a Numerical Model of Vertical Flow in Fractured and Karstic Limestone.

* Caputo, M C (maria.caputo@ba.irsa.cnr.it), National Research Council (CNR), Water Research Institute (IRSA), via F. De Blasio, 5, Bari, 70123, Italy De Carlo, L (lorenzo.decarlo@ba.irsa.cnr.it), National Research Council (CNR), Water Research Institute (IRSA), via F. De Blasio, 5, Bari, 70123, Italy Masciopinto, C (costantino.masciopinto@ba.irsa.cnr.it), National Research Council (CNR), Water Research Institute (IRSA), via F. De Blasio, 5, Bari, 70123, Italy Nimmo, J R (jrnimmo@usgs.gov), US Geological Survey, 345 Middlefield Road, MS 421, Menlo Park, CA 94025, United States

Karstic and fractured aquifers are among the most important drinking water resources. At the same time, they are particularly vulnerable to contamination. A detailed scientific knowledge of the behavior of these aquifers is essential for the development of sustainable groundwater management concepts. Due to their special characteristics of extreme anisotropy and heterogeneity, research aimed at a better understanding of flow, solute transport, and biological processes in these hydrogeologic systems is an important scientific challenge. This study integrates a geophysical technique with an infiltrometer test to better calibrate a mathematical model that quantifies the vertical flow in karstic and fractured limestone overlying the deep aquifer of Alta Murgia (Southern Italy). Knowledge of the rate of unsaturated zone percolation is needed to investigate the vertical migration of pollutants and the vulnerability of the aquifer. Sludge waste deposits in the study area have caused soil-subsoil contamination with toxics. The experimental test consisted of infiltrometer flow measurements, more commonly utilized for unconsolidated granular porous media, during which subsoil electric resistivity data were collected. A ring infiltrometer 2 m in diameter and 0.3 m high was sealed to the ground with gypsum. This large diameter yielded infiltration data representative of the anisotropic and heterogeneous rock, which could not be sampled adequately with a small ring. The subsurface resistivity was measured using a Wenner-Schlumberger electrode array. Vertical movement of water in a fracture plane under unsaturated conditions has been investigated by means of a numerical model. The finite difference method was used to solve the flow equations. An internal iteration method was used at every time step to evaluate the nodal value of the pressure head, in agreement with the mass- balance equation and the characteristic functional relationships of the coefficients.