Search

This review focuses on investigations of groundwater flow and solute transport in karst aquifers through laboratory scale models (LSMs). In particular, LSMs have been used to generate new data under different hydraulic and contaminant transport conditions, testing of new approaches for site characterization, and providing new insights into flow and transport processes through complex karst aquifers. Due to the increasing need for LSMs to investigate a wide range of issues, associated with flow and solute migration karst aquifers this review attempts to classify, and introduce a framework for constructing a karst aquifer physical model that is more representative of field conditions. The LSMs are categorized into four groups: sand box, rock block, pipe/fracture network, and pipe-matrix coupling. These groups are compared and their advantages and disadvantages highlighted. The capabilities of such models have been extensively improved by new developments in experimental methods and measurement devices. Newer technologies such as 3D printing, computed tomography scanning, X-rays, nuclear magnetic resonance, novel geophysical techniques, and use of nanomaterials allow for greater flexibilities in conducting experiments. In order for LSMs to be representative of karst aquifers, a few requirements are introduced: (1) the ability to simulate heterogeneous distributions of karst hydraulic parameters, (2) establish Darcian and non-Darcian flow regimes and exchange between the matrix and conduits, (3) placement of adequate sampling points and intervals, and (4) achieving some degree of geometric, kinematic, and dynamic similitude to represent field conditions.

Karst aquifers are geologic media that consist of enlarged solution conduits. The geometry of such conduits is very difficult to delineate and its hydraulic parameters such as hydraulic conductivity (K) and specific storage (Ss) are difficult to estimate. Hydraulic Tomography (HT) shown to successfully map the hydraulic heterogeneity of porous and fractured rock aquifers may also be effective in karst aquifers. In this study, we utilize two synthetic karst aquifers with the Branchwork and Network Conduit Patterns (BCP and NCP) to explore the ability of HT to map conduit patterns and their hydraulic parameters. Prior to the HT analysis, forward models are built to study groundwater flow through synthetic karst aquifers to generate data and to gain insights into flow patterns. Forward modeling results reveal that: (1) flow rates at the outlet of the BCP model is higher than the NCP model even when the same water injection rates are applied at the wells; (2) the location of the injection well has a significant impact on the flow rate at the outlet; and (3) flow lines mainly follow the conduit pattern. These results suggest that the accurate mapping of karst networks is critical. Next, the HT analysis of these data is conducted using a set of six models (BCP and NCP) to map the heterogeneity of different conduit systems under steady state and transient flow conditions. Inverse modeling results reveal that the estimated K maps become more reliable as the number of injection tests and the corresponding monitoring data are increased in the HT analysis. Moreover, the results of the BCP model is closer to the original conduit pattern in comparison to the NCP model. A comparative study of three inverse models with different number of observation wells and assuming the availability of geological data has shown that a more reliable pattern of the conduits can be mapped by adding geological information even when a small number of observation wells are available. Overall, our results confirm the feasibility of HT in detecting major conduits as well as being able to differentiate between the BCP and NCP network types.

The final publication is available at Elsevier via https://dx.doi.org/10.1016/j.jhydrol.2018.10.050 © 2018. This manuscript version is made available under the CC-BY-NC-ND 4.0 license https://creativecommons.org/licenses/by-nc-nd/4.0/

Pumping test is a well-known method for determining hydraulic characteristics of an aquifer. Diagnostic plot technique is applied to facilitate the identification of an appropriate groundwater flow conceptual model, and for interpretation of the pumping test data. The diagnostic plots of six observation wells around a pumping well at the Salman Farsi Dam Site showed simultaneous presence of integer and non-integer flow dimensions of pumping test around the pumping well. Type of flow pattern on two observation wells shows two-dimensional flow and that on four observation wells follows one-and-a-half-dimensional flow. The results revealed that flow pattern around the pumping well in the karst study area changes from two-dimensional within a distance of less than about 100 m from the pumping well to one-and-a-half dimensional at a distance from the pumping well larger than about 100 m. The change of flow dimensions with increasing distance from the pumping well is due to radial flow around the observation wells within a distance of less than about 100 m from the pumping well, and bilinear flow around the observation wells with a distance of more than about 100 m from the pumping well. Radial and bilinear flows are related to cone of depression and preferential flow paths, respectively.

Dasht-e-Arjan Lake is located in the southern Zagros mountain ranges of Iran. It is a tectonic karst polje, fed by a single spring plus local surface runoff, and drained by a single ponor. The fate of sinking water from the ponor had not been determined before this study. Previous investigations suggested two alternatives: (1) underground flow ~ 24 km southwest to Parishan Lake and/or springs in the Kazeron area, and (2) underground flow ~ 20 km southeast to springs at Domeasb Gorge. To clarify this question, an integrated hydrogeological study including a dye tracer test was undertaken. Comparison of the discharge into the Arjan ponor and the discharge of the Domeasb springs shows a reliable correlation. Further, annual water balance analysis reveals that the discharge at the Domeasb springs is too great for its topographic catchment, requiring a major additional source. The dye tracer test, using 5 kg of uranine, was detected only at the Domeasb springs. Flow through time after the dye injection was 210 h. The tracer breakthrough curves are very simple and sharp, with a clear peak concentration of 7 ppb. The total tracer mass recovery and groundwater flow velocity were estimated to be 58% and 105 m h−1, respectively. The results indicate a simple and direct limestone karst hydraulic connection along the geologic strike between Dasht-e-Arjan Lake and the Domeasb springs. The results clearly suggest the possibility of designing a dam around the ponor to protect drying up the lake during drought periods.

The hydraulic properties of hard rock and karst formations are typically highly heterogeneous. Quantification of the groundwater flow in these media is a complicated process with numerous uncertainties. There are several approaches to overcome this problem. Stochastic simulation is one of the promising approaches for quantification of the intrinsic uncertainty in karst formations. In the present research, the sequential indicator simulation was used as a tool for the simulation of the hydrogeological and geotechnical properties of a karstic formation at the Karun IV dam site, Iran. In this regard, two variables of interest, RQD and Lugeon, were considered as criteria for the development of karst. One hundred realizations of these variables were produced and then combined to obtain an expected value and the probability of exceeding a threshold value for each cell. The results showed a weak negative correlation between the expected values of RQD and Lugeon. Also, it was determined that the geologic structures (i.e., faults) and lithology play an important role in the development of karst in the study area. A decrease in the simulated values of Lugeon was observed with increasing depth of karstic layers and the distance from faults. The results of this study helped to overcome the intrinsic uncertainty in karst media and to present a conceptual model of karst development. This approach can be applied in other dams for the evaluation of risk and uncertainty of hydrogeological parameters.

Data on hydrogeological and lithological characteristics of geological formations are obtainable at every dam site. In practice, it is essential to examine the data using an engineering-based approach such as geostatistical methods. Specifically, it is important to recognize the principal factors that significantly control hydrogeological characteristics, such as Rock Quality Designation (RQD) and average hydraulic conductivity of the rock mass (Lugeon value). In this study, the statistical properties of these parameters are examined as functions of depth, lithology and the spatial distribution of the data points. The results of this study show that the spatial distribution and lithology are the main controls. Using this approach one can determine the most permeable rock formations, potential karst locations and the areas that may warrant more extensive study. The results of stochastic simulation could be extensively enhanced by considering the spatial distribution of hydrogeological characteristics.

Earth science phenomena are primarily spatially dependent with variations occurring on varying scales. Geostatistics is a well-known approach for the assessment of spatial models of regionalized variables, such as porosity. In this study, we used the results of 953 Lugeon tests (402 tests in exploratory wells and 550 tests in grouting wells) to assess effective porosity and karst development at the Tangab Dam site, Iran. Lugeon values were first normalized, following which the variogram method (i.e., range, sill, and nugget effect) was used to identify the spatial variability of Lugeon values. A high correlation range of 200 m was obtained along the grout curtain galleries, while the range was about 40 m in the vertical direction. The simple kriging method with Gaussian variograms was determined to be the best method for estimating the Lugeon values in the study area using cross validation-criteria (e.g., RMSE = 0.835 and ρ = 0.914). Spatial variation of Lugeon values was mapped using a simulated annealing approach. The analysis revealed: (1) a higher potential for karst development on the left abutment of the Tangab Dam site, (2) the average of simulated Lugeon values decreased from values of about 180 close to the ground surface at 1,440 m a.s.l. to lower values of about 50 at 1,260 m a.s.l., and (3) high Lugeon values in the abutments of the dam site approximately followed the 15° bedrock dip, which confirms the potential development of karst features in the Asmari limestone. A schematic model for development of karst at the Tangab Dam site is proposed based on hydrogeological data and results of simulated Lugeon values around the dam site.

This study presents the effect of annual precipitation amount (P) on the characteristics of spring hydrograph such as recession coefficient (α), volume of dynamic flow (V d), base flow (V b), and quick flow (V q) in two karst springs, Pireghar and Dime from the Zagros Region, Iran. Length of the studied period in Pireghar and Dime Springs is 10 and 11 water years, respectively. Hydrograph characteristics such as α’s, V d, V b, and V q were computed according to the annual spring hydrograph for each water year. The results reveal negative correlation between P and α 1 and positive correlation between P and V d and V q in Pireghar Spring. There is no increasing trend between P and V d in Dime Spring. Results show that analysis of several annual hydrograph provides detailed knowledge about development of karst. The behavior of the karst system in Pireghar Spring reflects higher degree of karst development in comparison to Dime Spring based on the values of quick flow, infiltration rate, and time variation of the spring discharge. A schematic conceptual model was proposed for the studied springs responsible to water storage and movement. As a result, the hydrodynamic characteristics of spring hydrograph are controlled by type of recharge, type of storage, and mechanism of water movement in the karst aquifer.

A geostatistical approach was used to study temporal structures in a time series of discharge and electrical conductivity (EC) in 15 karst springs from the Zagros mountain range, Iran. Two types of temporal behaviors, a periodic structure and nugget effect, plus one or two temporal structures, were identified and interpreted. These correspond to characteristics of karst systems, such as the catchment area, percent of conduit flow, and general degree of karst development. Springs were grouped into three categories based on their ranges (e.g., residence time) obtained by variogram analysis. The first group of springs include those that present the same temporal behaviour in variograms of discharge and EC. These springs are characterized by generally constant EC with increasing discharge suggesting the existence of a large underground reservoir. The second group of springs are those with varying temporal periodic behavior in variograms of discharge and EC. Positive correlation between discharge and EC values is the main characteristic of these springs and is interpreted to result from a piston-flow system in poorly developed karst aquifers. The third group of springs includes those that exhibit different temporal behaviors when compared with the periodic and non-periodic variograms. This group exhibits a negative correlation in scatterplots of discharge versus EC values suggesting a well-developed solution-conduit system that facilitates rapid response of the karst system to precipitation events. This study's results document the role of variogram analysis in delineating temporal structures of spring behaviors by means of time series of discharge and EC. Variogram analysis can be considered as a valuable tool for hydrogeological investigations in karstic terranes. DOI: 10.4311/jcks2009es0079

The Asmari limestone formation is the major aquifer system at the Khersan 3 Dam site, Zagros, Iran. Characterization of the aquifer system and study of karst development are essential for forecasting leakage potential and to plan remediation works. The aquifer functioning and karst structure were evaluated by geology, well hydrodynamics and natural tracing studies, showing one unconfined and three artesian sub-aquifers, the last being characterized by rapid flow, with contributing old recharge water, and a recharge at higher elevation than the unconfined sub-aquifer. The anticline structure outcrops the artesian sub-aquifers downstream of the site. The confining layers disappear and the aquifer system discharges as a mix of all groundwater. Artesian groundwater is estimated to contribute about 80% (dry season) and 50% (wet season) of flow in the mixing zone. Very few karst features are observed in cores and galleries at the site, despite some karst landforms in the reservoir area. However the aqu...

The paper discusses the investigations into the Khersan 3 Dam site in the Zagros Zone, southern Iran, where the Rig, Shorom and Laki Anticlines meet. The Asmari Limestone is the main aquifer system in the area of the dam site and its reservoir. Marly bands within the Lower Asmari Unit create three confined sub-aquifers under artesian pressure overlain by one unconfined sub-aquifer. The impounded water will be in direct contact with the Asmari Limestone hence leakage may occur through existing karst features. As the recharge zone of the artesian sub-aquifers is higher than the normal water level in the reservoir, no direct leakage through the artesian sub-aquifers is anticipated. However, when the water head in the unconfined sub-aquifer rises with the construction of the dam, leakage may occur and further karstification develop.

The excellent topographic condition of the limestone canyons for dam construction may be rejected if they are karstified. Karst features cause the reservoir not to be impermeable enough to permit the water to fill it and leakage occurs and often increases with time. Moreover, karst features may involve the stability of the dam itself. A few operated dam sites at the Zagros Zone encountered a leakage problem. Furthermore, more than 30 dams are presently under study for construction in the Zagros Zone. Karst conditions and leakage potential were investigated at an under-study site (Khersan 3 Dam) for assessing the general methodology for the study of leakage potentials. Conventional methods for studying karst features, geological mapping, geomorphology and extensive borings were applied before the dam was constructed. These methods are not efficient enough to precisely reveal the karst structure, especially hidden and paleokarst, nor the hydrological behaviour of the karst structure in different settings of groundwater flow. Based on the present case study and previous applied approaches by other authors, this paper introduces a methodology by means of karst structure and functioning approaches at local and regional scales that cover the conventional methods and overcome their shortages. The proposed methodology should be applied before construction of a dam and should include three steps (a) recognition of geological and hydrogeological settings, (b) delineation and functioning of the karst system related to the future reservoir, and (c) assessment of the leakage potentials. Following this methodology, the most probable leakage zone(s) and path(s) at the dam site can be highlighted.

A dye-tracing test is employed to study the karst development and flow regime at an artesian limestone sub-aquifer, the Khersan3 Dam site, Zagros Zone, Iran. Tracer breakthrough curves showed an early dominant peak followed by a pronounced tailing effect. The peak concentration was a response to induced pressure during dye injection. The results suggest that the dye was pushed into the small pores and fissures around the injection point during dye injection. Hence, the dye moved out as a result of matrix and fissure diffusive processes and created a long pronounced tailing. The maximum flow velocity in the upper artesian sub-aquifer ranged from 0.97 to 2.9 m/h. However, the mean tracer velocity ranged from 0.19 to 0.51 m/h based on the mean residence tracer time; consequently, the flow regime in the artesian sub-aquifer was determined to be mainly diffusive. The results reveal (1) a low hydraulic gradient from upstream of the dam axis to downstream; consequently, there is no considerable flow; (2) poor karst development and diffuse flow at the tracing test area; (3) a discharge zone at a location downstream of the dam axis which is the main terminal of general flow direction at the dam site.