Integrating 2D and Pseudo-3D Electrical Resistivity Imaging to Determine the Recharge Potential of Karst Surface Fractures: An Example in the Northern Segment of the Edwards Balcones Fault Zone (BFZ) Aquifer

Abstract

This study investigates the hydraulic connection of surface karst features within the Northern segment of the Edwards Balcones Fault Zone Aquifer, using a combination of 2D and pseudo-3D Electrical Resistivity Tomography (ERT) at an outcrop near Salado, Texas. The study site features several surface fractures whose hydrological functions are not well understood. Nine ERT profiles and two pseudo-3D models were used to evaluate the connection between surface fractures and subsurface karst conduits. Karst features at the study site were physically evaluated using characteristics such as morphology, which resulted in the identification of three surface fractures (F1, F2, and F3). The ERT results showed several high-resistivity anomalies interpreted as a poorly fractured zone and low-resistivity water-filled conduits within the Edwards Formation. Furthermore, the result reveals that slow hydraulic connectivity exists in F1 and F2; however, F3 presents a low-resistivity zone that extends vertically into the subsurface, which suggests that F3 may serve as a potential recharge feature to the Edwards Aquifer. These findings are corroborated by a water percolation test, as water penetrated more at F3 compared to F1 and F2. This study showed that the combined application of 2D and pseudo-3D ERT can successfully delineate potential recharge pathways in an exposed karst system, thereby constituting a supportive approach providing critical insight into recharge and the vulnerability of karst aquifers to contamination.

1. Introduction

Karst aquifers provide water for hundreds of millions of people globally [1]. In karst aquifers, karst conduits are the main groundwater flow paths, and pollutants can move through them quickly [2,3], causing widespread groundwater contamination; hence, the development of karst conduits through the dissolution process in karst terrains makes these aquifers especially susceptible to contamination. Furthermore, karst aquifers are known to be more vulnerable to pollution than other aquifers because of the concentrated flow path that supplies fast and direct recharge to the aquifer. These concentrated or focused flow paths include karst features such as solution-widened fractures, caves, and sinkholes found within the aquifer. While many of these karst features can be shallow and small, they may also indicate a larger, well-developed flow through networks of conduits into the aquifer system [4], which means that rainfall can easily infiltrate into these karst features, thereby recharging the aquifer [5]. However, not all karst features within the aquifer’s recharge area have the potential to contribute equally to aquifer recharge, highlighting the need to evaluate these features to determine their suitability for recharging the aquifer. The recharge potential of karst features can be assessed using geomorphological and geophysical methods to determine whether hydraulic connectivity exists between the surface karst feature and the aquifer, and whether rapid infiltration may occur. Previous studies, such as Panizza et al. [6], Veni [7], and Lindley and Hovorka [8], have used physical and geomorphological characteristics, such as airflow, cave fauna, sediments, morphology, feature alignment, topography, and lithology, for identifying and evaluating karst features and their potential to recharge karst aquifers and springs. Veni [7], in particular, was able to use these geomorphological characteristics to identify 246 potential recharge karst features, including 110 fissures, on the Camp Bullis Military Training Installation in Texas, USA. The authors suggested that more comprehensive studies, including excavations, subsurface cave assessments, dye tracer testing, and geophysical methods, could help validate or enhance the technique. Electrical resistivity tomography (ERT) stands out as a particularly valuable geophysical approach that has been extensively applied in karst environments. ERT is an effective technique for identifying and detecting subsurface anomalies in karst terrain, such as voids and karst features [9,10,11,12]. ERT technology can identify and locate targets that have resistivity values distinct from their surroundings by measuring the voltage drop between the potential electrodes. If the target has a lower resistivity than the surrounding background, such as water sources, water-filled karst features, or areas affected by marine transgression, the acquired data will indicate lower voltage, corresponding to the low resistivity of the target [13,14,15,16,17]. Conversely, when the target has a higher resistivity than the background, like air-filled cavities, the data will display the opposite effect [18,19,20,21,22]. This study employs electrical resistivity tomography (ERT) to investigate the hydraulic connections between surface karst features and the Edwards Aquifer. The research focuses on an outcrop near Salado, Texas, located within the recharge area of the northern segment of the Edwards BFZ Aquifer. The study site is particularly important because it lies along a fractured losing-stream segment, a type of recharge environment that has been hypothesized, but rarely characterized in detail [23]. Losing streams with discrete fracture networks may provide focused recharge to the aquifer, yet their hydraulic behavior remains poorly understood, particularly in this least-studied portion of the Edwards BFZ Aquifer [23]. Unlike well-developed caves or clearly connected conduits documented in previous studies [24], this site contains subtle, surface-expressed fractures whose connectivity is uncertain. Investigating such features is critical for understanding localized recharge dynamics and vulnerability in heterogeneous karst systems. Therefore, studying this site fills a significant knowledge gap by evaluating a recharge pathway type that has received little direct geophysical or hydrologic investigation. In general, this study focuses on integrating surface fracture characterization with high-resolution 2D and pseudo-3D ERT to evaluate hydraulic connectivity in exposed limestone conditions. This approach aims to clarify which surface fractures function as active recharge pathways within the Northern Segment of the Edwards BFZ Aquifer.

2. Materials and Methods

2.1. Geological Setting of the Study Area

The Edwards Balcones Fault Zone (BFZ) Aquifer is a karstic aquifer that spans an outcrop area of 1566 square miles through Central Texas, and it lies along the normally faulted Balcones Fault Zone [25]. It consists of highly faulted and fractured carbonate rocks from the Cretaceous period [25]. The Edwards (BFZ) Aquifer is a heterogeneous and anisotropic aquifer that contributes to a significant variation in the permeability, transmissivity, and storativity of the aquifer. The aquifer is divided into three hydrogeological segments: the San Antonio Segment, the Barton Spring Segment, and the Northern Segment [26]. This study will focus on the Northern Segment of the Edwards (BFZ) Aquifer. The Northern Segment of the Edwards (BFZ) Aquifer has been the least explored, and as a result, is possibly the least understood portion of the Edwards (BFZ) aquifer [27]. Although it is less prolific than the San Antonio portion, it is the principal water supply for most of the population in and surrounding the Salado Creek basin in Bell County [27]. In addition, the aquifer is linked to several significant springs in the vicinity, which are popular tourist attractions in Salado Village [27]. The Northern Segment of the aquifer runs along Interstate (I-35), underlying the Northern Travis, Williamson, and Bell Counties [26]. The aquifer lies within the jurisdiction of the Clearwater and the Underground Water Conservation District (CUWCD) and contains both unconfined and confined portions. The unconfined portion has exposed Comanche Peak, Edwards Limestone, and Georgetown Formations. Brune and Duffin [25] defined the limited segment as the aquifer’s down-dip portion, where the Edwards Limestone and adjacent limestones are hydraulically connected but covered by the Del Rio Formation, also known as Del Rio Clay (Figure 1). The Walnut Formation, notably the Keys Valley Member (Figure 1), is the underlying confining layer and provides little or no water [28]. This formation is composed of carbonaceous clay, often known as marl. The overlying confining layer is the Del Rio Formation, also known as the Grayson Formation. Together, these two formations serve as aquitards for the Edwards Limestones, which contain the aquifer. The Eagle Ford Group, Buda Formation, and Austin Chalk, Cretaceous units that surface in the Salado Creek basin, sit above the Del Rio Formation. None of these Cretaceous units are classified as aquifers in this area. Figure 2 depicts a more detailed geology map of the northern portion.

The study area experiences a subtropical climate that ranges from sub-humid to semiarid conditions [26]. Across central Texas, temperatures average about 68 °F over the year. Winters are relatively mild, with typical lows around 41 °F, whereas summers are characterized by high heat, with average maximum temperatures reaching roughly 95 °F. Long-term precipitation data collected by the U.S. National Weather Service over a 76-year period (1900–1976) indicate an annual average of approximately 33.5 inches [25]. Rainfall is usually at its lowest during July and August, while the heaviest precipitation commonly occurs in May and September.

2.2. Methodology

2.2.1. Physical Evaluation of Karst Features

Karst features in the study site were assessed based on their morphological characteristics, following one of the approaches adopted by Veni et al. [7] as a geomorphological technique for evaluating karst features. This morphology-based strategy emphasizes the size and shape of the karst features to determine their hydraulic connectivity. Elongated or narrow karst features are likely to form along individual fractures and may not indicate significant hydraulic connectivity to the subsurface, whereas karst features with near-circular or irregular shapes suggest the intersection of fractures, which may result in greater permeability. As a result, surface fractures were identified using this morphology-based technique. Thereafter, the electrical resistivity method was utilized to investigate the internal structure of these surface fractures to determine their hydraulic connectivity with the Edwards Aquifer.

2.2.2. Electrical Resistivity Method

Electrical resistivity survey is a reliable geophysical technique for detecting karst features [29,30]. The procedure involves introducing an electrical current into the ground through two stainless steel electrodes while simultaneously measuring the voltage drop or potential difference between two other metal electrodes inserted into the ground. As a result of the voltage drop and the current flow between the electrodes, the variation in the electrical resistivity of the subsurface can be identified and mapped. The resistivity survey for this study was conducted using a 56-electrode, 8-channel Advanced Geosciences Inc. (AGI, Cedar Park, TX, USA) SuperSting™ R8 resistivity meter, powered by a 12 V battery, with an electrode spacing of 2 m. The electrode spacing of 2 m was selected to provide sufficient resolution for imaging shallow near-surface karst features, which typically occurs in the Edwards Limestone formation. This spacing offers a balance between the depth of investigation and the ability to resolve narrow fracture zones expected in exposed limestone terrains. A dipole–dipole array was selected for data acquisition due to its superior vertical and lateral resolution compared to other electrode configurations, making it especially effective for imaging karst fractures and cavities. Considering that the study site is dominated by outcropping Edwards Limestone (Figure 3A), it was necessary to drill into the rock to properly install the electrodes. To ensure reliable data collection, we performed a contact resistance check after installation. This involved measuring the resistance at each electrode to verify that it had established a strong electrical connection with the ground. Electrodes with high contact resistance were re-seated to reduce resistance and improve conductivity. Maintaining low contact resistance across all electrodes was critical to achieving high-quality resistivity measurements, particularly in the challenging conditions presented by the hard limestone surface. Furthermore, while conducting the resistivity survey, a Leica GS18T RTK Rover was used to record the GPS coordinates of each electrode along the survey array. The elevation data obtained during the ER survey were later used to correct the resistivity measurements for topographic variations across the site. A total of nine-oriented 2D resistivity profiles were measured (Figure 3B). Six of these profiles were positioned parallel to each other, intersecting two identified fracture zones (F1 and F2). The remaining three profiles were also arranged parallel to one another, but were designed to cross the identified F3. In other words, the resistivity profiles were arranged parallel to each other as a pseudo-3D configuration setting to cover the full spatial extent of the identified surface fractures. Because these karst features are not a single continuous fracture but a cluster of short, intersecting fractures scattered across the area, the resistivity profile were strategically positioned to ensure that the entire fracture cluster was captured within the imaging footprint, thereby boosting the likelihood of intersecting any vertically continuous conductive anomalies. All 2D resistivity data were processed or inverted using EarthImager-2D™ V2.4.2 software developed by Advanced Geoscience Inc. (AGI), Austin, TX, USA. The goal of the inversion iterations was to reduce the mismatches between the modeled and measured resistivity data for every profile. The root mean square (RMS) misfit was used to evaluate the goodness of fit. After multiple trials, each profile in this investigation had a final RMS misfit of less than 10%. Beyond the 2D resistivity images, pseudo-3D resistivity images were also created. This pseudo-3D approach involved combining multiple parallel 2D resistivity profiles to produce a comprehensive 3D subsurface visualization. For this study, we utilized the nine parallel 2D resistivity profiles with a profile line spacing of 4 m, which is twice the electrode spacing of 2 m. In environmental investigations, the recommended maximum profile line spacing for pseudo-3D resistivity surveys is four times the electrode spacing [31]. The pseudo-3D resistivity data processing was accomplished by inputting the 2D resistivity data into AGI EarthImager3D V1.5.3 software. The generated pseudo-3D resistivity images display the resistivity distribution that most accurately corresponds to the field measurements.

2.2.3. Water Percolation Test

A water percolation test was carried out after the ERT survey to evaluate the hydraulic function of karst features by determining their hydraulic connectivity to the underlying aquifer. For the percolation test, a tank containing 260 gallons of water was transported to the study site, where 60 gallons of water were discharged into each selected karst feature using a 100-foot hose (Figure 4), and infiltration was monitored for two hours. The 60-gallon water volume used for the percolation test was based on the estimated surface expression and aperture of each fracture cluster. The volume was selected to ensure that sufficient water was applied to test the infiltration behavior across the entire fracture opening. The objective was to determine how quickly water infiltrates into these karst features, which would indicate the feature’s capacity to transmit water and its likely connection to the aquifer system. Infiltration rates were assessed qualitatively through visual estimation, based on the observed reduction in surface water over the monitoring period. This combined approach provided both surface and subsurface insights into the hydrologic function of the karst feature.

3. Results and Discussion

3.1. Physical Evaluation

A field survey was conducted at the site to identify karst features such as fractures, sinkholes, or dissolution cavities that may serve as potential recharge features to the Edwards Aquifers. Karst features identified during the survey were documented based on its type, location, and physical characteristics such as size and morphology. This procedure led to the identification of three prominent surface fractures designated F1, F2, and F3. The fractures were observed to be filled with soft materials such as loose mud, soil, leaves, and sticks, which may facilitate the infiltration of surface water into the subsurface (see

Table 1. Summary of the result for the physical evaluation of karst features at the site.

Feature Type	Formation	Trends	Aperture (ft)	Infills
Fracture (F1)	Edwards	N31$°$E	0.08–3.67	Clustered fractures filled with leaves, sticks, loose or soft mud
Fracture (F2)	Edwards	N60$°$E	<0.10	Clustered fractures filled with leaves, sticks, loose or soft mud
Fracture (F3)	Edwards	N36$°$E	<0.10	Clustered fractures filled with leaves, sticks, loose or soft mud

). Additionally, the fractures were oriented NE-SW and exhibited irregular shapes and clustering (Figure 5). While the surface fractures (F1, F2 and F3) occur as irregular clusters rather than clean, planar surfaces, a reliable dip angle could not be measured during the field survey. Although the clustering may suggest that the features may form at the intersection of fractures, which could enhance infiltration, further evaluation is needed to determine the veracity of the physical evaluation, especially since it is solely dependent on surficial physical characteristics. Hence, the introduction of ERT to image the underground structures of these fractures to determine their suitability to serve as a potential recharge zone to the Edwards Aquifer.

3.2. ERT Results

The resistivity section of ERT Profile 1 (Figure 6) reveals a surface layer with low resistivity values (<40 ohm-m, shown in dark blue), interpreted as a clay-rich zone. This low-resistivity zone was observed at the start of the profile in the SW end and extends for approximately 28 m. Between profile lengths of 28 m and 37 m, this low-resistivity layer thins out significantly, thereby corresponding with a visible surface depression. This low-resistivity layer is interpreted as an exposure of the underlying Edwards Limestone at the surface in the study site, where the resistivity values rise above 150 ohm-m and are displayed in green tones. Beyond the profile length of 38 m, the low-resistivity clay-rich zone reappears and continues toward the NE end of the profile. Beneath the clayey layer, the Edwards Limestone is visible throughout much of the subsurface in green to yellow shades, with resistivity values ranging from 150 to 1000 ohm-m, consistent with compact limestone. The distribution of resistivity indicates that the limestone is relatively continuous beneath the profile. A distinct low-resistivity anomaly (10–45 ohm-m, deep blue) is present between distances 10 m and 28 m along the profile with an elevation of 190 m and is interpreted as a potential conduit zone within the Edwards Limestone. The shape of this anomaly suggests an inclined orientation; however, it does not appear to reach the surface in this cross-section. No surface fracture expression was observed at this location during the resistivity survey for Profile 1, confirming that the anomaly likely represents a subsurface structural feature. In Profile 2 (Figure 7), this same feature appears to continue but with reduced thickness. Furthermore, this low-resistivity zone does not show hydraulic connectivity to F1, which is located at a profile length of 32 m. In Profile 3 (Figure 8), the ERT resistivity section is characterized by prominent zones of high-resistivity anomalies. A distinctive high-resistivity anomaly reaching up to 10,000 ohm-m was observed between surface distances of 10 m and 33 m along the profile at an elevation between 181 m and 192 m. Theoretically, such high-resistivity zones in limestones or other carbonate rocks may result not only from cavities or air-filled voids, but also from non-fissured or poorly fractured parts of the rock where little or no secondary porosity exists. However, considering the location and consistent patterns observed in Profiles 1 and 2, this anomaly is interpreted as a section of poorly fractured Edwards Limestone, which is likely less saturated compared to the surrounding material.

The resistivity section of Profile 4 intersects F2, identified at the site where the Edwards Limestone crops out between distances 26 and 38 m (Figure 9). Beneath the location of this surface fracture, zones exhibiting both low- and high-resistivity anomalies are present. The low-resistivity zones (less than 70 ohm-m) nearest to F2 appear between distances 4 and 24 m, where the anomaly seems to extend vertically in depth, and between 34 and 43 m, where the anomaly appears to be oval-shaped. Additional small segments of low-resistivity zones are located farther from the fracture, at distances between 50 and 60 m and between 70 and 86 m, all occurring at elevations below 192 m. These low-resistivity zones are interpreted as flooded conduits within the Edwards Limestone. The location of this low-resistivity zone and the composition of the Edwards Limestone bedrock as a non-marly limestone [25] further confirm these anomalies as water-filled conduits rather than wet-clay conduits that have similar low-resistivity responses. The low-resistivity zones between 50 and 90 ohm-m (light blue) surrounding these conduits are interpreted as the projection of karst weathering due to the dissolution or karstification process that formed the conduits (cave-like channels). Importantly, the resistivity section of Profile 4 shows no indication of connectivity between the subsurface conduits and the surface fracture, indicating that these features do not serve as potential recharge pathways to the Edwards Aquifer at this location. The small pockets of low-resistivity anomalies identified in Profile 4 between distances of approximately 50–60 m and 70–86 m appear to persist in Profiles 5 and 6. In Profile 5 (Figure 10), these anomalies are shown in light blue, with resistivity values ranging from 90 to 110 ohm-m, which is characteristic of karst weathering caused by dissolution processes that have led to the formation of subsurface conduits within the Edwards Formation. A similar trend was observed in Profile 6 (Figure 11), where the low-resistivity anomalies reduced in size and geometry. Even though the anomalies are still present, they are less pronounced in Profile 6, indicating that the karstic features may be less developed in this area where the profile is located. This suggests a spatial variability in the extent and intensity of the dissolution processes affecting the Edwards Formation.

The pseudo-3D-resistivity section shows a detailed visualization of the subsurface conditions at the study site. In other words, this pseudo-3D-imaging can reveal vital information about the hydrologic interconnectivity between the identified surface fractures and subsurface structures within the Edwards Limestone formation. In Figure 12A, the pseudo-3D-resistivity section displays the resistivity distribution throughout the surveyed volume, with values ranging from 22 ohm-m to 10,000 ohm-m. The resistivity distribution surrounding these fractures shows predominantly moderate to high resistivity values (green to yellow colors, approximately 474–2177 ohm-m), indicative of Edwards Limestone, which is exposed at the surface in the survey area. Figure 12B shows clearly a single subvertical low-resistivity zone underneath the location of F1 and F2, indicating a water-bearing feature (conduit) within the subsurface. These low-resistivity zones appear as disconnected bodies that do not extend to the surface, where F1 and F2 are located. In other words, there is a clear absence of vertical connectivity between the surface fractures and the deeper low-resistivity anomalies. The disconnection between surface fractures and subsurface conditions corroborates the findings from the 2D resistivity sections. Although these fractures were visible at the surface in the study site, they do not appear to act as a potential recharge feature to the Edwards Aquifer at this location. The low-resistivity anomaly visible in the deeper subsurface may be interpreted as water-filled conduits formed through karst dissolution processes; however, their lack of connection to F1 and F2 suggests that they are part of a deeper groundwater system with recharge occurring elsewhere in the aquifer system.

In Profile 7 (Figure 13), four segments of the low-resistivity zone are observed and are interpreted as karst conduits. The shape of these low-resistivity anomalies (approximately 24.4 to 110 ohm-m) appears to be inclined, indicating the possibility that the karst conduits are vertically oriented and are likely saturated with groundwater at this location. Two karst conduits that appear to extend deeper into the subsurface are observed near the center of the resistivity profile, suggesting a well-developed channel. Above these two major conduits located at the center of the profile, a distinct high-resistivity anomaly with resistivity values up to 10,000 ohm-m is evident and is interpreted as a poorly fractured Edwards Limestone due to its resistivity contrast with the surrounding materials. This poorly fractured zone indicates a drier or more-competent rock compared to the surrounding rock. In Profile 8 (Figure 14), three low-resistivity anomalies are observed, which also indicate potential karst conduits within the Edwards Limestone Formation. These anomalies are generally broader and subvertically elongated, which also suggests that the conduits in Profile 7 may be deeply incised. The resistivity section of Profile 8 intersects F3 at a distance of 24 m, and this feature appears to align with a vertically continuous low-resistivity anomaly extending from the surface into the subsurface. This alignment may suggest a strong possibility of a hydraulic connection between the surface and subsurface at this location, indicating that F3 may serve as a potential recharge feature for the Edwards Aquifer and can enhance infiltration of surface water into the subsurface conduits. Profile 9 (Figure 15) shows a resistivity distribution that indicates a relatively continuous subsurface with fewer and less pronounced low-resistivity anomalies compared to Profiles 7 and 8.

The pseudo-3D-resistivity section for Profiles 7–9 further provides compelling evidence that a hydraulic connectivity exists between F3 and the subsurface. Figure 16A shows the general distribution of resistivity across the three profiles with resistivity values ranging between 35 and 10,000 ohm-m; however, in Figure 16B, the connectivity of the low-resistivity anomaly becomes more apparent. Figure 16B shows a low-resistivity zone (<100 ohm-m) that extends downwards from the surface location of F3 into the subsurface, suggesting that recharge infiltrates downwards through this vertically oriented low-resistivity zone, thereby serving as a possible recharge pathway into the Edwards Aquifer. Furthermore, this low-resistivity anomaly may represent a zone of increased water saturation in the Edwards Formation, thereby corroborating the interpretation that F3 is not an isolated feature, but rather hydraulically connected to deeper portions of the Edwards Aquifer in the Northern Segment.

3.3. Water Percolation Test

To provide support for the geophysical interpretation, water percolation test was conducted to provide qualitative data on infiltration capacity of the identified surface fractures and to also assess their hydraulic conductivity to the subsurface. The test involves injecting 60 gallons of water into the identified F1, F2, and F3, and infiltration was monitored for two hours. The result of the water percolation test for F1 showed minimal water infiltration (approximately <5%) into the subsurface after two hours of monitoring the fracture location, thereby providing direct empirical corroboration of the resistivity survey results showing the absences or lack of connected vertically low-resistivity anomaly to deeper subsurface zone. The result suggest that F1 may be terminated or sealed at shallow depth and may not act as recharge pathway to the Edwards Aquifer at this site. Similarly, the water percolation test for F2 also showed limited infiltration of approximately <5% within the two-hour period. Furthermore, visual observation at the location of F1 and F2 confirmed that the majority of the water remained stagnant at the surface following the test, corroborating the inference from the ERT data that F1 and F2 are not hydraulically connected to the subsurface conduit. In contrast, the water percolation test of F3 showed rapid water infiltration into the subsurface, which indicates stronger permeability capacity compared to F1 and F2. The combination of the water percolation and ERT results provides evidence that suggests that F3 is an active karst feature that probably serves as a recharge pathway for the Edwards Aquifer in the Northern Segment. The ability of this recharge feature to facilitate rapid water infiltration makes it a critical conduit in the hydrogeologic framework of the study area.

4. Summary and Conclusions

In this study, we successfully integrated the Electrical Resistivity Method with water percolation tests to evaluate the hydraulic connectivity of surface fractures within the Edwards Limestone formation in the Northern Segment of the Edwards Balcony Fault Zone (BFZ) Aquifer. The combined application of the 2D and pseudo-3D Electrical Resistivity Tomography provided a detailed imaging of subsurface resistivity anomalies within the Edwards Limestone formation, where low-resistivity anomalies were interpreted as karst conduit and high-resistivity anomalies as poorly fractured Edwards Limestone. Three surface fractures, F1, F2, and F3, were identified during the physical assessment of the study site. Among the three identified surface fractures, only F3 showed substantial evidence of hydraulic connection with the deeper portion of the Edwards Aquifer at the study site. This conclusion was supported by both the ERT data and water percolation test, where the ERT result revealed a vertically low-resistivity anomaly that extended downward from F3, and the water percolation test that revealed rapid infiltration into the subsurface at the surface location of F3. In contrast, F1 and F2 lacked the downward projection of a vertically oriented resistivity anomaly and also exhibited minimal infiltration during the field test, which indicates that they are not significant recharge features. Additionally, field observations showed that F1, F2, and F3 exhibit similar types of surface infill materials, including soil, loose mud, leaves, and small woody debris. Therefore, the presence of these materials alone does not explain differences in hydraulic behavior. Despite having comparable infills, only F3 displayed a vertically continuous low-resistivity anomaly indicative of active hydraulic connectivity with the underlying aquifer. This suggests that hydraulic connectivity in this setting is controlled more by fracture openness, aperture continuity, and subsurface structural alignment than by the type of infilling material present at the surface. By integrating 2D and pseudo-3D ERT imaging with direct water-percolation testing, this study provides one of the few comprehensive characterizations of a fractured losing-stream system and its potential for discrete recharge through individual surface fractures. This approach not only documents a site type seldom evaluated in the literature, but also offers new insights into how losing-stream fractures can function as localized recharge pathways within heterogeneous karst terrains. These research findings attest to the spatial variability of recharge potential in karst terrains and highlight the significance of conducting site-specific studies when evaluating and assessing recharge mechanisms and the vulnerability of the karst aquifer.