Introduction

Linear features on the earth surface attract the attention of many scientists in the field of geology since early 20th century (Hobbs 1904; O’Leary et al. 1976; Zakir et al. 1999). The first usage of the term “linear” is probably utilized by Hobbs (1904) to refer the linear feature on the earth surface. O’Leary et al. (1976) defined the lineament as a mappable, simple, or composite linear feature of a surface whose parts are aligned in a rectilinear or slightly curvilinear relationship and which differ from the pattern of adjacent features and presumably reflect some subsurface phenomenon. Sabins (1997) used the term “lineament” to refer to any linear feature on the earth’s surface that reflects an underlying geological structure such as fault, fracture, or joints. Satellite imageries and aerial photos are usually utilized for lineament extraction either manually or through automated processes. Moderate spatial resolution satellite data (e.g., Landsat and Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER)) imageries are commonly used for lineament extraction (Madani 2009). Radarsat and Shuttle Radar Topography Mission (SRTM) elevation data are also utilized to recognize and delineate the topographic lineaments (Masoud and Koike 2011). High-spatial-resolution panchromatic imageries such as Ikonos and QuickBird are used for linear feature extraction. Lineament mapping is an important issue for many geological applications such as (i) selection of the optimum wavelengths used for automatic lineament extraction (Madani 2002), (ii) utilization of ASTER panchromatic band for lineament extraction and studying their relationship to mineral locations (Madani and Emam 2011), (iii) study of the lineament characterization and their tectonic significant (Solomon and Ghebreab 2006), (iv) drainage network geometry versus tectonic elements (Musumeci et al. 2003; Hodgkinson et al. 2006; Ribolini and Spagnolo 2008), and (v) lineament analyses for assessment of groundwater potential (Anudu et al. 2011). This study aims to (i) evaluate the relationship between the lineament trends and the drainage network geometry at Bulghah area, Saudi Arabia, using panchromatic SPOT-5 imagery and SRTM elevation data and (ii) locate the most promising sites for groundwater targeting at the study area.

Study area

Initial examination of the false color composite SPOT-5 imagery helps greatly in defining the study area (Fig. 1a, b). We have selected the southwestern part of the Bulghah gold mine area, Saudi Arabia, to test the purposes of this study. The study area lies approximately 520 km west of Riyadh city, the capital of Saudi Arabia. It is divided into three sectors namely A, B, and C (Fig. 2). In each sector, we are dealing with the main lineament trends and the drainage network geometry. Sector “A” is the upstream part of many small wadies that strike in NW–SE direction and drained in SE direction crossing sectors “B and C” toward the main channel of Wadi Bulghah which has NE–SW strike (Fig. 1d). These wadies are traversed through different rock units such as Hulayfah volcanics, diorite intrusion, and alkali granites. The oldest rock unit exposed at the study area is represented by Hulayfah volcanics. They are made up of the older Afna Formation and the younger Nuqrah Formation. They are intruded by N–S fault-bounded diorite intrusion (Fig. 1c; Delfour 1977). Harbi and Madani (2014) presented a new geological map for the study area based on field data and fusion of SPOT-5 data. The study area is important for gold production. It contains Bulghah gold mine and Bulghah North deposit. Gold production at the Bulghah mine started on 2002 and reached 1.735-t tonnage with 0.7 g/t Au in 2007 (SRC 2007). Water is an important issue used to perform several mining activities at Bulghah gold mine. The present study aims to locate the most promising groundwater potential sites. It will provide decision makers at Bulghah mine with right location to look for groundwater.

Fig. 1
figure 1

Study area location. a Saudi Arabia. b SPOT-5 false color composite 4, 3, and 2 in RGB imagery. c Geological map for the study area modified after Delfour (1977). d The entire drainage network at Wadi Bulghah area

Full size image
Fig. 2
figure 2

A 3-D perspective view showing locations of A, B, and C sectors

Full size image

Regional geology

Figure 1c represents a simplified geological map for the study area modified after Delfour (1977). The country rocks are represented by Hulayfah volcanics which consist of Nuqrah Formation and Afna Formation. The older Afna Formation occurs mainly in the western part of the study area and is made up of andesitic volcanic rocks and volcaniclastic derivatives. Nuqrah Formation consists of acidic volcaniclastic rocks. Discontinuous ridges of gossans are exposed at sector “B” and are distinguished on FCC SPOT imagery by their red image signature (Fig. 2). Discontinuous marble bands forming nearly N-S to NNW trending ridges are exposed at sector “C” to the east of the diorite intrusion. They are commonly occurring within the middle part of Nuqrah Formation.

Hulayfah volcanics are unconformable overlained by Murdama Group which is represented by rhyolite, andesite, sandstone, conglomerate, and marble. Diorite and alkali granite intrusions belonging to Idah (620–615 Ma) and Abanat (585–570 Ma) suites are intruded in the previously described rock units.

Materials and methods

SPOT-5 panchromatic imagery (5-m spatial resolution) and SRTM elevation data are the main remote sensing data used throughout this study for lineament extraction and drainage network derivation. Integration of the processed remote sensing and field data is used to clarify the relationship between the lineament trends and the drainage network orientations at the study area. Many authors utilized digital satellite data for automatic lineament extraction using the following: (i) directional filtering, nonlinear, and morphological spatial filters (Süzen and Toprak 1998; Wahby et al. 2001); (ii) edge enhancement algorithms and linking methods (Koike et al. 1995); and (iii) Fourier transformation and knowledge-based systems (Morris 1991). LINE module (Geomatica package) is used to extract lineaments automatically by performing the following: (i) edge detection, (ii) threshold, and (iii) linear feature extraction steps. The extraction process is performed under user-defined parameters including the following: (i) edge filter radius = 5, (ii) minimum edge gradient = 20, (iii) minimum line length = 5, (iv) line fitting tolerance = 3, (v) maximum angular difference = 15, and (vi) maximum linkage difference = 50. In the first step, edge detection algorithm is applied to produce an edge strength image. Panchromatic SPOT-5 imagery is filtered with a Gaussian function and radius equal to 5. In the second step, the edge strength image is thresholded using value equal to 20 to obtain a binary image. Each ON pixel of the binary image represents an edge element. In the third step, lineaments are extracted from the binary edge image. This step consists of several substeps. First, a thinning algorithm is applied to the binary edge image to produce pixel-wide skeleton lines. Then, a sequence of pixels for each line is extracted from the image. Any line with the number of pixels less than 5 is discarded from further processing. An extracted pixel line is converted to vector form by fitting piecewise line segments to it. The resulting polyline is an approximation to the original pixel line where the maximum fitting error is 3. Finally, the LINE algorithm links pairs of polylines which satisfy the following criteria: (1) two end segments of the two polylines face each other and have similar orientation (the angle between the two segments is less than 15) and (2) the two end segments are close to each other (the distance between the end points is less than 50). This method failed to differentiate between geological and nongeological linear features. Manual editing is used to eliminate the nongeological linear features and to enhance the results. The rest of lineaments which represent the geological structures are recorded in vector segment used later for statistical analyses and rose diagram construction.

Drainage network for the study area is generated automatically from 90-m resolution SRTM digital elevation data using Geomatica software. Two drain modules are used. The first module (DWCON) produces four data sets which are as follows: (i) depression-less digital elevation data, (ii) flow direction for each pixel, (iii) a flow accumulation data set in which each pixel receives a value equal to the total number of cells that drain to it, and (iv) a delta value data set in which each pixel equals the increase in flow accumulation value in the flow direction. DRAIN module is used later to generate the drainage network layer using the flow accumulation data. Vector layer for drainage network orientations is generated and used for statistical analyses and rose diagram construction.

Results

Sector A

Sector A constitutes the upstream part of the study area (Fig. 2). It is characterized by slightly elevated hills dissected by small channels draining mainly toward the SE direction. Parallel and subparallel drainage patterns are predominant in this sector. Rock units that exposed at sector A are made of Nuqrah Formation. They are made up of acidic volcaniclastic rocks including agglomerate, fine-laminated tuffs, and intercalations of jasper and cherty tuffs. It is clearly observed that the structural elements especially faults are the main controlling factor affecting the drainage network orientations in this sector.

Figure 3 shows the main trends of the automatically extracted lineaments over sector A using panchromatic SPOT-5 band with their rose diagram. The main lineament trends are NE–SW and NW–SE constituting the 19.86 and 19.53 %, respectively. The second major trends recorded in sector A are N–S and E–W which constitute 13.65 and 11.76 %, respectively. ENE–WSW, NNW–SSE, and WNW–ESE are subordinate directions constituting 8.65, 8.32, and 7.32 % of total lineament number, respectively. The lineament frequency rose diagram shows the high degree of preferred orientation in the conjugate systems NE–SW and NW–SE, and N–S and E–W directions.

Fig. 3
figure 3

The main lineament trends extracted automatically over sector “A” using panchromatic SPOT-5 imagery with their lineament frequency rose diagram

Full size image

Figure 4 shows the drainage lines digitized from SRTM-derived drainage network for sector A. The rose diagram shows the preferred orientations along NE–SW and NW–SE conjugate system with high frequency in E–W trend. Visual inspection of high-spatial-resolution SPOT-5 imagery revealed the presence of many small E–W drainage channels drained toward the east direction. These channels are responsible for the high E–W drainage frequency.

Fig. 4
figure 4

The main trends of the drainage lines over sector “A” digitized from the drainage network with their drainage frequency rose diagram

Full size image

The histograms comparing the frequencies of the directional trends of lineaments and drainage over sector A are shown in Fig. 5. Both lineament lines and drainage pattern are dominated by NE–SW and NW–SE trends, as indicated by their numbers (179 and 176: lineaments; 22 and 14: drainage) and lengths (35.5 and 41.6 km: lineaments; 7.2 and 4.3 km: drainage). The second fracture systems observed at the study area are N–S and E–W trends.

Fig. 5
figure 5

Histograms comparing the frequencies of the directional trends of lineaments and drainage over sector “A”

Full size image

Sector B

Sector B is the most distinguished sector on the FCC SPOT-5 imagery (Fig. 2). It has a red image signature. It is characterized by plain area with some scattered gossan ridges (Nuqrah Formation) aligned in N–S direction. It is dominated by several NW–SE offsets. In this sector B, the main lineament trend is N–S to NNE (Fig. 6). It constitutes 39.7 % of the total lineament number. The conjugate system NE–SW and NW–SE is also observed and found that they constitute 18.49 and 15.60 % of the total lineament number, respectively. Lineament frequency rose diagram shows the high degree of preferred orientation in the NE–SW, NW–SE, and N–S to NNE trends. Figure 7 shows the drainage lines digitized from drainage network for sector B. They are frequently oriented in N–S and E–W directions representing 39.02 and 19.51 % of total digitized drainage lines. Figure 8 shows the histograms comparing the frequency of the directional trends of lineaments and drainage cover sector B area. When lump, the N–S and NNE lineaments and drainage numbers, we can conclude that

Fig. 6
figure 6

The main lineament trends extracted automatically over sector “B” using panchromatic SPOT-5 imagery with their lineament frequency rose diagram

Full size image
Fig. 7
figure 7

The main trends of the drainage lines over sector “B” digitized from drainage network with their drainage frequency rose diagram

Full size image
Fig. 8
figure 8

Histograms comparing the frequency of the directional trends of lineaments and drainage cover sector “B” area

Full size image
  1. 1.

    N–S to NNE trends are the main directions for both lineaments and drainage networks as indicated by their numbers (69 and 44) while their lengths are 15 and 11.6 km, respectively.

  2. 2.

    NE–SW and NW–SE trends occupied the second rank.

Sector C

It contains large varieties of rock units. It is covered by Hulayfah volcanics, marble ridges, diorite, and alkali granite intrusions. Small wadies having NW and E–W trends are traversing these rock units. They drain east toward the main channel of Wadi Bulghah which has NE–SW strike. The area is characterized by presence of several dike swarms having different orientations. Figure 9 shows the main trends of the automatically extracted lineaments over sector C using panchromatic SPOT-5 band. The main lineament trends are NE–SW and NW–SE constituting 29.68 and 19.22 % of total number, respectively. The second major trends recorded at sector C are N–S and E–W which constitute 10.75 and 10.85 % of total number, respectively. The lineament frequency rose diagram shows the high degree of preferred orientation in the conjugate directions NE–SW and NW–SE, and N–S and E–W. Figure 10 shows the drainage lines digitized from SRTM-derived drainage network for sector C. The rose diagram shows the high degree of preferred orientations in NE–SW and NW–NE conjugate system. Figure 11 shows the histograms comparing the frequency of the directional trends of lineaments and drainage cover sector C area. The lineament and drainage pattern of the study area is dominated by NE–SW and NW–SE trends, which show up in both numbers and lengths.

Fig. 9
figure 9

The main lineament trends extracted automatically over sector “C” using panchromatic SPOT-5 imagery with their lineament frequency rose diagram

Full size image
Fig. 10
figure 10

The main trends of the drainage lines over sector “C” digitized from drainage network with their drainage frequency rose diagram

Full size image
Fig. 11
figure 11

Histograms comparing the frequency of the directional trends of lineaments and drainage cover sector “C” area

Full size image

Field observation

Faults

Field study revealed the presence of three main fault systems observed at the study area. The first fault system is recorded at sector B and is represented by N–S left-lateral strike–slip fault following Nabitah belt trend. The second fault system observed at the study area is represented by left-lateral strike–slip Najd fault system trend in NW–SE direction. The third fault trend has E–W direction. Conjugate fault system strikes in NE–SW and NW–SE directions are observed at sector C crosscuting the diorite intrusion (Fig. 12a). They are characterized by the presence of brownish hematitic clay material along the fault plane (Harbi and Madani 2014).

Fig. 12
figure 12

Field observations from sector “C.” a Conjugate faults with hybrid joint systems crosscut diorite intrusion. b N–S acidic dike cut through alkali granite. c Joint sets in granitic rocks

Full size image

Dikes

Three main types of dike swarms are observed at sector C and are easily distinguished on the FCC SPOT imagery. The first type is porphyritic andesite, about 10 m thick, has NW–SE strike, and is observed at the Bulghah north open-pit site cutting diorite intrusion. The second type of dikes has N–S direction and is represented by acidic dikes cutting alkali granite (Fig. 12b). Pink acidic dikes that have NE–SW direction are also observed.

Joints

Three main joint sets are observed in the granitoid rocks exposed at sector C (Fig. 12c). They are mainly subvertical and have NE–SW, NW–SE, and N–S directions. Hybrid joint systems (extensional and shear) are also recorded at the study area.

Discussion

Lineament–drainage relationships

Visual inspection of the lineaments and drainage maps as well as their rose diagrams at sectors “A, B, and C” revealed the presence of a close relationship between the main lineament trends and the drainage network orientations. Most of the drainage channels are structurally controlled by and follow Nabitah and Najd fault structures. At sectors “A and C,” the conjugate system NE–SW and NW–SE represents the main lineament trends as well as the main drainage network orientations. At sector B, the lineament trends and the drainage network orientations are completely coincide with N–S to NNE, NW–SE, and E–W directions.

Tectonic implications of the lineaments

Sector B is considered as the clue of the structural history of the study area. It contains the most interesting feature observed on FCC SPOT image “strike–slip bend basin” (Fig. 2). The basin elongates in N–S direction and is dissected by NW–SE Najd faults in its southern parts. Madani (2011) studied in detail the structural elements in the study area and its relation to gold mineralization. His study revealed the presence of two main deformational events which had affected the rock units exposed at the study area. The first deformational event is represented by N–S to NNE–SSW Nabitah fault system. Generally, Nabitah belt divided the Arabian Shield into two zones. The inner zone is characterized by sigmoidal fish-shaped shear zones whereas the outer zone is characterized by the presence of gneiss domes (Quick 1991). The older Nabitah structure has been offseted by NW–SE Najd fault system (Figs. 2 and 13) which represents the second deformational event recorded at the study area. Najd fault system is considered as one of the most important feature crosscuting the Arabian Shield and is originated during the Najd tectonic cycle. Generally, it is approximately 300 km wide and extends more than 1000 km in NW–SE trend. It is associated with a complex array of secondary structures including strike–slip, oblique-slip, normal faults and dike swarms. The distribution of secondary structures indicates that compressional and extensional conditions are prevailing at sectors A and C, respectively (Fig. 13). These major fractures are highly susceptible to weathering and form Wadi valleys following N–S, NE–SW, NW–SE, and E–W directions.

Fig. 13
figure 13

Structural interpretation of the major lineaments at the study area

Full size image

Lineament density and groundwater exploration

In fractured aquifer, fractures play an important role in groundwater production. They act as conduits for groundwater flow. Several authors investigated the relationship between the lineaments and the groundwater occurrences (Lattman and Parizek 1964; Woodruff et al. 1974; Odeyemi et al. 1985; Magowe and Carr 1999; Bassey et al. 1999; Anomohanran 2013). They concluded that the groundwater accumulation and, consequently, well production increase in highly fractured areas. In this study, results of lineament extraction process show only the dimensions, directions, and spatial distribution of the lineaments over the study area. To convert this form into density form, a grid with spatial area of about 800 m2 is generated and the cumulative length of lineaments (km) in each grid cell is calculated. Lineament density map (Fig. 14) is generated using density module of Geomatica package to give an idea about the concentration of lineaments in the study area. Areas having high lineament density (white and light grey colors) represent areas with relatively high groundwater potentials. Two promising sites for groundwater targeting are detected based on lineament density map. The first site lies at the intersections of NW–SE, N–S, and E–W drainage orientations at sector A, whereas the second site lies at sector C and occupies the intersection of NW–SE and E–W drainage orientations.

Fig. 14
figure 14

The most promising sites for groundwater targeting for the study area based on lineament density map

Full size image

Conclusion

Results of the present study reveal the presence of strong and close relationship between lineament trends and drainage network orientations at Bulghah area, Saudi Arabia. Most of drainage channels are structurally controlled and follow either Nabitah structures or Najd fault system. At sectors A and C, the drainage channels are completely coincide with the lineament trends and follow NW–SE, NE–SW, and E–W directions. At sector B, the N–S to NNE and NW–SE channel systems are recorded representing Nabitah and Najd fault systems. This study utilized the lineament density map to locate the most promising locations for groundwater potentiality. A site in sector A and another one in sector C are found to be promising spots for groundwater potentiality.