AUTOMATIC LINEAMENTS MAPPING AND EXTRACTION IN RELATIONSHIP TO NATURAL HYDROCARBON SEEPAGE IN UGWUEME, SOUTH-EASTERN NIGERIA

The study focus on the integration of Remote Sensing and Geographic Information System for identification and delineation of lineaments in relation to natural hydrocarbon seepage, which occur in Ugwueme, South-Eastern Nigeria. To achieve this objective, remotely sensed data (ASTER Digital Elevation Model and Landsat 8 OLI/TIRS) were used to depict the surface expression of faults, folds and fractures which are expressed in the form of lineaments. The global positioning system (GPS) was also used for ground verification. The geology map of the study area, which is elucidated in the geology of Nigeria was used to show the distribution of rocks and other geologic structures. The delineation of lineament features was done automatically with the PCI Geomatica while the Rock ware was used to generate the Rose diagram for demonstration of the direction of the extracted lineaments. The classification of the lineaments density and the lineaments intersection analysis were categorized as very low, low, moderate, high and very high classes respectively. Areas classified as very high to high lineaments density are potential zone, which act as conduits for hydrocarbon seepage. The result shows that a total lineament frequency of 947 km and a total lineament length of 946 km were delineated from the satellite data. The result further shows that areas with high lineaments density are concentrated in the southwest, south, central and northern part of the study area while areas with low lineament density were found within the eastern part of Ugwueme. The Rose diagram highlight the major trend in the (NE-SW), (N-S) and (NW-SE) directions, and the minor trend in the (W-E) direction. These directional trends depict the directions of lineaments which act as conduits zones for hydrocarbon seepage in the region. The overall findings of the study shows that lineament density, lineament intersection and rose diagrams are concepts applicable in hydrocarbon oil and gas seepages.


Introduction
The need for studying geologic structures such as lineaments which include faults, joints and fractures cannot be overemphasized. This is because, they do not only act as zones for ores and minerals deposits, but also act as reservoirs for oil, gas and water storage. Lineaments are straight or nearly -straight line features, visible on the earth's surface as faults, folds and fractures which exposes the underlying geologic structure (Hung et al., 2005). These faults, folds and fractures are weak geological plane which play a role as conduits for hydrocarbon oil and gas seepage. The word "lineament" was first used by Hobbs (1904), who describe it as a significant linear landscape that exposes the architecture of different rock basement. for oil, gas, minerals, ores and for ground water studies (Manghany et al., 2009). The surface expression of faults, fractures, folds and foliations are expressed in the form of lineaments with remotely sensed data, such as satellite imageries and the Digital Elevation Model (DEM) (Solomon, 2003). For decades, the existence of lineaments have been of interest to earth scientists. The attention and interest of linear structures to earth scientists grew rapidly with the introduction of aerial photographs and satellite data. Gupta (2003) reported that the extraction of lineament was first observed with stereo pairs, transparencies and light tables, which was the first generation of satellite data. Lineaments identification on satellite data relates to geological zones of weakness on the earth's crust, which represent fracture traces (Sabins, 1997). Schumacher (1999) documented that there is a direct relationship between lineament and hydrocarbon seepage ( Figure 1). Under high pressure, hydrocarbon oil and gas tend to seep within zones of lineament along fractures and faults of soils and sediments, to form seepages at the earth's surface (Okeke & Enoh, 2016). According to Shi et al. (2010), the vertical movement and seepage of hydrocarbon oil and gas within zones of lineament along fractures and faults is called Chimney effect. Awadh et al. (2010) quoted that oil and gas which escape from the earth's impermeable reservoir, seep and remain hidden in the subsurface or migrate through zones of lineaments and weak geological planes to form seepages. Okieimen and Okieimen (2005) stated that hydrocarbon that seeps on the earth's surface or near surface often produce changes associated with the soil chemistry.
Remote Sensing have been widely adopted in the field of geoscience, to express the correlation between lineaments and hydrocarbon seepage (Schumacher, 1999). The techniques provide information for identification, mapping and extraction of lineaments within hydrocarbon seepage prone zones at both distinct and regional level. When compared with the conventional methods, remote sensing technique offers a synoptic, systemic, rapid and repetitive area coverage for identification and extraction of lineaments (Leblanc et al., 2003;Tweed et al., 2007). Remote Sensing technique provides an advantage of spatial, spectral and temporal availability of remotely sensed satellite data, which covers a vast and inaccessible locations within a short time frame (Onyedim & Ocan, 2001). Remotely sensed satellite data, which are acquired from a variable wavelength intervals of electromagnetic spectrum, are more adequate for extracting lineaments than the conventional aerial photographic maps (Casas et al., 2000). Remote Sensing image processing and analysis involves discrimination of lineaments with vary wavelengths intervals (Casas et al., 2000). The various steps involves in the discrimination of lineaments from remotely sensed maps are the enhancement analysis or manual techniques. These techniques are important in lineament analysis for interpretation of areas associated hydrocarbon oil and gas seepage. Remotely sensed data such as satellite imageries and the Digital Elevation Model (DEM) can be used to delineate lineaments by alignment of features. These features include topography, vegetation and soil moisture, which can be studied with tone, color, texture, pattern and topography attributes (Khan & Glenn, 2006). Three methods have been identified for mapping and extracting lineaments from satellite imageries. These methods are the manual technique ; the semi -automatic technique (Jordan & Schott, 2005); and the automatic technique (Rayan, 2013). The automatic technique is the most commonly used method for extracting lineament among researchers. Different automated and semi -automated lineament delineation algorithms exists (Wang & Howarth, 1990). These algorithms include the Hough Transform, Segment Training Algorithm (STM), PCI LINE and the image objective line extraction (Mallast et al., 2011).
The study was conducted in Ugwueme, in the South-Eastern part of Nigeria. It focus on the integration of Remote Sensing and GIS for automatic delineation of faults, fractures and folds features with Line module from the PCI Geomatica (Hashim et al., 2018). The delineated faults, fractures and folds are represented as lineaments, which may either be linear or curvilinear on remotely sensed data. Thematic maps such as the rose diagram, lineament density map and the lineament intersection map which were prepared from these delineated lineaments often act as conduits for hydrocarbon seepage.

Location and extent
Ugwueme is situated on top of Awgu escarpment, in Enugu in the South-Eastern part of Nigeria. Geographically, the region is bound by latitude 6° 0′ 00″ N and 60° 03′ 00′′ N and Longitude 7° 24′ 00′′ E and 70° 28′ 00′′ E coordinates (Figure 2). The region is bordered by Awgu and Onoli towns in the Eastward direction, Nkwe and Mgbidi towns in the Northward direction and to Lokpanta in the South-West direction. Ugwueme has an aerial coverage of 85.6 km, characterized with low built -up area, with an estimated population of 13 000 people (Nigerian Population Commission, 2006).

Climate, drainage, vegetation and soil
Ugwueme experiences two main climatic conditions, the dry and the wet seasons. During the dry season (November-March), the study area receive little or no raindrops with an average temperature of 26.6 °C annually. During this period, the region experiences dryness which is associated with the North-East trade wind that blows from the Sahara (Viessman et al., 1972). In the wet season (April-September), Ugwueme experiences repeated heavy rainfall, with a record of 1800 mm (Enugu State University of Science and Technology, 2018). This heavy rainfall in the study area often aid flooding, leaching, erosion and infiltration in the zone. Ugwueme is identified with many streams, which assume its sources from the apex of Awgu escarpment. Streams which drain the area are Ogwunnu, Echie, Ndumoku, Obae, Ngene Uhie, Aguta -Lokpanta and Iyiohimiri. These water bodies are dendritic in nature, due to the sedimentary formation of the region. The study area is a large forested area, characterized with tall grasses which remains un -cultivated (Okeke & Enoh, 2016). The area has mostly ferralitic soils known as Red Earth, which is poorly drained and mainly suitable for cultivation of cash crops.

Geological setting
The geology (lithology) of Ugwueme is highlight in Nigeria's geological map ( Figure 4). Stratigraphically, the geologic sequence of Ugwueme, consists of Asu river, Ezeaku and Awgu deposits. Others are Nkporo, Mamu, Ajali sandstone, Nsukka, Imo shale, Ameki and Benin formation and Alluvium (Reyment, 1965). The Asu river deposits are composed of micaeous shale and ammonite fauna, which were believed to have been developed from basement complex (Kogbe, 1975). The Asu deposits are dark, sandy and fine in nature while the ammonite fauna exhibit an Albian age. Ezeaku deposit consists of hard shale and siltstone, characterized with different colors and different sediments. These color variation in the deposit is between grey and black while the sediments variations is between sandstone and sandy shale. According to Reyment (1965), Ezeaku exhibit a thick deposit, with variation estimated to be up to 1000 m. Beneath the Ezeagu sediment deposit, is the Awgu sediment formations. These formations are rich in ammonites and are characterized with bedded shale, which is estimated to be up to 400 m thick (Ojoh, 1992). The Nkporo sediment deposit, which is situated above the Awgu deposit is mainly maestrichtian in age. The Owelli sandstone deposits, Enugu and Asata shales deposits are lateral equivalent of Nkporo sediment deposit. Owelli sandstone deposits is estimated to be up to 250 m thick. These sandstone deposits are mainly ferruginous in nature, characterized with medium and coarse grained sandstones (Simpson, 1954). At the hydrocarbon seepage location, situated in Ugwueme, Owelli sandstone deposit is estimated to be 130 m thick (Okeke, 2006), settling on the Awgu shale ( Figure 3).

Data and methods
The first step for the methodology of the study is the selection of the initial input of data for the lineaments extraction. Lineament can be identified and mapped from different sources such as aerial photographs, satellite data and geophysical data etc. In this study, the satellite image was the preferred option for lineament mapping and extraction. The second step for the methodology is the extraction and mapping of lineament from satellite images. The final steps involves the evaluation of the lineament thematic maps. These maps include the lineament density map, direction map, intersection length map and the orientation analysis map. The workflow chart of the study is depict in Figure 5.

Data used for the study
The data utilized for the study were collected from both primary and secondary sources ( Table 1). The primary sources involves the use of GPS receiver to obtain the coordinates of the seepage positions, while the secondary sources include the satellite data, analogue maps covering the study area and the relevant literature which include text books, journals, magazines, weekly newspapers and research websites. The Landsat 8 OLI/TIRS satellite data were downloaded freely at path 188 row 55 and 56 covering the study area from the USGS (2019) official website with earth explorer USGS.gov. It is cloud free and has a spatial resolution of 30 m.

Field work
The field work is an integral part of the study. The field study was observed to validate the hydrocarbon seepage extent in Ugwueme, which was captured with satellite data. The materials adopted for the field study is depict in Table 2. The GPS was used in the field to take the geographic coordinate of the seepage location

Base map
The base map serve as a guide for the study Digital camera The digital camera was used in the field for taking photographs of outcrops and its associated features Field note book, pen and pencil These materials were used for recording purposes

Geology hammer
The geologic hammer is a hard steel hammer, characterized with two ends and a handle. It was used in the study for scooping out sample from loose materials

Major software adopted for the study
In the study, the software were selected based on their ability to solve existing problems, in order to achieve the predetermined objectives. Sequel to this, the following software (Table 3) were adopted for the study.

Pre-processing and processing of the satellite data
The preprocessing of the Landsat 8 OLI/TIRS satellite image were atmospheric and geometric corrections. The acquired satellite data was projected to the Universal Transverse Mercator (UTM) of zone 32 N on the ellipsoid of World Geodetic System (WGS) 1984. The processing steps had include: layer stacking, sub-setting, geo-referencing, mosaicking and band combinations. The layer stacking tool was utilized to combine the very near infrared (VNIR), short wave infra-red (SWIR) and long wave infra-red (LWIR) bands into a new multiband file that enables the manipulations of all bands simultaneously. The sub-setting tool was utilized to specify the extent of the image for analysis. Sub-setting was observed with software, such as ERDAS Imagine, ARCGIS 10.5 and the PCI Geomatica 14 software. The color composite images of the different bands combinations was done. This is the false color composite (FCC) which combines one visible (VIS), one near-infrared (NIR) and one short-wave infrared band R(7), G(5), and B(3) to depict a large color variation for different surface materials which will aid geological interpretation.

Lineament extraction and analysis
The Erdas Imagine, ENVI and the PCI Geomatica are the most widely used software for automatic lineament extraction from remotely sensed data (Hashim et al., 2018). In this study, the lineament extraction was done with line module associated with PCI Geomatica software. The line module of the PCI Geomatica software reduces subjectivity, save time and aid comparison and validation for lineament extraction (Hashim et al., 2018). The logic of this method is similar to STA, where the total number and total lengths of extracted lineaments depends on the input parameter values, represented by optional digits of the LINE modular in PCI Geomatica software. The algorithm of this modular consists of three stages: The edge detection stage, the threshold stage and the curve extraction stage (Hashim et al., 2018). LINE module of the PCI Geomatica often identify and extract lineaments from an image, before converting these features in vector form with the six optional parameters (Table 4). In the study, the automatic lineaments and extraction were statistically analyzed for the preparation of the lineament length map, lineament density map and lineaments intersection map. The Rockwork 14 software was used in the study to prepare the rose diagram, which indicate the direction of the delineated lineament.

Digital Elevation Model (DEM)
The Digital Elevation Model (DEM) can be acquired from different sources such as the SRTM (90×90) m global DEM, Advanced Space borne Thermal Emission and Reflection Radiometer (ASTER) 30 m by 30 m or by the manual interpolating method (Abdullah et al., 2010;Vittala et al., 2005). This study had adopted the ASTER DEM method for the acquisition of the DEM data. DEM is a numerical surface which store elevation for a particular location on a natural terrain ( Figure 6). DEM may store elevation as either Grid or TIN (Carter, 1988). With the grid storage format, DEM store elevation in regular arrays and with the TIN storage format, elevation is stored at irregular distributed points, which connect points with triangles (Carter, 1988). Because TIN are complicated, simple relief shading often prefer the grids storage format instead of TINS (Kratt et al., 2010). With the ArcGIS 10.5 software, the shaded relief image was prepared from DEM whose spatial resolution (30 by 30) m for lineament extraction (Figure 7). The shaded relief images extracted from DEM for the study, has an illumination direction (sun -azimuth) of 315° 0′ 0′′ and a solar elevation of 45° 0′ 0′′ for automatic lineament extraction which act as channels for hydrocarbon oil and gas seepage.  (Solomon, 2003). In the study, the fault traces were delineated from the ASTER Digital Elevation Model (ASTER DEM) and then grouped into four categories as no traces, minor traces, moderate traces and major traces of fault (Table 5). Zones with major traces highlight areas of high lineament density, which are highly prerequisite for hydrocarbon seepage flow. Zones of moderate fault traces depicts areas with moderate lineament density while areas with minor or No traces of fault indicate location with minimal or possibly No lineament features. Areas with No traces of faults, are locations which are not prerequisite for hydrocarbon seepage. Table 5 shows the statistical analysis of the surface traces of faults in Ugwueme. By observation, we see that zones with major traces of fault occupies an area of 119325 m 2 which is represented by 17.64%. Zones with moderate traces of fault account for 335638 m 2 which yield 49.63%. Zones of minor traces and No traces of faults have an aerial coverage of 156235 m 2 and 65132 m 2 accounting for 23.1% and 9.63% respectively.

Geology
Geology plays a vital role in the occurrence of hydrocarbon oil and gas seepage (Solomon, 2003). The geology of an area influences the ways by which lineaments can be mapped and interpreted in an areas associated with hydrocarbon seepage. Different groups of rocks such as Nkporo, Lower coal, false bedded sandston, Awgu -Nde aboh shale, Eze Aku Shale and Asu river have been identified and are elucidate in the geological map covering the study area (Nigeria Geological Survey Agency [NGSA], 2010). The description of the minerals present in each group are Nkporo Shale (Shale and mudstone), Lower Coal (coal, sandstone and shale), false bedded sandstone (false bedded sandstones, coal and shale), Awgu -Nde aboh shale (Shale and limestone), Eze Aku Shale (black shale and siltstone) and Asu river (shale and limestone). Table 6 depicts the classification of lineament frequency and lineament length in each geological feature in the study area. By the classification, Nkporo Shale highlight lineament frequency as 154 Hz which represent 16.26% and lineament density as 157 km which represent 16.60%. Lower Coal depict lineament frequency as 710 Hz, which account for 74.97% and lineament density as 712 km for 75.26%. False bedded sandstone are represented with a total frequency of 7 Hz accounting for 8.34% and a total lineament density of 70 km accounting for 7.4%. Awgu -Nde aboh shale and Eze Aku Shale groups exhibit lineament frequency each for 1 Hz representing 0.11% as well as lineament density, each for 2 Hz representing 0.21% respectively. The Asu river group represent lineament frequency as 2 Hz accounting for 0.21% and lineament density of 3 km accounting for 3%. The overall total lineament frequency of 947 km and total lineament length of 946 km have been extracted from the satellite data for the study. Figures 9 and 10 depicts the lineament length map and lineament frequency map of Ugwueme in each geological feature. The length of lineaments within the study were measured and plotted into grid cell for preparation of the lineament length map. By carefully observing the lineament length map (Figure 9) and the lineament frequency map (Figure 10), we see that the extracted lineaments, are more concentrated in the North-Southern part of the study area than other areas. This observation shows that the North-Southern part of Ugwueme are prerequisite and act as most conduit zones for hydrocarbon seepage.

Lineament density
The aim of the lineament density analysis is to calculate the frequency of lineament per unit area (Magesh et al., 2012;Yeh et al., 2016). The analysis is used to produce a map which shows the concentration of the lineaments within the grid cell in the study area. In the study, lineament density map was produced with the spatial analyst tool which is elucidated in the ARCGIS 10.5 software by counting the lines per unit area, and then plot them in the grid center with the same tool. The line designated within the neighborhood is valid during density calculation. If the line falls outside the neighborhood for a particular cell, it is labeled no data. Table 7 depicts the lineament density classification categorized from very high to high, moderate and from very low to low lineament density for hydrocarbon seepage. The classification result shows that areas with very high lineament density represent 1774 m 2 which account for 0.26%, areas with high lineament density depicts 19476 m 2 accounting for 2.82%, areas with moderate lineament density highlight 105132 m 2 yielding 15.18%, areas with low and low lineament densities account for 198175 m 2 and 368137 m 2 yielding 28.61% and 53.15% respectively. Lineament density map of all -round directions is produced and depicted with grids ( Figure 11). The Lineament density map indicate that high lineament density is

Lineament intersection density
Lineament intersection density is represented as the total number of intersection lineament, estimated per unit area (Yeh et al., 2016). In this study, the lineament intersection density was observed by plotting the intersection of two or more lineaments, considering it as points and then contour the points to produce different degree of fracture. Table 8 depicts the classification of lineament intersection for the study. The classification result was categorized from very low to low, moderate and from high to very high lineament intersection density. Areas with very high lineament intersection density were categorized as 1781 m 2 representing 0.26%, areas with high lineament intersection density accounted for 19325 m 2 yielding 2.79%, areas with moderate lineament intersection density produce 85132 m 2 resulting in 12.29%, while areas with low and very low lineament intersection density accounted for 141318 m 2 and 445138 m 2 representing 20.40% and 64.26% respectively.  Figure 12 depict the lineament intersection density map per unit area for the study. The lineament intersection map exposes the hidden underlying configuration, which appears as linear intersection features (Yeh et al., 2016). The aim of the intersection density map is to determine the areas which are of diverse fracture orientation. The intersection density map is shown as a plain map, if the fractures don't intersect within the area. By observing the lineament intersection map (Figure 12), we see that lineament intersections are more concentrated in the North-South region of the study area. Towards the eastern part of the study area, the lineament intersection is very low to low, and in some cases, it does not exist. Area with very high to high lineament density intersection are most conduit zones for hydrocarbon seepage.

Orientation analysis
Lineament orientation are often determined with Rose diagrams (Yeh et al., 2016). The Rose diagram tool which is elucidated in the Rose work software is often used to prepare the Rose diagram ( Figure 13). The Rose diagram in the study depicts the directional frequency of the automatic extracted lineament over the investigated area. These diagram highlight two important trends for directional orientation. These trends are the major trend, explicit in the (NE-SW), (N-S) and (NW-SE) directions, and the minor trends indicated in the (W-E) direction. The directional trends of the Rose diagram are utilized for testing the distribution of faults, folds and fractures, which are expressed as lineament features and which act as conduits for hydrocarbon seepage. Table 9 depicts the statistical summary of the extracted lineament for the study area.

Conclusions and discussions
The study shows that Remote Sensing and GIS are cost effective tools for mapping geological features such as lineaments, which include faults, fractures and folds. Hydrocarbon oil and gas which escape from the earth's impermeable reservoir often flow at high pressure along zones of faults and fractures to form seepages. In this study, remote sensing was integrated with GIS for identification and extraction of lineaments in relations to natural hydrocarbon seepage in Ugwueme, South-Eastern Nigeria. Lineaments delineated from remotely sensed data were used for preparation of the fault zone map, Rose diagram, lineament density map and lineament intersection maps. These maps exposes the subsurface configuration which exist in the form of linear features and which act as conduits for hydrocarbon seepage. The Rose diagram depicts the NE-SW, N-S and NW-SE major trend orientation and the W-E minor trend orientation. These directional trends are the determining factors for testing the distribution of lineaments, which are channels for the flow of hydrocarbon oil and gas. The lineament density map and the lineament intersection map were classified as very high, high, moderate, low and very low zones for hydrocarbon seepage. Zones of very high to high lineament density were diagnosed as areas of high degree of fractures and folds, which are prerequisite for hydrocarbon seepage.

Disclosure statement
The authors declare that no conflict of interest exist.