BVC is located in the Turkana rift zone which is within the Kenyan Rift system. The Kenya Rift is comprised of both the Turkana Rift Zone and the Kenya Rift Zone [6] . It generally follows the boundary between the Proterozoic Mozambique Belt and the Archean Tanzania Craton [7] . The northern, central, and southern segment of the Kenya Rift have half graben geometries due to earlier faulting and pronounced subsidence along the western boundary faults [6] . For most of the structural evolution of the Kenya Rift, faulting has been controlled by an E-W oriented position of minimum stress [8] .

Volcanic activity in Turkana rift, which was followed by activity moving southwards, was the initial step in the formation of the Kenya rift during the early Miocene (14 - 23 Ma BP). Faulting that occurred in conjunction with rifting took place in many stages, beginning with faulting on the western side, which was accompanied by basaltic and phonolitic volcanism on the crust of the uplift [9] . Major faults extended along the western side, generating a half graben that was bordered on the eastern side by a monoclinic flexure. At the same time, the development of massive basaltic-trachytic shield volcanoes was taking place. On the eastern side, major faults formed, which resulted in the transformation of the half graben into a complete graben and was followed by basalt-trachyte volcanism.

Around 5 million years ago (Miocene epoch), the graben structure began to take shape. Subsequently, fissure eruptions in the axis of the rift led to the creation of flood lavas between 2 and 1 million years ago (Quaternary period). Within the axis of the rift, volcanic activity has been growing increasingly intense over the past two million years. Large Quaternary shield volcanoes, notably the Barrier Volcanic Complex, began to form along the Kenya rift axis during this time period. The majority of these volcanoes are geothermal potential areas.

Volcanism in BVC commenced with the eruption of strongly undersaturated flood basalts from a series of fissures along the eastern margin of the trough. Trachyte volcanism commenced with the eruption of lavas and pyroclastic flows from Likaiu East and Kalolenyang volcanoes in the period extending from at least 1.3 Ma to 0.7 Ma [10] . These eruptions build up steep sided edifices, the summit of which later collapsed to form craters. Small volumes of basalts were erupted at a late stage. Rejuvenation of rift margin structures faulted and tilted this center before the next phase of activity.

Likau West (0.5 - 0.2 Ma) is poorly exposed. The summit area is covered by small trachyte and phonolite lava domes and associated pyroclastic deposits. Volcanism on the BVC culminated with the formation of Kakorinya volcano (>0.2 Ma to Recent). The construction of the lava shield was accompanied by the eruption of extensive pyroclastic flow deposit and air-fall tuff on the western flanks of the volcano. Collapse of the summit area of Kakorinya took place in two stages. Two outer ring faults formed immediately after the Upper trachytes were erupted and were associated with limited basaltic activity on the summit area (Figure 3). These fractures are post-dated by trachyte lava domes and widespread air-fall tuffs, the eruption of which triggered the formation of the caldera at 92 ka [10] , as shown in Table 1.

Table 1 below shows the generalized stratigraphy of the Barrier Volcanic Complex.

The digital elevation SRTM data was downloaded from Earth Explorer USGS website. The period of the data was 01-01-2000 to 31-12-2020, a range of 20 years. The image with less atmospheric noise was selected and downloaded for processing and analysis using the WGS-84 datum, and 37N UTM zone map projection. The study area was clipped out using the BVC shapefile. The clipped image was double-blurred to smooth it for clear geological structural identity.

PCI Geomatica was used for generating major geological structures in the area. The digital elevation model of BVC was imported into the software, run it, and the results was generated in terms of a shapefile of major geological structures i.e. faults in the area. ArcGIS was used for drawing the lineaments and generating a structural density map. The lineaments generated from ArcGIS were exported into Rockworks 17 in form of file for generation of a rose diagram.

The geological structural analysis was done in three steps namely; delineation of lineaments, determination of lineaments density and finally determination of major trends of lineaments. The results of the analyses gave a clear map on the trend of structures that influence the geothermal system in the study area. The steps used are further discussed in the proceeding text.

The GPS was used for collecting coordinates of the structures mapped on-site. The Branton Compass was used for recording the strikes and dips, see Table 2. The data on the location of hot springs and fumaroles was obtained from Ol Suswa Energy Limited which has been licensed to explore the geothermal resource in BVC. The raw data on strikes and dips was input into the Georose Software and the results were the generation of a Rose diagram. Figure 13 shows the distribution and location of the mapped structures.

The total number of lineaments generated in the study area were 260; whereby at an azimuth of 45˚ a total of 60 lineaments, at 90˚ was 95 lineaments, at 150˚ was 61, and at 315˚ was 44 lineaments were generated. The lineaments generated for different Azimuths are shown in Figures 4-7. The results were compiled into one map as represented in Figure 8.

The hot springs in the southern part of the area occurring in the East-West direction are evident with results generated by PCI Geomatica but not evident with ArcGIS drawn lineaments. It is also evident that the geothermal manifestations i.e., hot springs and fumaroles (Figure 9) in the area occur along the delineated geological structures. The E-W occurring structures indicated by hot springs can be as a result of folding.

For analysis and construction of lineaments density map the lineaments shown in Figure 8 were used. From the color ramp of Figure 10, the highest density areas are represented in red while low density areas are represented in green, and lowest/no lineament areas are in white color. From the results, there is localised high concentration of lineaments in the four volcanic centers which shows high potential for occurrence of geothermal resource. However, there could be an impermeable layer that separates Kalolenyang to the western part of the area with the other three volcanic centers hence having a separate geothermal system with distinct recharge.

The results from the two previous steps were used to determine the major trends in the study area. The delineated lineaments were compiled and processed to generate a Rose diagram that shows the dominant orientation of the lineaments as shown in Figure 11.

From Figure 11, the geological structures at BVC are trending in the NNE-SSW, and N-S. These trends of geological structures might have been as a result of faulting towards the east as reported by (10). The North-South trending is associated with the North trending normal faulting at the rift margin that led to the formation of the inner trough of BVC as well basalts of Logipi.

To ensure the accuracy of the data, ground truthing was necessary for the calibration and validation of remote sensing models. This increased the precision and reliability of the analysis and result in more informed decision-making processes by cross-referencing data from remote sensing with information from the ground (See Figure 12). Ground truthing was done by measuring the strikes and dips at BVC in order to confirm the major trends (See Table 2). The raw data on strikes and dips was input into the Georose Software and the results were the generation of a Rose diagram shown in Figure 13.

This involved collection of on-site data to enhance the accuracy of data obtained remotely and was done by visiting the study area and taking measurements of strikes and dips, coordinates of the mapped structures, and general ground topography (Table 3). 12 accessible structures in the area were mapped and the result was a generation of a rose diagram shown in Figure 13. The trend of structures is in the NE-SW direction which supports the structural trends in Figure 11. This validated the reliability and legitimacy of information extracted from satellite imagery.