The Pre-Tertiary play has emerged as a promising exploration target for rejuvenation within the western flank of the Malay Basin, offshore Terengganu, Peninsular Malaysia. One of the primary challenges associated with this interval is the ambiguous subsurface stratigraphy and the poorly defined seismic reflectors, which hinder accurate geological interpretation. To address these limitations, further subsurface characterization is essential, particularly to improve the representation of fracture geometries, their evolution patterns, and to support volumetric estimations. Given that the closest geological analogue to this play is located within the Eastern Belt of Peninsular Malaysia, we conducted acquisition of Unmanned Aerial Vehicle (UAV) Light Detection and Ranging (LiDAR) data at several location from the northern to southern extents of the belt.

This initiative aims to enhance the resolution and understanding of the structural framework, thereby supporting the evaluation of the Pre-Tertiary play. The study encompasses several objectives, including the application of advanced geospatial technologies and the advance of technical methodologies. The primary focus, however, is to facilitate fracture reservoir analysis by validating geologically plausible fracture systems and gaining insights into their characteristic. This paper presents the setup of fracture modelling parameters prior to executing fracture simulations. It also details the acquisition strategies and interpretation workflows for UAV LiDAR, close range photogrammetry, UAV multispectral and ground spectral sampling, along with a comparative analysis against field-derived data to assess uncertainties. The findings underline the limitations and uncertainties encountered throughout the project, providing a critical evaluation of the accuracy and reliability of the methodologies employed.

The study area is in Kapas Island, approximately 5 km off the coast of Marang. The island lies within the Eastern Belt of Peninsular Malaysia. It is predominantly underlain by strongly deformed Permo-Carboniferous metasedimentary rock and unconformably overlying mildly deformed conglomerate sequences (Shuib, 2004). The metasediments are interpreted to have been deposited in a shallow marine to deltaic environment, while the Kapas Conglomerate likely formed in a fluvial or alluvial fan setting (Kamal Roslan et al., 1999; Shuib, 2004). Although no fossils have been identified on the island, regional lithostratigraphic correlation suggests a Permo-Carboniferous age for the metasedimentary sequence (Shuib, 2003). Structurally, the island exhibits multiple deformation phases, including tight folding and faulting, attributed to transpressional tectonics. These processes have generated extensive fracture networks, enhancing the secondary porosity and permeability of the host rocks. Such characteristics warrant further investigation into their potential as fractured hydrocarbon reservoirs.

In general, a very detailed and precise elevation model couple with high resolution true colored and multispectral images are required in the study. UAV or drone was used as main survey equipment in this project due to its flexibility to map terrain and outcrop in proximity with precise measurement and in high resolution. The analysis commenced with a review of the processed datasets. A surface differential model, generated by subtracting the DTM from the DSM, was combined with hydrological flow pattern data to delineate potential fracture zones. This approach enabled the identification of structurally controlled drainage patterns, particularly rectangular geometries, which are indicative of subsurface faulting and jointing.

To further refine the structural interpretation, close-range photogrammetry was employed to visualize and quantify fracture orientations, including dip and dip azimuth. This dataset also facilitated the estimation of fracture length and, with supporting field observations, fracture height. Based on empirical measurements, a height-to-length ratio of 1:2 was adopted for this study. However, fracture aperture could not be reliably measured due to limitations in the resolution of the photogrammetry dataset. In future survey, a higher camera resolution (eg. 45 MP) can be deployed to provide better pixel resolution thus improving texture generated in 3D mesh model.

The integration of UAV LiDAR, close-range photogrammetry, and multispectral datasets has significantly improved structural interpretation on Pulau Kapas. The surface differential model and flow pattern analysis enabled confident identification of fracture zones, while photogrammetry provided accurate bedding orientation and fracture geometry estimates. Although multispectral data was limited in fracture detection due to spectral homogeneity, it supported bedding orientation analysis. Overall, the multi-source geospatial approach has proven effective in characterizing the structural framework of the study.