To systematically examine the paradigm shift of sedimentary paleogeography research driven by artificial intelligence, and clarify the technical integration pathways and core challenges, this study combines theoretical analysis and typical case analysis to summarize the multi-dimensional characteristics of the paradigm shift and explore a new framework for intelligent paleogeographic reconstruction. The results show that the paradigm of sedimentary paleogeography research is undergoing a systematic transformation across four dimensions: the data foundation shifts from experience-driven to data-driven, the cognitive mode from qualitative description to quantitative characterization, the prediction target from static representation to dynamic prediction, and the decision-making mode from single-discipline analysis to systematic decision-making. Accordingly, a comparative framework of traditional empirical, digital, and intelligent paradigms is constructed. In three key aspects, i.e. paleogeographic reconstruction, sedimentary process simulation, and reservoir parameter prediction, the methods such as knowledge graphs, physics-informed neural networks, and ensemble learning have demonstrated significant integration effectiveness. Furthermore, a technical workflow framework for intelligent paleogeographic reconstruction is proposed, encompassing core modules including multi-source data fusion, knowledge graph constraints, machine learning prediction, expert validation, and feedback iteration. This work provides a systematic theoretical reference and technical pathway for the digital-intelligence transformation of sedimentary paleogeography.

Based on the core, cast thin section, CT, loggings, test and seismic data, the sedimentary-diagenetic characteristics and controls on favorable reservoirs in semi-restricted carbonate ramp setting were elucidated, through a case study of the Lower Cretaceous Yamama Formation in Oilfield A of the Central Arabian Basin. During the Early Cretaceous, the study area was a carbonate ramp in semi-restricted environment, where low- to medium-energy shallow-water lithofacies were common, and the depositional facies were dominated by large-scale lagoon, locally with grain shoal, point reef, back shoal and tidal flat. Bioclastics were diverse, with algae, benthic foraminifera, bivalve, bacinella, and peloids being the most abundant. The Yamama Formation in the study area underwent intense diagenesis during the penecontemporaneous stage, with cementation and dissolution coupled to control the formation and preservation of secondary pores. The reservoirs in the Yamama Formation are composed of packstone, wackstone and bindstone, indicative of frequently varying lithology with poor lateral correlatability. The reservoirs are porous, dominated by micropores, moldic pores, and skeletal pores, with a low abundance of primary intergranular pores, and the pore throats dominated by medium- and micro-throats. The physical properties generally exhibit low to medium porosity, and low to ultra-low permeability. The medium-high permeability reservoirs are underdeveloped. Favorable reservoirs in the Yamama Formation are controlled by local high-energy sedimentation, soluble bioclastic enrichment, intense dissolution, and abnormal-high pressure. Local high-energy grain shoals contain well-preserved primary intergranular pores with no intense cementation, forming small-scale favorable reservoirs. In contrast, low- to medium-energy facies such as lagoon and back shoal are locally rich in soluble bioclastics such as algae and bacinella. The bioclastics were intensely dissolved, forming a large number of moldic or skeletal pores, which effectively improved the reservoir physical properties, thus facilitating the formation of large-scale favorable reservoirs. The favorable reservoirs of Yamama Formation are mainly discovered in YA and YB sections, and large-scale reservoirs thereof are located in the central-northern part of the study area. These represent key targets for subsequent exploration and development.