<![CDATA[Chlorite-bearing sandstone reservoirs often pose challenges in hydrocarbon zone interpretation due to their dual effects: porosity preservation and pore-throat blockage at depth. For a better understanding, this study integrates multiple methodologies, including mineralogical analysis of core samples and petrophysical logging, to classify and to assess the influence of high chlorite content on reservoir quality; a comprehensive well log interpretation approaches and machine learning (ML) to water saturation estimation in chlorite-affected intervals. Based on porosity and permeability measurements, thin-section analysis, X-ray diffraction (XRD), scanning electron microscopy (SEM), as well as special core analysis (SCAL), the morphologies of chlorite are mainly grain-coating, rosetted boxwork, and floccules. Furthermore, the reservoirs are classified into four groups according to chlorite impact levels. Classes 1 and 2 exhibit minimal chlorite influence, whereas classes 3 and 4 show a significant chlorite effect. Notably, class 4 demonstrates lower permeability than class 3, despite higher porosity and a stronger impact of chlorite due to pore-throat blockage by abundant chlorite floccules. Water saturation analysis reveals that in high-chlorite zones, the dual-water model provides more accurate results than Archie’s equation, whereas in low-chlorite intervals, both methods yield comparable outcomes. Additionally, ML-based predictions using the state-of-the-art LightGBM tool on well-log data exhibit strong alignment with core analysis and empirical calculations, with R²= 0.8804 and RMSE = 0.0572. These findings significantly enhance water saturation estimation in chlorite-impacted reservoirs and demonstrate broad applicability for assessing chlorite distribution in wells with limited core data or well log datasets.]]>
Copyrights © 2026
