Smart water flooding is a promising eco-friendly method for enhancing oil recovery in carbonate reservoirs. The optimal salinity and ionic composition of the injected water play a critical role in the success of this method. This study advances the field by employing machine learning and data analytics to streamline the determination of these critical parameters, which are traditionally reliant on time-intensive laboratory work. The primary objectives are to utilize data analytics to examine how smart water flooding influences wettability modification, identify key parameter ranges that notably alter the contact angle, and formulate guidelines and screening criteria for successful lab design. This design aims to shift rock conditions from oil-wet to water-wet by choosing the optimal salinity and ionic concentrations for smart water flooding. Analyzing a comprehensive dataset of 484 data points from 48 studies, primarily focused on calcium carbonate (72.7%), We assessed properties of rocks and oils, injection brine characteristics, and contact angles before and after smart water injection from 2010 to 2023. Our findings highlight a strong correlation between the initial contact angle and the change post-smart water flooding, indicating the critical role of the rock's initial wettability. We identified that high concentrations of sodium, calcium, and chloride ions adversely affect wettability alteration, while sulfate ions contribute positively. Furthermore, outliers were removed, and after comprehensive analysis, application guidelines were developed for each significant parameter for calcium carbonate, limestone, and dolomite rock. Our research further reveals that calcium carbonate exhibits intermediate initial wettability, whereas dolomite and limestone show more muscular initial water-wet conditions. Smart water ionic composition analysis revealed distinct responses in calcium carbonate compared to dolomite and limestone, underlining the importance of tailoring smart water composition to specific rock types. The analysis highlights that oils with lower api gravity, acid numbers, and viscosities exhibit more excellent responsiveness in altering wettability. Overall, this study significantly advances smart water flooding in carbonate reservoirs, offering a framework for optimizing eor techniques by salinity and ionic composition. Additionally, it establishes a screening criterion for optimal salinity and ionic ranges, potentially reducing time-consuming experiments.


Geosciences and Geological and Petroleum Engineering

Second Department

Chemical and Biochemical Engineering

Keywords and Phrases

Carbonate reservoirs; Ionic composition; Machine learning; Optimal salinity; Smart water flooding

International Standard Book Number (ISBN)


Document Type

Article - Conference proceedings

Document Version


File Type





© 2024 Society of Petroleum Engineers, All rights reserved.

Publication Date

01 Jan 2024