Geotechnical performance of a two-row grout curtain in karstified limestone–marl: insights from stage-grouting and clay-filled discontinuities at Chamshir RCC Dam

Curtain grouting is widely applied for seepage control in dam foundations; however, quantitative performance assessment in heterogeneous karstified formations remains methodologically challenging. This study presents an integrated evaluation of a two-row grout curtain constructed in the karstified limestone–marl foundation of the 151-m-high Chamshir roller-compacted concrete (RCC) gravity dam in Iran. The methodology combines Lugeon permeability testing, cement take analysis, real-time monitoring of descending (downstage) stage-grouting sequences, borehole deviation surveys (conducted on less than 9% of holes), and control borehole verification with pre-grouting and post-grouting comparisons. Rather than viewing split-spacing merely as a geometric layout, we analyze the physical governing mechanism: the progressive, sequential clogging of fracture networks under systematically increasing pressures—moving from wider, low-resistance pathways in primary stages to narrower fractures—which successfully reduced the average cement take by 80% (from 155 kg/m in primary holes to 31 kg/m in quaternary holes). Crucially, stage-by-stage geotechnical and hydraulic feedback captured during the sequential segmenting of boreholes provided vital real-time data to dynamically calibrate injection pressures and adaptively refine grout mix rheology during construction. Notably, several test intervals exhibited zero or low Lugeon permeability (< 1 Lu) but unexpectedly high cement take, reaching 254 kg/m at 45–50 m depth in borehole P7. This inverse relationship is attributed to clay-filled discontinuities that displace under grouting pressure, creating artificial permeability that clean-water testing fails to detect. Post-grouting control boreholes confirmed that permeability was successfully reduced to below 3 Lu across most of the treated zone. However, deviation surveys revealed that 13% of the tested holes (1 out of 8) exceeded the 2% acceptance criterion, with a maximum deviation of 4.18% at 62 m depth, highlighting critical quality assurance gaps in complex geological settings. This study demonstrates that integrating multi-disciplinary datasets—linking structural geology, stage-grouting feedback, and deviation monitoring—provides a far more reliable framework for curtain evaluation than conventional Lugeon testing alone.