Physical Modeling of Seepage Control Using Upstream Blanket and Cutoff in Earth Dams: A Hele–Shaw Experimental Study

Seepage beneath earth dams founded on pervious strata can cause excessive under-seepage, elevated downstream exit gradients, and high phreatic levels, thereby increasing susceptibility to internal erosion and piping. This study presents a Hele–Shaw laboratory investigation of seepage-control efficiency for an upstream impervious blanket used alone and in combination with a vertical cutoff (blanket–cutoff system). The experimental geometry reproduces a zoned earth dam cross-section at a scale of 1:200. Five foundation thickness ratios (T/B = 0.184–1.00) were tested. For the blanket-only system, four blanket length ratios (Lb/B = 0.50–1.25) were examined. For the blanket–cutoff system, cutoff depth ratios (S/T = 0.20–0.80) were investigated using (i) a representative blanket length Lb/B = 0.75 across all foundation depths and (ii) a deep-foundation case T/B = 1.00 across all blanket lengths. Seepage discharge, head loss due to seepage-control measures, maximum exit gradient at the downstream toe, and phreatic line location were measured at steady state and expressed in dimensionless form using the equivalent Hele–Shaw hydraulic conductivity. Relative to the no-measure reference case, the upstream blanket reduced dimensionless discharge by 20.8–70.2%, reduced the exit-gradient indicator by 6.4–50.2%, and reduced the downstream seepage-surface height by 58.9–92.8%. Adding a vertical cutoff provided further reductions relative to the blanket-only configuration, up to 34.4% in discharge and to 29.8% in exit-gradient indicator at Lb/B = 0.75—while increasing head loss across the upstream control system. Regression-based correlations and main-text design maps are proposed for preliminary sizing. The proposed correlations and design maps are intended for screening-level use only within the tested ranges 0.18 ≤ T/B ≤ 1.00, 0.50 ≤ Lb/B ≤ 1.25, and 0.20 ≤ S/T ≤ 0.80. Because the Hele–Shaw model is a two-dimensional viscous-flow analog of saturated seepage, the results provide a physical basis for relative comparison of seepage-control measures rather than a direct substitute for site-specific analysis of heterogeneous three-dimensional foundations. Accordingly, the agreement discussed in this paper is qualitative and trend-based, and the proposed tools are intended to complement rather than replace quantitative FEM for site-specific design.