Experiments seepage reducing measures canal bed : column tests and CST permeability tests

This report is a product within sub-project 3.1 Measures of SITO-PS study “Measures to Reduce Seepage Issues Along Canals (ARK)”, which uses the Amsterdam-Rhine Canal (ARK in Dutch) between Breukelen and Nigtevecht as a case study. The surrounding area experiences significant seepage and other excess water problems, linked to soil rupture and groundwater boils in the polders along the canal. This report describes an experimental investigation into "soft solutions" aimed at reducing water seepage along the Amsterdam-Rhine Canal near Nigtevecht. The research evaluates the effectiveness and durability of various sealing materials in reducing the permeability of the canal bed under dynamic conditions. The testing followed a two-phase approach: • Capillary Suction Time (CST) tests: Small-scale static tests used to characterize the intrinsic permeability of material "recipes" and determine optimal sand-to-additive ratios. • Column tests with dynamic loading: Larger-scale tests (1.5 m pressure head) utilizing a rotating vane to mimic the hydrodynamic shear stresses (0.6 to 2.4 m/s) induced by passing ships. Five primary materials were evaluated in various mixtures and configurations: RONA ZBM (sand-bentonite mixture), OCMA bentonite, S1 clay, natural dredged mud, and granulite (Greenbase). The highest vane speed of 2.4 m/s in the column tests is 1.7x higher than the highest measured ship-induced flow velocity, but the corresponding bed shear stress from the vane corresponds to the highest measured shear stress in the 9-day field campaign in ARK (Deltares 2025a). Given the limited measurement period of 9 days in the field campaign it is possible that incidentally higher ship-induced flow speeds occur in the ARK. The occurrence of the high shear stresses in the vane protocol is more frequent than in the field measurements. Hence, the applied vane protocol is considered to be representative, but on the conservative side for ARK conditions. The experimental investigations revealed a distinct performance hierarchy where sodium activated bentonite outperforms clay, mud, and granulite due to active swelling properties that require significantly less material mass for effective sealing. When subjected to dynamic ship induced shear, these materials typically followed a three-phase behavioral pattern involving initial consolidation, stepwise erosion during peak 2.4 m/s flow events, and subsequent stabilization. Although the hydrodynamic stress often erodes away the bulk of the initial capping layer, the system's long-term sealing capacity was maintained by a thin, robust "effective layer" that formed at the base interface and was continuously reinforced by the "self-healing" deposition of suspended fines. The durability of this seal was much improved by protective configurations; specifically, "sandwiching" pure clay or mud beneath a 30 cm sand layer increased resistance times to 80–90 days compared to just ~10 days for unprotected ZBM (for ZBM the 30 cm sand cover provided temporary improvement, but not towards the end of the test at which moment the resistance time also was ~10 days – similar to unprotected ZBM), while a gravel armor layer on top of ZBM yielded the highest observed resistance of ~221 days. For practical application in the ARK, these findings underscore the advantage of applying a protective overburden (sand or gravel) to dampen forces and ensuring a "resting time" to allow consolidation before exposure to heavy shipping traffic starts. The report recommends further research into placement procedures (e.g., jetting or flume tests), the influence of background currents, and the impact of different sand cover thicknesses. A field pilot in the ARK is strongly advised to monitor in-situ erosion, deposition, and spatial redistribution of these materials over time.