We present an Electrical Resistivity Tomography (ERT) survey conducted near Larsen sheet pile walls along the Wał Zawadowski embankment in Wilanów, Warsaw. The survey profiles are 120 m long and the depth of investigation differs depending on the applied measurement protocol, reaching approximately 26 m for the gradient xl array and up to about 30 m for the dipole-dipole array. By inverting the apparent resistivity data, we identified a three-layer sub-surface structure (sands; water-saturated sands with clay and silt; muds). The field-scale tests show that the gradient protocol produced an RMS misfit below 5 %, whereas the dipole-dipole protocol yielded a misfit of about 15 %, mainly due to noise from buried utilities. This work allows direct comparison of the same geological profile under identical site conditions by two types of geophysical measurements. We argue that such comparisons are rarely reported for urban floodplain environments, yet they can provide valuable guidance for planning ERT investigations where drilling and sounding access is limited and their locations must be selected carefully. The presented approach supports and optimises decision-making processes in the design, monitoring and maintenance of hydraulic and geotechnical structures.