This paper describes pilot-scale studies that examined three integrity test methods for: (1) quantifying virus removal by nanofiltration (NF) and reverse osmosis (RO) membrane systems when arranged in single element unit and two-stage system configurations, and (2) determining change in virus removal capability of such systems when subject to different types of membrane/o-ring compromisation and fouling. The three methods evaluated included one biologic type (MS-2 phage), that has been employed previously; and two, new non-biologic types (24-nanometre polystyrene fluorescent dyed microspheres and fluorescent Rhodamine WT [R-WT] dye, molecular mass 496 daltons). All three surrogates were employed in a manner intended to show a minimum of 4-logs removal by the NF and RO membranes selected for test. Methods of compromisation included a pinhole induced through one membrane leaf in the spiral wound NF/RO element, and both cracking of and removal of sections from one of the permeate tube o-rings. Testing was conducted on two source waters, representing brackish surface water and effluent categories: a microfiltered secondary effluent and a river water. The river water is characterized by low to moderate TDS and high TOC and was treated with conventional alum coagulation, flocculation, sedimentation and granular media filtration for subsequent membrane processing.