A simulation-based study concerning three non-destructive approaches by which water in spent nuclear fuel assemblies might be quantified with neutrons is described. Three fuel types have been considered: thirty-six spent Advanced Gas-cooled Reactor (AGR) fuel pins contained in a stainless-steel can; a prototype fast reactor (PFR) spent fuel assembly and a light water reactor (LWR) spent fuel assembly – with the PFR and LWR assemblies containing mixed oxide fuels. The three approaches are investigated with MCNP5 based on neutron interrogation: fast neutron detection of the perturbed fast neutron flux and two approaches based on the 2.223 MeV γ-ray photon emission resulting from the 1H(n,γ)2H capture reaction, stimulated with moderated and unmoderated neutrons. For each of these approaches the following perspectives have been considered: neutron transmission, the influence of radiation emitted by the associated spent fuel inventory, and the dependencies on water quantity and location. The moderated γ-ray assay technique is observed to return the most significant distinction between scenarios with and without water in the fuel assemblies, notwithstanding the need for efficiency response functions to be overlaid on the neutron models as part of the interpretation of the MCNP model results. Water present in all of the spent fuels considered is observed to perturb the incident neutron flux to a degree significant within the simulation uncertainties. The AGR and PFR cases are not undermined by the intrinsic neutron field from the fuel; whereas for the LWR-MOX case, the simulations suggest that this will limit its potential. Successful detection of water quantities down to 10 g is anticipated for AGR, whereas reduced mass sensitivity is forecast for PFR and LWR-MOX, where it is assumed that the water will be more dispersed. Water located in fewer locations is observed to be harder to discern, suggesting that small, localised quantities of water could pose a challenge for these techniques.