||The COSTA target areas exhibit different slope failure events, which reflect different triggering mechanisms including gas hydrates dissociation. Gas hydrates stability law depends on temperature, pore pressure, gas chemistry, and pore water salinity. Any change in the equilibrium parameters may convert the hydrate to gas plus water, causing significant weakening of the sediment, and generating a rise of pore pressure. A significant consequence of the hydrate melting is the gas production, which alters significantly the behaviour and mechanical properties of the marine sediments and could be very hazardous when the sediment is unloaded in undrained conditions. The behaviour of the unsaturated marine sediment has to be accurately identified in order to quantify the geological risks associated to the hydrate dissociation [see, for instance, Sultan, N., Cochonat, P., Foucher, J.P., Mienert, J., this volume. Effect of gas hydrates melting on seafloor slope instability]. Thus, in this work, five tests were carried out on synthetic marine sediments in order to identify the gas effect on the hydro-mechanical behaviour of unsaturated marine sediments. Based on the experimental results obtained from this study, on previous constitutive models [Alonso, E.E., Gens, A., Josa A., 1990. A constitutive model for partially saturated soils. Geotechnique 40(3), 405-430; Gens, A., Alonso, E.E., 1992. A framework for the behaviour of unsaturated expansive clays. Can. Geotech. J. 29, 1013-1032] and on the relationship of Van Genuchten [Van Genuchten, M.Th., 1980. A closed-form equation for predicting the hydraulic conductivity of unsaturated soils. Soil Sci. Sue. Am. J. 44, 992-898], a new mathematical formulation was proposed to simulate the behaviour of the unsaturated soil. Qualitative comparison between experimental results and model predictions shows the capacity of the proposed model to reasonably reproduce the essential features of the hydro-mechanical behaviour of marine unsaturated sediment. Experimental and simulation results show that unloading (tidal cycles, erosion, natural slope instabilities, excavation) a submarine unsaturated sediment slope in shallow water is much more hazardous than unloading the same submarine slope in deep water.