The release of methane from the seafloor into the water column within the hydrate stability field (HSF) is a natural and widely observed process. The subsequent bubble dissolution rate determines how far upwards gas is transported and the vertical distribution of this methane source to the water column. Understanding this process is essential to describing natural deep sea seeps, to assess the hazard potential from blowouts in offshore drilling activities for gas and oil, and to refine past and future scenarios of global change involving large-scale destabilization of gas hydrates and free methane gas. We report on in situ experiments on single methane and argon bubbles within and above the HSF for depths from 400 to 1500 m. Single bubbles were injected from the ROV Ventana into an attached, back-illuminated, flow-through imaging box. The ascent of individual bubbles within the imaging box was recorded and analyzed for rise velocity, V B , and radius shrinkage rate, dr/ dt. For all bubbles V B = ~ 30 cms − 1 . Methane bubbles released within the HSF had markedly enhanced lifetimes, attributed to hydrate skin formation. dr/ dt varied from − 7.5 μm s − 1 above the HSF to – 1 μm s − 1 at 1500 m, well within the HSF. Bubble longevity within the HSF increased with distance into the P/T-space from the hydrate phase boundary. There was a delay prior to the slower dissolution, where dr/ dt for methane bubbles was comparable to dr/ dt above the HSF, which was interpreted as a time delay before the onset of hydrate skin formation. Although variable, the onset time generally decreased with distance from the hydrate phase boundary. No delay was observed for the deepest releases. We relate these findings to formal calculations of methane solubility and density as a function of pressure. Pressure-dependent deviations from ideal gas law and Henry's law were implemented in a numerical bubble-propagation model which incorporated the effects of decreased solubility and surface mobility after hydrate formation. Inclusion of these effects greatly improved model prediction of observed methane bubble behavior within the HSF. The effect of these individual depth-related effects on bubble dissolution rate and longevity then was assessed quantitatively, and assumptions of different gas solubility prior and after hydrate nucleation or an effect of hydrates at the interface on the hydrodynamics at the surface were tested. Here, our approach implements our current knowledge on physicochemical and hydrodynamic control and does not seek the best fit for the given data sets, thus it also reveals current uncertainties in methane bubble processes in the HSF.