In chapter 4, I linked seismic interval velocities with physical rock models in order to estimate the hydrate and gas saturations across the BSR. The use of acoustic velocity information alone enabled me to estimate hydrate saturations for different micromechanical models of hydrate deposition in the pores. However, it cannot be used by itself to differentiate between the models and, ultimately, the obtained differences in saturation. Consequently, I use additional amplitude information, in particular AVO analysis, in an attempt to differentiate between the proposed rock-physics models of internal hydrate deposition. This analysis is performed by simple 1-D modeling. Based on given saturations, which were calculated in chapter 4, and assuming a porosity of 45%, I calculate elastic velocities and densities. Using Kirchhoff modeling, I generate synthetic seismograms and investigate the BSR AVO response obtained from the different models. This investigation shows that seismic amplitudes cannot differentiate between hydrate being part of the fluid (model A) and hydrate becoming part of the frame (model B), but give a distinctly different AVO trend in case of hydrate cementing the grains (model C). The actual seismic amplitudes can be qualitatively reproduced by either model A or B. These models cause the sediment structure at the Blake Outer Ridge to be weak and might cause the hydrate to clog large pore space conduits, since both models A and B are less permeable than model C. This could explain why free gas is trapped underneath the BSR: the hydrate presents an impermeable barrier.