Understanding the processes and phases of sediment transport and drift formation is essential for a reconstruction of their evolution. Our approach to evaluate the contribution of different processes to drift growth, especially the alongslope vs. downslope transport component, is numerical modelling.For this work, we have selected the area of ODP Leg 178 on the western side of the Antarctic Peninsula, where a dense grid (~4000 km) of seismic profiles and numerous cores document the presence of 8 sediment mounds along the continental slope. These mounds are interpreted to be drifts, though a contrasting model proposes a turbidity-current-dominated origin. We concentrate on Drift 7, which was covered by ~1500 km of seismic reflection lines, four CTD sites and two ODP Leg 178 sites (1095 and 1096). Numerical modelling of the processes involved could favour one of the proposed models and answer the question regarding mound origin.The forward modelling process needs a 3-D framework of seismostratigraphic and sedimentological input parameters to define the input data and boundary conditions of the numerical model. These include: (a) boundaries of seismostratigraphic units or event horizons (e.g. events of erosion or non-deposition), (b) seismostratigraphic unit thicknesses, (c) granulometric changes i.e. grain-size distributions within seismostratigraphic units (d) physical properties of sediments at event horizons and, (e) lithological changes.In general, we aim at the inversion of an observed sediment structure into the depositional process. I.e. from the observation we want to deduce information on the generating current. This information comprises of current (steady and tidal) velocity and direction, the epoch when the current was active and its duration. This will lead to a better understanding of the processes responsible for sediment transport and erosion by oceanic currents. The problems we want to solve can be summarised as follows:1. Is an initial topography (large-scale or mesoscale topography or inhomogeneity in sediment distribution) necessary for the formation of a structure like a sediment drift?6. What current velocities are needed to a) transport and b) deposit sediment?7. Does the current velocity need to be maintained for the whole period or is the shaping of the sediment drift favoured by episodic changes in current velocity?4. How long would an episode of current activity have to be in order to generate the morphology observed in a sediment drift?Preliminary work on the high-resolution seismic reflection data and granulometric work within the seismicstratigraphic units has shown clear variability in the grain size that could be correlated, if only tentatively at this time, to know changes in the development of the Antarctic ice sheet in response to climatic change. Further analyses and correlation between the geological and geophysical data form work in progress.