Ocean-forced melt of ice shelves is key to the stability of the Antarctic Ice Sheet (AIS); however, most Earth system models (ESMs) do not simulate ocean circulation in the cavities beneath ice shelves. This forces ice-sheet models to rely on ice-shelf basal melt parameterizations that use ocean temperatures external to the ice shelf (far-field temperatures) extrapolated into ice-shelf cavities. In this work, we evaluate the impact of this limitation by utilizing an ESM that includes ice-shelf cavities. We use existing simulation results from the cavity-resolving MPAS-Ocean (MPAS-O) model within the Energy Exascale Earth System Model (E3SM) v2.1 under the SSP3-7.0 greenhouse gas scenario through 2100 to force the MPAS-Albany Land Ice (MALI) ice-sheet model and perform AIS projections. We test the limitation of the standard parameterization-based approach by performing experiments in which MALI is forced by 1) ice-shelf basal melt rates simulated by MPAS-Ocean, 2) parameterized ice-shelf basal melt rates using temperatures within the full cavity domain and 3) the parameterization using far-field temperatures at the ice-shelf front (a community standard approach). Results for scenario 1 show that the choice of melt-rate extrapolation has a significant impact on AIS evolution due to differences in ice-shelf extent and geometry between the fixed-cavity MPAS-Ocean and MALI as it evolves. Due to biases in the ocean model melt rates, all experiments using these melt rates gain mass through 2100; however, when melt-rate extrapolation is not implemented, the sea-level equivalent mass gain is ~7x larger at 2100 compared to when extrapolation is turned on (see figure). Ongoing work is evaluating scenarios 2 and 3 which includes an evaluation of bias corrections to ocean temperatures. Our results provide a stepping stone towards a fully coupled ice-sheet and ocean realization and provide insight on how to manage inconsistencies in the representations of ice shelves in ice-sheet and ocean models.