Ice loss from glaciers results in highly nonuniform patterns of sea level change due to the effects of self-attraction and loading. To quantify these spatial effects, it is necessary to obtain an ice load model that is both spatially realistic and regionally complete. We demonstrate a technique to produce such a model for the Alaskan glaciers by combining mass balance rates from a Gravity Recovery and Climate Experiment (GRACE) mascon solution with realistic glacier geometries. This load model can be used to solve the “sea level equation” to determine gravitationally self-consistent sea level and gravity rates. The model predicts a significant drop in relative sea level in the near field of the glaciers, with coastal rates of around −9 mm/yr (compared to a global average rise of 0.2 mm/yr) and significant differences to those predicted by a coarser model. The magnitude and sensitivity of these near-field rates imply that the near-field tide gauge records could contain significant information about the spatial distribution of ice loss. Comparison of model gravity rates with an independently produced, spherical harmonic, GRACE solution verifies that our technique can successfully capture the mass changes estimated in the mascon solution within our higher-resolution model. Finally, we use our ice load model to examine the possibility of detecting the effects of ice loss in estimates of ocean bottom pressure (OBP) from GRACE. We use the model to simulate the effects of GRACE signal leakage and show that the OBP signal from leakage has a similar pattern to, but larger amplitude than, the sea level “fingerprint” expected from ice loss.