We predict gravitationally self-consistent global geoid and relative sea level (RSL) perturbations due to present-day melting of ice complexes, including the Antarctic and Greenland ice sheets and a suite of mountain glaciers and ice sheets. Classic analyses of sea level change indicate that these perturbations will depart significantly from eustatic (i.e., geographically uniform) trends Woodward, 1888; Farrell and Clark, 1976>, although this result has not always been appreciated in modern analyses. Mass flux of individual ice reservoirs will produce unique geometries of sea level change, and this distinctiveness admits the possibility of using global geoid, sea surface, and RSL signatures of recent climate change to infer the ongoing mass balance of each reservoir rather than simply the net mass flux. As an example, we show that perturbations to the geoid arising from noneustatic water loads associated with each ice reservoir are sufficiently large (at low degrees) to be theoretically measurable within 5 years by the GRACE satellite mission. We complete the study by reanalyzing tide gauge data at 23 sites selected by Douglas <1997> in a recent analysis of global RSL rise. Traditionally, estimates of global sea level rise are generated by taking the mean of a set of secular tide gauge trends that have been corrected for the influence of ongoing glacial isostatic adjustment (GIA) related to the late Pleistocene glacial cycles. The common assumption in such studies is that the geographic scatter in the residual, GIA-corrected trends is due to errors in the GIA model or unmodeled processes (e.g., tectonics). We consider a large suite of GIA model predictions and apply a least squares approach to the GIA-corrected tide gauge trends to estimate the weighting of various present-day sea level signatures. We find that the fit to the residual RSL trends is significantly improved and that the procedure is able to resolve a long-standing observation of anomalously low sea level rates in Europe. This preliminary analysis, which is relatively insensitive to changes in the assumed geometry of the present-day mass balance, assumes that ocean thermal expansion is globally uniform. However, the procedure can be easily extended to incorporate a realistic steric contribution once the geometry of the process is sufficiently well constrained. ¿ 2001 American Geophysical Union