© The Author(s), 2014. This article is distributed under the terms of the Creative Commons Attribution License. The definitive version was published in Cryosphere 8 (2014): 1375-1392, doi:10.5194/tc-8-1375-2014. ; In Antarctica, uncertainties in mass input and output translate directly into uncertainty in glacier mass balance and thus in sea level impact. While remotely sensed observations of ice velocity and thickness over the major outlet glaciers have improved our understanding of ice loss to the ocean, snow accumulation over the vast Antarctic interior remains largely unmeasured. Here, we show that an airborne radar system, combined with ice-core glaciochemical analysis, provide the means necessary to measure the accumulation rate at the catchment-scale along the Amundsen Sea coast of West Antarctica. We used along-track radar-derived accumulation to generate a 1985–2009 average accumulation grid that resolves moderate- to large-scale features (>25 km) over the Pine Island–Thwaites glacier drainage system. Comparisons with estimates from atmospheric models and gridded climatologies generally show our results as having less accumulation in the lower-elevation coastal zone but greater accumulation in the interior. Ice discharge, measured over discrete time intervals between 1994 and 2012, combined with our catchment-wide accumulation rates provide an 18-year mass balance history for the sector. While Thwaites Glacier lost the most ice in the mid-1990s, Pine Island Glacier's losses increased substantially by 2006, overtaking Thwaites as the largest regional contributor to sea-level rise. The trend of increasing discharge for both glaciers, however, appears to have leveled off since 2008. ; This research was supported at UW by NSF OPP grants ANT-0631973 (B Medley, I. Joughin, E. J. Steig, and H. Conway) and ANT-0424589 (B. Medley and I. Joughin). Work at WHOI was supported by NSF OPP grant ANT-0632031 and NASA grant NNX10AP09G (S. B. Das and A. S. Criscitiello). D. H. Bromwich and J. P. Nicolas were supported by NASA grant NN12XAI29G and NSF grant ANT-1049089.