Background
The authors note that mass balance estimates of the Antarctic Ice Sheet (AIS) in the 2007 report of the Intergovernmental Panel on Climate Change, as well as estimates in more recent reports, lie between +50 and -250 Gt/year for the period 1992-2009, which range is approximately 15% of the annual mass input to the AIS and 0.8 mm/year sea level equivalent (SLE). Providing a bit more detail, they additionally note that two estimates (+28 and -31 Gt/year) from radar altimeter measurements made by European Remote-sensing Satellites (ERS) lie in the upper part of the 1992-2009 range, while estimates from the Input-minus-Output Method (IOM) and the Gravity Recovery and Climate Experiment (GRACE) lie in the lower part (-40 to -246 Gt/year) of that range.

What was done
Striving to determine which set of observations is the most correct, Zwally and Giovinetto, as they describe it, "compare the various estimates, discuss the methodology used, and critically assess the results," while they also "modify the IOM estimate using (1) an alternate extrapolation to estimate the discharge from the non-observed 15% of the periphery, and (2) substitution of input from a field data compilation for input from an atmospheric model in 6% of the area."

What was learned
The two U.S. researchers from the Cryospheric Sciences Branch of NASA's Goddard Space Flight Center found that "the modified IOM and a GRACE-based estimate for observations within 1992-2005 lie in a narrowed range of +27 to -40 Gt/year, which is about 3% of the annual mass input and only 0.2 mm/year SLE." And they indicate that their preferred estimate for 1992-2001 is -47 Gt/year for West Antarctica, +16 Gt/year for East Antarctica, and -31 Gt/year overall.

What it means
In concluding their analysis, Zwally and Giovinetto state that "although recent reports of large and increasing rates of mass loss with time from GRACE-based studies cite agreement with IOM results," they say that their evaluation "does not support that conclusion." And with the great potential for extremely large errors in GRACE-based studies, which typically suggest overly large ice sheet losses -- see our reviews of Ramillien et al. (2006), Velicogna and Whar (2006) and Quinn and Ponte (2010) -- it would appear that the result of Zwally and Giovinetto's analysis is probably as close to whatever the truth happens to be in this case than any of the analyses that rely heavily on GRACE data.

References
Quinn, K.J. and Ponte, R.M. 2010. Uncertainty in ocean mass trends from GRACE. Geophysical Journal International 181: 762-768.

Ramillien, G., Lombard, A., Cazenave, A., Ivins, E.R., Llubes, M., Remy, F. and Biancale, R. 2006. Interannual variations of the mass balance of the Antarctica and Greenland ice sheets from GRACE. Global and Planetary Change 53: 198-208.

Velicogna, I. and Wahr, J. 2006. Measurements of time-variable gravity show mass loss in Antarctica. Science 311: 1754-1756.