I asses the feasibility of multi-variate scaling relationships to estimate glacier volume from glacier inventory data. Scaling laws are calibrated against volume observations optimized for the specific purpose of estimating total global glacier ice volume. I find that adjustments for continentality and elevation range improve skill of area-volume scaling. These scaling relationships are applied to each record in the Randolph Glacier Inventory which is the first globally complete inventory of glaciers and ice caps. I estimate that the total volume of all glaciers in the world is 0.35±0.07 m sea level equivalent, including ice sheet peripheral glaciers. This is substantially less than a recent state-of-the-art estimate. Area volume scaling bias issues for large ice masses, and incomplete inventory data are offered as explanations for the difference.
Citation: Grinsted, A. (2013): An estimate of global glacier volume, The Cryosphere, 7, 141–151, doi:10.5194/tc-7-141-2013
More details on the study can be found under this link where I discuss:
Figure caption: The volume fraction stored in all the glaciers larger than a given area. Dark cyan shows the results of this study, and thin bright cyan excluding regions with many glacier complexes in RGI v2. The distribution from Huss and Farinotti (2012) is shown in green.
This figure shows that ~85% of the global glacier volume is stored in ~1000 largest RGI glacier complexes (>100 km2). In the paper I suggest that we can improve the glacier estimate through detailed studies of those complexes.