ALBERT

Description: Recent studies have determined mass changes of Arctic ice caps and glaciers from satellite altimetry. Determining regional mass balance of ice caps and glaciers using this technique is inherently difficult due to their size and geometry. Furthermore these studies have mostly relied on one method or the same types of methods to determine the regional mass balance, by extrapolating elevation changes using their relation to elevation. This makes the estimation of mass balance heavily dependent on the method used to regionalize the elevation changes. Left without consideration large discrepancies can arise in the mass change estimates and the interpretation of them. In this study we use Ice, Cloud, and land Elevation Satellite (ICESat) derived elevation changes from 2003–2009 and determine the impact of different regionalizing schemes on the mass change estimates of the Arctic ice caps and glaciers. Four different methods, based on interpolation and extrapolation of the elevation changes were used to quantify this effect on the regional mass changes. Secondly, a statistical criteria was developed to determine the optimum method for each region in order to derive robust mass changes and reduce the need of external validation data. In this study we found that the range or spread of the estimated mass changes, for the different regions, was highly correlated to the inter-annual variability of the elevation changes, driven by the different climatic conditions of the regions. Regions affected by a maritime climate show a large range in estimated values, on average 1.5–2 times larger than the predicted errors. For regions in a continental regime the opposite was observed, and the range of the values lies well inside the error estimates. We also found that the extrapolation methods tend on average to produce more negative values than the interpolation methods and that our four methods do not fully reproduce the original histogram. Instead, they produce more negative distributions than the original which may indicate that previous and these current estimates using ICESat observations might be overestimate by as much as 4–19%, depending on region. This should therefore be taken into account when deriving regional mass balance from satellite altimetry in regions which show high inter-annual variability of elevation changes. In these regions several different independent methods should be used to capture the elevation change pattern and then analyzed to determine the most suitable method. For regions in a continental climate regime, and with low variability of elevation changes, a single method may be sufficient to capture the regional elevation change pattern and hence mass balance.