Errors in Regional Extrapolation Methods Using Data From the Western Chugach Mountains, Alaska, USA.

We used airborne laser altimetry measurements to determine the volume changes of 23 glaciers in the Western Chugach Mountains, Alaska, USA between 1950/57 to 2001/2004. We used these data to test several regional extrapolation methods by removing individual glaciers from the sample and estimating their changes from averages of the remaining measurements. Regional extrapolation methods included those which accounted for variations in surface area distribution with elevation, those which only considered total glacier changes, and those which scaled changes in glacier area to changes in volume. We found that although different methods produced widely varying estimates for individual glaciers, errors in total regional volume changes were similar for each method (15 to 19%). This is because on a regional scale the sum of extrapolation errors for individual glaciers (both positive and negative) approximately balanced each other. We found that the range of errors introduced by the choice of regional extrapolation method were on the same order of magnitude as errors in our outlines of the glacier surface area. We therefore stress the need for accurate maps of glacier extent as a key component of any regional glacier volume change study. We used an ensemble of regional extrapolation methods to arrive at an estimate of the total rate of volume loss of Western Chugach glaciers (7.4±1.1 km3 yr-1 water equivalent, causing a rise in sea level of 0.02±0.003 mm yr-1 between 1950/1957 to 2001/2004). Nearly half of this change was dominated by Columbia Glacier, a large and rapidly retreating tidewater glacier (above sea level rate of volume loss of 3.1±0.08 km3 yr-1 water equivalent). The tidewater glacier cycle is largely independent of climate and can result in glacier changes that are not coherent with others in a region. Our data illustrate how the cycle is not synchronized within a region, as we observed with two adjacent tidewater glaciers, one advancing and the other retreating. It is therefore important to separate tidewater glaciers from regional extrapolations and consider their contributions to sea level change individually.