Disentangling the debris-cover anomaly in High Mountain Asia

Rocky debris covers 30% of glacier ablation areas in High Mountain Asia and generally suppresses melt. However, remote sensing observations have shown no statistical difference in glacier thinning rates between areas with and without debris cover; the ‘debris cover anomaly’. This pattern is apparent at subregional and regional scales, even after controlling for the elevation differences between debris-covered and clean ice. 

Two primary hypotheses to explain this behaviour have interpreted the thinning patterns in terms of melt or ice supply differences. First, rapid melt at supraglacial ponds and ice cliffs could enhance ablation in debris-covered areas, and therefore thinning as well. These features cannot entirely compensate for the melt reduction under debris, so a second hypothesis interprets the anomaly to indicate differences in emergence velocity between debris-covered and clean ice. However, complete understanding of the problem is challenged by a scale gap: the prior process studies have focused on single glaciers, whereas the anomaly has been identified for subregional- to regional spatial scales. Furthermore, these hypotheses neglect numerous other differences between debris-covered and clean glaciers (e.g. topo-climatic situation, accumulation mechanisms), which could bias this comparison.

We overcome these limitations through a direct assessment leveraging diverse large datasets and modelling. We firstly estimate emergence velocities and map ice cliffs and supraglacial ponds on a glacier-by-glacier basis across High Mountain Asia. We additionally assess other factors that could contribute to unexpected specific mass balance patterns: thin debris melt enhancement, distinct topo-climatic settings and the importance of avalanching for debris-covered ice. To determine the contribution of each factor to the debris-cover anomaly, we develop a statistical metric of how anomalous sub-debris ablation rates are, based on the difference in ablation rates between debris-covered and clean ice, as well as its altitudinal pattern. We use this metric and systematically remove the influence of the above hypothesized controls from each glacier’s emergence-corrected thinning data (specific mass balance) in a full-factorial investigation.

Our results firstly demonstrate that although emergence velocity differences between clean and debris-covered ice are systematic across the region, they do not resolve the debris-cover anomaly at the subregional or regional scale (altitudinal ablation rates are more negative for debris than clean ice). We find that accounting for any additional factor reduces the strength of the debris anomaly at regional and subregional scales, and our full-factorial analysis suggests that multiple factors combine to explain the debris cover anomaly.  Our results indicate that both hypotheses are correct in their process understanding at the glacier scale (reduced emergence velocity under debris, substantial ice cliff and pond ablation contribution), but not in their interpretation of the debris cover anomaly. Rather, our results underline previous suggestions that debris-covered glaciers fundamentally differ from clean ice glaciers in terms of mass supply mechanisms (i.e. supported by avalanching) and ablation patterns, leading to distinctive geometric expression and dynamics, and that the debris anomaly results from the integration of these patterns across scales.