Sunlight Absorption on the Greenland Ice Sheet Experiment (SAGE) – Tracing black carbon from emissions to deposition
Résumé
The SAGE project seeks to understand the impact of light absorbing impurities on the Greenland Ice Sheet (GrIS). In general, the project has found that black carbon and dust concentrations in snow were low in the dry snow zones of the GrIS during 2012-2014 and that their concentrations do not appear to be trending relative to observations of these concentrations in snow over recent decades. We provide a revised analysis of MODIS albedo trends on the GrIS using new collection 6 data. These indicate that observed albedo of dry snow is not substantially trending. Sensor drift which had been present in collection 5 data has been substantially removed and the observed albedo of dry snow on the GrIS is now showing near zero trend.
Episodic enhancements in BC deposition are, however, found in specific layers in our extensive snow pit observations. These peak enhancements include concentrations of up to ~40 ng/g BC and would have reduced the albedo of the snow by 0.01-0.02. If timed correctly, the deposition of such a layer could be an important factor in initiating a melt-albedo feedback. Here we present an overview of synthesis work seeking to trace the formation of such a layer back to emission sources and call attention to multiple presentations making up the project. Collectively, the work traces a specific enhanced deposition event occurring on the northwest region of the ice sheet in early August 2013 to source fires in Canada. We summarize the multi-modal approach including remote sensing of aerosols, atmospheric trajectory modeling, chemical transport modeling, and coupled Earth system modeling. The emission, transport, and deposition of the enhanced event is observed and predicted by these tools and we find general agreement between these several modes of sensing and predicting. Further investigations explore other events where BC was emitted and even transported over the ice sheet but did not cause deposition events, resulting in no BC signature in the snow. We also show the spatial distribution of enhanced BC deposition from the event we studied, and highlight its limited nature. The results have significance for understanding the role episodic BC deposition might play in interannual variability of ice sheet surface melt and for interpreting the relationship between fire-frequency and fire signatures in ice cores.