General rightsThis document is made available in accordance with publisher policies. Please cite only the published version using the reference above. Full terms of use are available: http://www.bristol.ac.uk/pure/about/ebr-terms compounds made up 62% of the dissolved organic carbon exported from the glacier surface 36 through streams. We therefore conclude that microbial communities were the primary driver 37 for labile dissolved organic carbon production and recycling on glacier surfaces (up to 1.12 ± 38 0.14 mg C L −1 d −1 carbon production), and that glacier dissolved organic carbon export is 39 dependent on active microbial processes during the melt season.

doi: 10.7185/geochemlet.1611Darkening of glacier and ice sheet surfaces is an important positive feedback to increasing global temperatures. Deposition of impurities on glaciers is primarily believed to reduce surface albedo, resulting in greater melt and mass loss. However, no study has yet included the effects of biological activity in albedo reduction models. Here, we provide the first experimental evidence that microbial activity can significantly decrease glacier surface albedo. Indeed, the addition of nutrients at ice meltwater concentrations to microbe-impurity mixtures resulted in extensive microbial organic carbon fixation and accumulation in Greenland Ice Sheet surface debris. Accumulated organic carbon, over the period of a melt season, darkened the glacial debris in our experiments from 31.1 % to 15.6 % surface reflectivity (used as an analogue for albedo in our calculations), generating a strongly absorbing surface. Our experiments are the first to quantify the microbially-induced potential melt increase for the Greenland Ice Sheet (up to an average of 17.3 ± 2.5 Gt yr -1 at present and up to ~85 Gt yr -1 by 2100, based on our first order calculations). Mass loss from glaciers will conceivably intensify through enhanced microbial activity, resulting from longer melt seasons and fertilisation from anthropogenic sources.

The first molecular-based studies of microbes in snow and on glaciers have only recently been performed on the vast Greenland Ice Sheet (GrIS). Aeolian microbial seeding is hypothesized to impact on glacier surface community compositions. Localized melting of glacier debris (cryoconite) into the surface ice forms cryoconite holes, which are considered ‘hot spots’ for microbial activity on glaciers. To date, few studies have attempted to assess the origin and evolution of cryoconite and cryoconite hole communities throughout a melt season. In this study, a range of experimental approaches was used for the first time to study the inputs, temporal and structural transformations of GrIS microbial communities over the course of a whole ablation season. Small amounts of aeolian (wind and snow) microbes were potentially seeding the stable communities that were already present on the glacier (composed mainly of Proteobacteria, Cyanobacteria, and Actinobacteria). However, the dominant bacterial taxa in the aeolian samples (Firmicutes) did not establish themselves in local glacier surface communities. Cryoconite and cryoconite hole community composition remained stable throughout the ablation season following the fast community turnover, which accompanied the initial snow melt. The presence of stable communities in cryoconite and cryoconite holes on the GrIS will allow future studies to assess glacier surface microbial diversity at individual study sites from sampling intervals of short duration only. Aeolian inputs also had significantly different organic δ13C values (-28.0 to -27.0‰) from the glacier surface values (-25.7 to -23.6‰), indicating that in situ microbial processes are important in fixing new organic matter and transforming aeolian organic carbon. The continuous productivity of stable communities over one melt season makes them important contributors to biogeochemical nutrient cycling on glaciers.