Glacier meltwater and monsoon precipitation drive Upper Ganges Basin dissolved organic matter composition

Author Posting. © The Author(s), 2019. This is the author's version of the work. It is posted here by permission of Elsevier Ltd. for personal use, not for redistribution. The definitive version was published in Geochimica et Cosmochimica Acta 244 (2019): 216-228, doi:10.1016/j.gca.2018.10.012.Mountain glaciers store dissolved organic carbon (DOC) that can be exported to river networks and subsequently respired to CO2. Despite this potential importance within the global carbon cycle, the seasonal variability and downstream transport of glacier-derived DOC in mountainous river basins remains largely unknown. To provide novel insight, here we present DOC concentrations and molecular-level dissolved organic matter (DOM) compositions from 22 nested, glaciated catchments (1.4 – 81.8 % glacier cover by area) in the Upper Ganges Basin, Western Himalaya over the course of the Indian summer monsoon (ISM) in 2014. Aliphatic and peptide-like compounds were abundant in glaciated headwaters but were overprinted by soil-derived phenolic, polyphenolic and condensed aromatic material as DOC concentrations increase moving downstream. Across the basin, DOC concentrations and soil-derived compound class contributions decreased sharply from pre- to post-ISM, implying increased relative contribution of glaciated headwater signals as the monsoon progresses. Incubation experiments further revealed a strong compositional control on the fraction of bioavailable DOC (BDOC), with glacier-derived DOC exhibiting the highest bioavailability. We hypothesize that short-term (i.e. in the coming decades) increases in glacier melt flux driven by climate change will further bias exported DOM toward an aliphatic-rich, bioavailable signal, especially during the ISM and post-ISM seasons. In contrast, eventual decreases in glacier melt flux due to mass loss will likely lead to more a soil-like DOM composition and lower bioavailability of exported DOC in the long term.We thank Britta Voss (WHOI) for assisting with sample collection; Travis Drake (FSU), and Ekaterina Bulygina (Woods Hole Research Center) for laboratory assistance; and the NHMFL ICR user program (NSF-DMR-1157490) for aiding in data acquisition and analysis. This study was partly supported by NSF-DEB-1145932 to R.G.M.S. J.D.H. was partially supported by the NSF Graduate Research Fellowship Program under grant number 2012126152, with additional support in the form of travel grants awarded by the MIT Houghten Fund and NHMFL. All data used in this study are available in the Supporting Information Tables S1 and S2