Comparing the performance of three methods to assess DOM dynamics within two distinct glacierized watersheds of the tropical Andes

• Published in: Environmental pollution (1987) 2020-10, Vol.265, p.115052-115052, Article 115052
• Main Authors: Rodriguez-Avella, K.A., Baraer, M., Mark, B., McKenzie, J., Somers, L.
• Format: Article
• Language: English
• Subjects: Dissolved organic matter; Fluorescence spectroscopy; Fluorescent dissolved organic matter; Parallel factor analysis; Water quality; Watershed scale
• ISSN: 0269-7491; 1873-6424
• DOI: 10.1016/j.envpol.2020.115052

Dissolved organic matter (DOM) is recognized as a good indicator of water quality as its concentration is influenced by land use, rainwater, windborne material and anthropogenic activities. Recent technological advances make it possible to characterize fluorescent dissolved organic matter (FDOM), the fraction of DOM that fluoresces. Among these advances, portable fluorometers and benchtop fluorescence excitation and emission spectroscopy coupled with a parallel factor analysis (EEM-PARAFAC) have shown to be reliable. Despite their rising popularity, there is still a need to evaluate the extent to which these techniques can assess DOM dynamics at the watershed scale. We compare the performance of in-situ measurements of FDOM with laboratory measurements of fluorescence spectroscopy within the context of two distinct glacierized watersheds in Peru. Glacierized watersheds represent unique testing environments with contrasting DOM conditions, flowing from pristine, vegetation-free headwaters through locations with obvious anthropogenic influences. We used an in-situ fluorometer and a portable multimeter to take 38 measurements of FDOM, pH and turbidity throughout the two catchments. Additionally, samples were analyzed in the laboratory using the EEM-PARAFAC method. Results were compared to dissolved organic carbon (DOC) measurements using standard high-temperature catalytic oxidation. Our results show that the three techniques together were able to capture the DOM dynamics for both studied watersheds. Taken individually, all three methods allowed detection of the watershed DOM main points of sources but in a more limited way. Due to the narrow bandwidth of the portable fluorometer used in the study, FDOM measurements were almost non-detectable to protein-like substances. Indeed, the more demanding EEM-PARAFAC was able to both differentiate between potential sources of DOM and provide an estimate of relative concentrations of different organic components. Finally, similar to FDOM but to a lesser extent, the DOC measurements showed some limits where protein-like substances make up most of the DOM composition. [Display omitted] •FDOM readings capture dynamics comparable to those of the whole DOM.•Protein-like component intensities increase in anthropogenically impacted sites.•FDOM in situ sensors with narrow bandwidth may miss a part of the FDOM pool.