Long-term trends in dissolved organic carbon concentration: a cautionary note

Dissolved organic carbon (DOC) plays an important role in surface water chemistry and ecology and trends in DOC concentration have been also associated with shifts in terrestrial carbon pools. Numerous studies have reported long-term trends in DOC concentration; however, some studies consider changes in average measured DOC whereas other compute discharge weighted concentrations. Because of differences in reporting methods and variable record lengths it is difficult to compare results among studies and make regional generalizations. Furthermore, changes in stream discharge may impact long-term trends in DOC concentration and the potentially subtle effect of shifts in stream flow may be missed if only measured DOC concentrations are considered. In this study we compare trends in volume-weighted vs. average measured DOC concentration between 1980 and 2001 at seven headwater streams in south-central Ontario, Canada that vary in wetland coverage and DOC (22-year mean vol. wt.) from 3.4 to 10.6 mg l(-1). On average, annual measured DOC concentrations were 13-34% higher than volume-weighted values, but differences of up to 290% occurred in certain years. Estimates of DOC flux were correspondingly higher using measured concentration values. Both measured and volume-weighted DOC concentrations increased significantly between 1980 and 2001, but slopes were larger in measured data (0.04-0.35 mg l(-1) year(-1) compared with 0.05-0.15 mg l(-1) year(-1)) and proportional increases at the most wetland-influenced sites ranged from 32 to 43% in volume-weighted DOC and from 52 to 75% in measured DOC. In contrast, DOC flux did not change with time when estimated using either method, because of the predominant influence of stream flow on DOC export. Our results indicate that changes in stream flow have an important impact on trends in DOC concentration, and extrapolation of trend results from one region to another should be made cautiously and consider methodological and reporting differences among sites.