Know the flow - spatial and temporal variation of DOC exports and the importance of monitoring site specific discharge

Sammanfattning: The export of dissolved organic carbon (DOC) is an important component in both water quality and carbon cycling within boreal catchments. DOC exports are mainly controlled by discharge. Therefore, the resolution and quality of discharge data is essential when quantifying DOC exports as well as for the understanding of its spatial and temporal variation. Although there are many studies quantifying DOC exports, many fail to consider the variability of discharge between different catchments within a landscape, instead often assuming uniform specific discharge. In this study, spatial and temporal variation in DOC concentrations and exports was studied for 13 nested subcatchments of a boreal catchment located in northern Sweden. The subcatchments were studied during a nine-year observation period (2009-2017). Discharge was monitored at each site and exports were calculated using daily DOC concentrations interpolated from frequent water sampling. Positive correlations between discharge and DOC concentrations were observed for most of the sites. Higher concentrations correlated with increased wetland coverage. Seasonal variations in DOC concentrations were found between subcatchments dominated by forest compared wetlands. The annual mean DOC export ranged between 26 and 108 kg ha-1 yr-1 and correlated positively with wetland coverage (R2 = 0.67, p < 0.001). Spatial variation on both annual and seasonal timescales, could to a large extent be explained by the catchment characteristics using principal component analysis. Forest and wetland coverage, as well as the size and elevation of each subcatchment explained most of the spatial variation of DOC exports. The importance of these catchment characteristics varied with season, where the two-month long spring season contributed between 32 to 52% of the annual DOC export. For annual exports, the proportion exported during spring season was larger for forest-dominated catchments compared to wetland catchments. Comparing export calculated adopting variable discharge to exports adopting uniform specific discharge revealed an average absolute difference of 18% ranging between 5 to 32% for individual subcatchments. DOC exports from wetland-dominated catchments were underestimated while forest-dominated catchments were overestimated. This study clearly demonstrates that adopting uniform specific discharge provides erroneous quantifications of DOC exports by not being able to capture the spatial variability within the system. This highlights the importance of discharge data with high resolution and quality when estimating element fluxes.