Dissolved inorganic carbon export from rivers of Great Britain: Spatial distribution and potential catchment-scale controls
A.M. Tye, Jennifer Williamson, Helen P. Jarvie, Nancy B. Dise, Dan Lapworth, Don Monteith, Richard Sanders, Daniel J. Mayor, Michael Bowes, Michael Bowes, Annette Burden, Nathan Callaghan, Gareth Farr, Stacey L. Felgate, Stuart W. Gibb, Peter Gilbert, Geoff Hargreaves, Patrick Keenan, Vassilis Kitidis, Monika D. Jürgens, Adrian P. Martin, Ian Mounteney, Philip D. Nightingale, M. Glória Pereira, Justyna Olszewska, Amy Pickard, Andrew P. Rees, Bryan M. Spears, Mark Stinchcombe, Debbie White, Peter J. Williams, Fred Worrall, Chris Evans
Abstract
Dissolved inorganic carbon (DIC) fluxes from the land to ocean have been quantified for many rivers globally. However, CO2 fluxes to the atmosphere from inland waters are quantitatively significant components of the global carbon cycle that are currently poorly constrained. Understanding, the relative contributions of natural and human-impacted processes on the DIC cycle within catchments may provide a basis for developing improved management strategies to mitigate free CO2 concentrations in rivers and subsequent evasion to the atmosphere. Here, a large, internally consistent dataset collected from 41 catchments across Great Britain (GB), accounting for ∼36% of land area (∼83,997 km2) and representative of national land cover, was used to investigate catchment controls on riverine dissolved inorganic carbon (DIC), bicarbonate (HCO3−) and free CO2 concentrations, fluxes to the coastal sea and annual yields per unit area of catchment. Estimated DIC flux to sea for the survey catchments was 647 kt DIC yr−1 which represented 69% of the total dissolved carbon flux from these catchments. Generally, those catchments with large proportions of carbonate and sedimentary sandstone were found to deliver greater DIC and HCO3− to the ocean. The calculated mean free CO2 yield for survey catchments (i.e. potential CO2 emission to the atmosphere) was 0.56 t C km−2 yr−1. Regression models demonstrated that whilst river DIC (R2 = 0.77) and HCO3− (R2 = 0.77) concentrations are largely explained by the geology of the landmass, along with a negative correlation to annual precipitation, free CO2 concentrations were strongly linked to catchment macronutrient status. Overall, DIC dominates dissolved C inputs to coastal waters, meaning that estuarine carbon dynamics are sensitive to underlying geology and therefore are likely to be reasonably constant. In contrast, potential losses of carbon to the atmosphere via dissolved CO2, which likely constitute a significant fraction of net terrestrial ecosystem production and hence the national carbon budget, may be amenable to greater direct management via altering patterns of land use.