Carbon dioxide dynamics driven by groundwater discharge in a coastal floodplain creek

Document Type


Publication details

Atkins, ML, Santos, IR, Ruiz-Halpern, S & Maher, DT 2013, 'Carbon dioxide dynamics driven by groundwater discharge in a coastal floodplain creek', Journal of Hydrology, vol. 493, pp. 30-42.

Published version available from:


Peer Reviewed



Dissolved carbon dioxide (CO2) may be highly enriched in groundwater. However, the contribution of groundwater discharge as a source of CO2 to rivers, estuaries and coastal waters is poorly understood. We performed high resolution measurements of radon (222Rn, a natural groundwater tracer) and the partial pressure of CO2 (pCO2) in a highly modified tidal creek and estuary (North Creek, Richmond River, New South Wales, Australia) to assess whether CO2 in surface waters was driven by groundwater discharge. A spatial survey revealed increasing 222Rn activities (up to 17.3 dpm L−1) and pCO2 (up to 11,151 μatm) in the upstream direction. The enrichment occurred in a drained coastal acid sulphate soil wetland upstream of a mangrove forest. Time series experiments (24-h) were performed at two stations upstream and downstream of the pCO2 enrichment area. Upstream measurements demonstrated a significant correlation between pCO2 and 222Rn while downstream values resulted in a significant inverse relationship between pCO2 and dissolved oxygen apparently as a result of respiration in nearby mangroves. Measurements taken 2 days after a 245 mm precipitation event revealed the highest recorded 222Rn activities (up to 86.1 dpm L−1) and high pCO2 (up to 11,217 μatm), showing a strong groundwater influence after flooding. These observations imply that groundwater discharge drove CO2 dynamics at the upstream station while multiple complex processes drove CO2 at the downstream station. A 222Rn mass balance model demonstrated that groundwater discharge accounted for about 76% of surface water in this floodplain creek. The CO2 evasion rates (799 ± 225 mmol m−2 d−1) were driven primarily by currents rather than wind. Groundwater-derived CO2 fluxes into the creek averaged 1622 mmol m−2 d−1, a value twice as high as atmospheric CO2 evasion and consistent with carbon uptake within the creek and downstream exports. These results demonstrate that groundwater seepage was a major factor driving CO2 evasion to the atmosphere from the creek. Groundwater discharge should be accounted for in CO2 budgets in coastal systems.

Grant ID