Radon as a tracer of submarine groundwater discharge into a boat basin in Donnalucata, Sicily

Abstract

Estimates of groundwater discharge into a boat basin in Donnalucata, southeastern Sicily, based on radon measurements are reported. A continuous monitor was deployed at several locations that made repeated integrated measurements of radon concentration in the seawater once per hour. We observed large differences in the concentrations (60–2500 Bq m −3) and inventories (90–3200 Bq m −2) of 222Rn at various locations within the boat basin. The station closest to shore had the highest concentration by far (up to over 3000 Bq m −3), reflecting inputs of radon-rich groundwater into this zone. We constructed a radon mass balance by assuming that radon may be advected via groundwater discharge into all compartments of the boat basin and the mixing outputs from an inner “box” would be an input into the adjacent box. We also made allowances for atmospheric evasion losses that were occasionally very high due to high winds and huge concentration gradients across the air–sea interface. Once the radon fluxes have been estimated, we could calculate groundwater discharge by dividing these fluxes by the radon concentration of the groundwater. Should the groundwater be entering the basin by widely disseminated seepage through bottom sediments, it would be appropriate to use values of pore water derived by sediment equilibration techniques. These values (2700±200 Bq m −3) are close to those from several samples from shallow wells collected near the boat basin (1700–2700 Bq m −3). However, we measured higher radon levels than this in the waters of the innermost portion of the basin. We thus elected to assume that the main radon entry was either via direct spring input with activities (15 100 Bq m −3) similar to those measured from a natural spring on a nearby beach or a blend of spring water with surficial aquifer water. The resulting range in our calculated groundwater discharges into the boat basin is 1200 to 7400 m 3 day −1. These estimates are higher than those based on seepage meters (range=300–1000 m 3 day −1), probably because of the wide spatial variability of the spring and seep inputs.