Comparison of Different Air–Water Gas Exchange Models to Determine Gaseous Mercury Evasion from Different European Coastal Lagoons and Estuaries

Abstract

Mercury (Hg) evasional fluxes from coastal waters are an important feature of the global Hg cycling; however, they remain poorly characterized due to limitations in datasets and a lack of suitable gas exchange models for such an environment. The objective of this study was to improve this evasional assessment by examining the extent and variability of gaseous Hg exchanges, computed with different gas exchange models, over various European estuaries (Gironde, Scheldt, Rhine) and coastal lagoons (Arcachon, Thau). The major outcomes of this work are the comparison of different models used to compute gaseous Hg exchanges and also the estimation of the variability of Hg fluxes at the air–water interface using a large database composed of 425 measurements acquired over a wide range of coastal ecosystems and seasons. The relative variance, expressed as the relative standard deviation of Hg° fluxes, was used to estimate the variability among models, seasons and sites. The inter-model relative variance (101 %) was the lowest compared to inter-site (182 %) or inter-season (172 %) variability. The highest fluxes were found during spring for the Gironde Estuary (73.7 pmol m−2 h−1), during summer for the Scheldt Estuary, the Arcachon and Thau Lagoons with respectively, 53.3, 55.3 and 23.6 pmol m−2 h−1 and during winter for the Rhine Estuary (50.1 pmol m−2 h−1). They were mainly explained by a combination of greater gaseous Hg production in the water column and wind speed. Overall, the results demonstrate that the sites and seasonal variations will generally overcome the uncertainty arising from the model selection. Generic models are therefore suitable for evasional assessments at the regional scale, while site-specific models should be used for local studies, when accurate mass balances are required.