Variable export and consumption of alkalinity in coastal wetlands: insights from multi-year, high-frequency observations in an intertidal saltmarsh

Coastal wetlands laterally export a major portion of their fixed atmospheric CO₂ to coastal oceans as inorganic carbon via tidal exchange, which is considered as a potential mechanism of blue carbon storage in the ocean due to the long residence time (thousands of years) of inorganic carbon. Such an export from saltmarshes can be evaluated by fluxes of total dissolved inorganic carbon (DIC) and total alkalinity (TA) in tidal creeks. The exported TA should be distinguished from DIC to represent a long-term carbon sink in the ocean, but knowledge of TA exports remains limited due to limited direct measurements. Furthermore, most of the previous estimates of TA exports were based on short-term studies (e.g., over a few tidal cycles at different seasons). However, we find that long-term high-resolution measurements are critical to avoid any biases in these estimates because of their extreme heterogeneity. Herein, we used multi-year, high-frequency measurements to resolve the TA exports from an intertidal saltmarsh, which also enables a comparison to the high-frequency DIC exports. Our study site is located at the Sage Lot Pond (SLP), Massachusetts, USA, where in situ water quality sensors were deployed from 2012 to 2016 and bottle data of TA and DIC were collected over multiple tidal cycles across seasons in the tidal creek. This comprehensive dataset allows us to develop a machine learning algorithm to predict high-frequency TA time series for subsequent calculations of TA exports. We observed that (1) consistent monthly trends in TA exports across years, with a net consumption of 1.0 mol m^-2 yr^-1 (i.e., TA sink) from October to June and an export of 7.1 mol m^-2 yr^-1 from July to September; (2) similar annual TA exports averaging 1.9 mol m^-2 yr^-1 across different years; (3) annual TA exports 16 times less than those of DIC. The finding of the particularly lower TA exports relative to DIC is in contrast with the recent global synthesis and many previous studies. There was a substantial decrease in TA from marsh sediment porewater to the tidal creek, indicating a large TA removal during porewater exchange and transport. We propose several mechanisms to explain such a net TA removal. Firstly, large amounts of aerobic respiration in the tidal creek and surface sediment can remove TA.  Secondly, groundwater discharge into the tidal creek likely supplies many reduced compounds (i.e., Mn²⁺ and Fe²⁺), which can be oxidized causing a net TA removal. Lastly, sulfate reduction is the primary mechanism for TA production in porewaters, and the resulting S^2- may be oxidized prior to the formation and burial of pyrite, decreasing the TA exports. The low TA exports observed here could be collectively driven by these processes at different spatiotemporal scales. These results raise the question, between DIC and TA, which estimated flux may better represent long-term CO₂ sink from coastal wetlands. It also generates further interest in studying the fate of wetland exported DIC as it holds the key to credit lateral carbon export as blue carbon.