Benthic macrofaunal carbon fluxes and environmental drivers of spatial variability in a large coastal-plain estuary

Abstract. While the importance of carbon cycling in estuaries is increasingly recognized, the role of benthic macrofauna remains poorly quantified due to limited spatial and temporal resolution in biomass measurements. Here, we ask: (1) To what extent do benthic macrofauna contribute to estuarine carbon cycling via respiration and calcification? and (2) How well can routinely collected environmental variables predict their biomass? We analyzed data from 8128 benthic samples collected from the Chesapeake Bay between 1995 and 2022 and estimated associated carbon fluxes using empirical relationships. We then used generalized additive models to relate observed and modeled environmental variables to the biomass. Biomass was highest in the upper mainstem of the Bay (Upper Bay) and upper Potomac River Estuary, the largest tidal tributary of the Bay. In the Upper Bay, benthic macrofauna respired 18 %–45 % of the estimated organic carbon supply. Calcification-driven alkalinity reduction reached 6.31 ± 2.84 mol m−2 yr−1 in the Potomac River Estuary, aligning with prior estimates of alkalinity sinks in the tributary and highlighting the potential importance of calcifying fauna in alkalinity dynamics. Estimated CO2 production in the Upper Bay from benthic respiration and calcification (151 g C m−2 yr−1) also exceeded observed air–sea CO2 fluxes (74.5 g C m−2 yr−1). Generalized additive models revealed that low salinity, moderate dissolved oxygen, and elevated nitrate best predicted high-biomass zones, with the three predictors explaining 52 % of biomass deviance. These predictive relationships offer a pathway to estimate macrofaunal biomass and associated carbon fluxes in systems where direct biomass measurements are sparse. Our findings demonstrate that benthic macrofauna play a substantial and spatially structured role in estuarine carbon cycling. Incorporating their contributions into estuarine biogeochemical models will improve predictions of ecosystem responses to environmental and anthropogenic change.