Challenges in Quantifying Air‐Water Carbon Dioxide Flux Using Estuarine Water Quality Data: Case Study for Chesapeake Bay

Abstract

Estuaries play an uncertain but potentially important role in the global carbon cycle via CO2 outgassing. The uncertainty mainly stems from the paucity of studies that document the full spatial and temporal variability of estuarine surface water partial pressure of carbon dioxide ( pCO2). Here, we explore the potential of utilizing the abundance of pH data from historical water quality monitoring programs to fill the data void via a case study of the mainstem Chesapeake Bay (eastern United States). We calculate pCO2 and the air‐water CO2 flux at monthly resolution from 1998 to 2018 from tidal fresh to polyhaline waters, paying special attention to the error estimation. The biggest error is due to the pH measurement error, and errors due to the gas transfer velocity, temporal sampling, the alkalinity mixing model, and the organic alkalinity estimation are 72%, 27%, 15%, and 5%, respectively, of the error due to pH. Seasonal, interannual, and spatial variability in the air‐water flux and surface pCO2 is high, and a correlation analysis with oxygen reveals that this variability is driven largely by biological processes. Averaged over 1998–2018, the mainstem bay is a weak net source of CO2 to the atmosphere of 1.2 (1.1, 1.4) mol m−2 yr−1 (best estimate and 95% confidence interval). Our findings suggest that the abundance of historical pH measurements in estuaries around the globe should be mined in order to constrain the large spatial and temporal variability of the CO2 exchange between estuaries and the atmosphere.

Mechanisms Driving Decadal Changes in the Carbonate System of a Coastal Plain Estuary

“…Variability in biological modifications and seawater chemistry, rather than changing atmospheric CO 2 , explains the spatiotemporal patterns of decadal changes in the Chesapeake Bay surface carbonate system. In the summer and in the middle Bay, surface DIC and p CO 2 levels are lower than other seasons and regions due to high primary production (Brodeur et al., 2019; Friedman et al., 2020; Herrmann et al., 2020). For a seasonally uniform increase in atmospheric CO 2 , there can be more uptake of CO 2 in the summer and in the middle Bay where there is a biologically driven undersaturation of p CO 2 relative to the atmosphere, which leads to greater increases in surface DIC (Figure 6a).…”

“…It is more challenging to compare the results of this modeling study with existing observationally derived trends. For example, a change in measurement protocols around 1996 prohibits a comparison of pH data collected before and after this date (see Herrmann et al., 2020 for details). The analysis in the present study thus avoids using pH data from those early years, but there remains a possibility that the change in methodology after 1996 contributed to artificially high pH values in the lower Chesapeake Bay in the analyses of Waldbusser et al.…”

“…Waldbusser et al (2011) conducted linear regressions of glass electrode pH data . Although they found significant decreasing trends in surface pH in the lower Bay, the changes in pH sampling methodology prior/after 1996 (Herrmann et al, 2020) likely bias the trends calculated using these data. Herrmann et al (2020) excluded these early pH data in the Chesapeake Bay and analyzed the spatiotemporal patterns of the calculated CO 2 partial pressure (pCO 2 ) and air-water CO 2 exchange over the years 1998-2018.…”

“…Although they found significant decreasing trends in surface pH in the lower Bay, the changes in pH sampling methodology prior/after 1996 (Herrmann et al, 2020) likely bias the trends calculated using these data. Herrmann et al (2020) excluded these early pH data in the Chesapeake Bay and analyzed the spatiotemporal patterns of the calculated CO 2 partial pressure (pCO 2 ) and air-water CO 2 exchange over the years 1998-2018. The Bay was found to be a weak source of CO 2 to the atmosphere, with more outgassing during the early part of the record, perhaps in DA ET AL.…”

Effects of Wind-Driven Lateral Upwelling on Estuarine Carbonate Chemistry

“…Overall, the complicated non-linear dependence of CO 2 flux on the wind speed and strong spatial variability make it difficult to identify a simple relationship between the CO 2 flux and wind forcing in this estuary. Although numerical model estimates of airsea@ CO 2 flux can help overcome challenges in making similar estimates from limited and error-prone observations (Herrmann et al, 2020), assigning causation to these changes under dynamic physical forcing is difficult even with model simulations.…”