skip to main content


The NSF Public Access Repository (NSF-PAR) system and access will be unavailable from 11:00 PM ET on Thursday, June 13 until 2:00 AM ET on Friday, June 14 due to maintenance. We apologize for the inconvenience.

Title: Temporal Variability of Air‐Sea CO 2 flux in the Western Tropical North Atlantic Influenced by the Amazon River Plume

The partial pressure of carbon dioxide (pCO2) was surveyed across the Amazon River plume and the surrounding western tropical North Atlantic Ocean (15–0°N, 43–60°W) during three oceanic expeditions (May–June 2010, September–October 2011, and July 2012). The survey timing was chosen according to previously described temporal variability in plume behavior due to changing river discharge and winds.In situsea surfacepCO2and air‐sea CO2flux exhibited robust linear relationships with sea surface salinity (SSS; 15 < SSS < 35), although the relationships differed among the surveys. Regional distributions ofpCO2and CO2flux were estimated using SSS maps from high‐resolution ocean color satellite‐derived (MODIS‐Aqua) diffuse attenuation coefficient at 490 nm (Kd490) during the periods of study. Results confirmed that the plume is a net CO2sink with distinctive temporal variability: the strongest drawdown occurred during the spring flood (−2.39 ± 1.29 mmol m−2 d−1in June 2010), while moderate drawdown with relatively greater spatial variability was observed during the transitional stages of declining river discharge (−0.42 ± 0.76 mmol m−2 d−1in September–October 2011). The region turned into a weak source in July 2012 (0.26 ± 0.62 mmol m−2 d−1) when strong CO2uptake in the mid‐plume was overwhelmed by weak CO2outgassing over a larger area in the outer plume. Outgassing near the mouth of the river was observed in July 2012. Our observations draw attention to the importance of assessing the variable impacts of biological activity, export, and air‐sea gas exchange before estimating regional CO2fluxes from salinity distributions alone.

more » « less
Author(s) / Creator(s):
 ;  ;  ;  ;  ;  ;  ;  ;  ;  
Publisher / Repository:
DOI PREFIX: 10.1029
Date Published:
Journal Name:
Global Biogeochemical Cycles
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Analysis of the time-dependent behavior of the buoyant plume rising above Grotto Vent (Main Endeavour Field, Juan de Fuca Ridge) as imaged by the Cabled Observatory Vent Imaging Sonar (COVIS) between September 2010 and October of 2015 captures long term time-dependent changes in the direction of background bottom currents independent of broader oceanographic processes, indicating a systematic evolution in vent output along the Endeavour Segment of the Juan de Fuca Ridge. The behavior of buoyant plumes can be quantified by describing the volume, velocity, and orientation of the effluent relative to the seafloor, which are a convolved expression of hydrothermal flux from the seafloor and ocean bottom currents in the vicinity of the hydrothermal vent. We looked at the azimuth and inclination of the Grotto plume, which was captured in three-dimensional acoustic images by the COVIS system, at 3-h intervals during October 2010 and between October 2011 and December 2014. The distribution of plume azimuths shifts from bimodal NW and SW to SE in 2010, 2011, and 2012 to single mode NW in 2013 and 2014. Modeling of the distribution of azimuths for each year with a bimodal Gaussian indicates that the prominence of southward bottom currents decreased systematically between 2010 and 2014. Spectral analysis of the azimuthal data showed a strong semi-diurnal peak, a weak or missing diurnal peak, and some energy in the sub-inertial and weather bands. This suggests the dominant current generating processes are either not periodic (such as the entrainment fields generated by the hydrothermal plumes themselves) or are related to tidal processes. This prompted an investigation into the broader oceanographic current patterns. The surface wind patterns in buoy data at two sites in the Northeast Pacific and the incidence of sea-surface height changes related to mesoscale eddies show little systematic change over this time-period. The limited bottom current data for the Main Endeavour Field and other parts of the Endeavour Segment neither confirm nor refute our observation of a change in the bottom currents. We hypothesize that changes in venting either within the Main Endeavour Field or along the Endeavour Segment have resulted in the changes in background currents. Previous numerical simulations (Thomson et al., J. Geophys. Res., 2009, 114 (C9), C09020) showed that background bottom currents were more likely to be controlled by the local (segment-scale) venting than by outside ocean circulation or atmospheric patterns. 
    more » « less
  2. Abstract

    Existing analyses of salt marsh carbon budgets rarely quantify carbon loss as CO2through the air–water interface in inundated marshes. This study estimates the variability of partial pressure of CO2(pCO2) and air–water CO2fluxes over summer and fall of 2014 and 2015 using high‐frequency measurements of tidal waterpCO2in a salt marsh of the U.S. northeast region. Monthly mean CO2effluxes varied in the range of 5.4–25.6 mmol m−2marsh d−1(monthly median: 4.8–24.7 mmol m−2marsh d−1) during July to November from the tidal creek and tidally‐inundated vegetated platform. The source of CO2effluxes was partitioned between the marsh and estuary using a mixing model. The monthly mean marsh‐contributed CO2effluxes accounted for a dominant portion (69%) of total CO2effluxes in the inundated marsh, which was 3–23% (mean 13%) of the corresponding lateral flux rate of dissolved inorganic carbon (DIC) from marsh to estuary. Photosynthesis in tidal water substantially reduced the CO2evasion, accounting for 1–86% (mean 31%) of potential CO2evasion and 2–26% (mean 11%) of corresponding lateral transport DIC fluxes, indicating the important role of photosynthesis in controlling the air–water CO2evasion in the inundated salt marsh. This study demonstrates that CO2evasion from inundated salt marshes is a significant loss term for carbon that is fixed within marshes.

    more » « less
  3. Abstract

    The Virgin Islands basin (VIB) includes several Marine Protected Areas (MPAs) of interest as biologically unique spawning aggregation sites. The ecological structure in and around these MPAs is regulated by several factors, including changes in near‐surface water properties. Anomalously low near‐surface salinity is observed in the VIB during April 2009/2011, and March 2010, with a salinity signature consistent with Amazon plume waters. Other low salinity events in the region are found during 2007–2017 using output from an ocean reanalysis. The reanalysis shows that horizontal salinity advection explains near‐surface salinity variability in the VIB to a high degree, including events observed in the in situ measurements. We use a Lagrangian Particle tracking model to track particles over the 2007–2017 period and identify the source and pathways of water imports to the VIB. We describe three pathways. The northernmost one is often associated with advection of salty Atlantic waters. The two southernmost paths are associated with advection of low salinity waters from the Amazon into the VIB. The latter two pathways arrive to the Caribbean Sea as described in previous studies on low salinity advection to the wider Caribbean from the Amazon River; we find that once in the Caribbean Sea, the low salinity water makes its way into the VIB when steered northward by mesoscale features. This results in Amazon River waters regulating salinity variability in the VIB during April–November. During December–March, when mesoscale activity is at its minimum, the Atlantic inflow regulates the salinity variability within the VIB instead.

    more » « less
  4. Abstract

    Shifting baselines in the Arctic atmosphere‐sea ice‐ocean system have significant potential to alter biogeochemical cycling and ecosystem dynamics. In particular, the impact of increased open water duration on lower trophic level productivity and biological CO2sequestration is poorly understood. Using high‐resolution observations of surface seawater dissolved O2/Ar andpCO2collected in the Pacific Arctic in October 2011 and 2012, we evaluate spatial variability in biological metabolic status (autotrophy vs heterotrophy) as constrained by O2/Ar saturation (∆O2/Ar) as well as the relationship between net biological production and the sea‐air gradient ofpCO2(∆pCO2). We find a robust relationship between∆pCO2and∆O2/Ar(correlation coefficient of −0.74 and −0.61 for 2011 and 2012, respectively), which suggests that biological production in the late open water season is an important determinant of the air‐sea CO2gradient at a timeframe of maximal ocean uptake for CO2in this region. Patchiness in biological production as indicated by∆O2/Arsuggests spatially variable nutrient supply mechanisms supporting late season growth amidst a generally strongly stratified and nutrient‐limited condition.

    more » « less
  5. null (Ed.)
    The first eruption at Kīlauea’s summit in 25 years began on March 19, 2008, and persisted for 10 years. The onset of the eruption marked the first explosive activity at the summit since 1924, forming the new “Overlook crater” (as the 2008 summit eruption crater has been informally named) within the existing crater of Halemaʻumaʻu. The first year consisted of sporadic lava activity deep within the Overlook crater. Occasional small explosions deposited spatter and small wall-rock lithic pieces around the Halemaʻumaʻu rim. After a month-long pause at the end of 2008, deep sporadic lava lake activity returned in 2009. Continuous lava lake activity began in February 2010. The lake rose significantly in late 2010 and early 2011, before subsequently draining briefly in March 2011. This disruption of the summit eruption was triggered by eruptive activity on the East Rift Zone. Rising lake levels through 2012 established a more stable, larger lake in 2013, with continued enlargement over the subsequent 5 years. Lava reached the Overlook crater rim and overflowed on the Halemaʻumaʻu floor in brief episodes in 2015, 2016, and 2018, but the lake level was more commonly 20–60 meters below the rim during 2014–18. The lake was approximately 280×200 meters (~42,000 square meters) by early 2018 and formed one of the two largest lava lakes on Earth. A new eruption began in the lower East Rift Zone on May 3, 2018, causing magma to drain from the summit reservoir complex. The lava in Halemaʻumaʻu had drained below the crater floor by May 10, followed by collapse of the Overlook and Halemaʻumaʻu craters. The collapse region expanded as much of the broader summit caldera floor subsided incrementally during June and July. By early August 2018, the collapse sequence had ended, and the summit was quiet. The historic changes in May–August 2018 brought a dramatic end to the decade of sustained activity at Kīlauea’s summit. The unique accessibility of the 2008–18 lava lake provided new observations of lava lake behavior and open-vent basaltic outgassing. Data indicated that explosions were triggered by rockfalls from the crater walls, that the lake consisted of a low-density foamy lava, that cycles of gas pistoning were rooted at shallow depths in the lake, and that lake level fluctuations were closely tied to the pressure of the summit magma reservoir. Lava chemistry added further support for an efficient hydraulic connection between the summit and East Rift Zone. Notwithstanding the benefits to scientific understanding, the eruption presented a persistent hazard of volcanic air pollution (vog) that commonly extended far from Kīlauea’s summit. 
    more » « less