Year-to-year salinity changes in the Amazon plume: Contrasting 2011 and 2012 Aquarius/SACD and SMOS satellite data
Introduction
The Amazon/Orinoco plume is a vast seasonal zone of surface water in the western tropical Atlantic with salinities that are several psu fresher than the water beneath the barrier layer (e.g. Lentz, 1995). Recent studies have linked the intensification of hurricanes to the presence of this plume due to the impact of haline stratification on reduction of vertical heat flux (e.g. Balaguru et al., 2012, Ffield, 2007, Grodsky et al., 2012). The seasonal extent of the plume is the result of several competing processes including changes in Amazon River discharge, advection, and turbulent mixing by the strongly seasonal winds (e.g. Hu et al., 2004, Nikiema et al., 2007, Zeng et al., 2008). Here we consider the potential role of nonlocal precipitation that changes the properties of the inflow to the region and changing plume advection in contributing to year to year variations in the plume.
The area of the combined Amazon/Orinoco plume, defined as the area covered by water with SSS < 35 psu, reaches its maximum northward and eastward extent in August–September when the zone of weak winds shifts northward. The plume contracts by November, coincident with the reappearance of the northeast trade winds and shifts in the surface currents (Dessier & Donguy, 1994) suggesting that diluted surface water is destratified by the strengthening winds of late fall (e.g. Grodsky et al., 2012). With the exception of rare surveys and thermosalinograph (TSG) observations along commercial shipping lanes the best information on year-to-year variations in the plume has come indirectly from tracking anomalous optical properties associated with coastal water (Hu et al., 2004, Muller-Karger et al., 1988, Salisbury et al., 2011, Zeng et al., 2008). These studies have suggested that year to year changes in Amazon River discharge play an important role driving year to year changes in the plume. However, since river discharge strongly affects water optical properties, this approach may only be detecting the portion of the plume associated with river discharge.
The Amazon River discharge varies seasonally from a minimum of 0.8 × 105 m3 s− 1 in November to a maximum of 2.4 × 105 m3 s− 1 in late May in response to the seasonal southward shift of the intertropical convergence zone (ITCZ, Lentz, 1995). During boreal winter, water in the plume is stored at the river mouth, trapped against the coast by onshore winds. In boreal spring, water in the plume is transported northwestward along the shelf and shelf break by the North Brazil current (NBC). By boreal summer this low salinity surface water is transported mainly by three export pathways (Fig. 1), whose seasonality is discussed by Foltz and McPhaden (2008). Some continue northwestward towards the Caribbean current (Hellweger & Gordon, 2002), some is transported northward by wind-driven Ekman currents and eddies into the barrier layer region east of the Lesser Antilles (Mignot, Lazar, & Lacarra, 2012), and some is carried eastward into the north equatorial counter current (NECC, Carton & Katz, 1990). Indeed, from August through October typically 70% of the Amazon plume water is deflected eastward in the NECC along the NBC retroflection (Lentz, 1995). These three export pathways are seen in the spatial pattern of September climatological SSS as low salinity directional lobes extending into the Caribbean, north subtropical Atlantic, and central tropical Atlantic along the NECC (Fig. 1). The Orinoco plume to the northeast also varies seasonally from a minimum of 1 × 104 m3 s− 1 in March to a maximum of 7 × 104 m3 s− 1 in August (e.g. Hu et al., 2004). The latter plume area may exceed 160,000 km2 in fall, but much of the Orinoco discharge is swept into the Southern Caribbean Sea. The NBC is the main route by which water is transported into the plume area. This suggests that water characteristics upstream in the South Equatorial Current region that feeds the NBC may impact salinity in the plume through this import pathway.
Interannual anomalies of plume size correlate (0.58) with anomalous Amazon flow at Obidos (Zeng et al., 2008) suggesting a significant impact from the Amazon, but leaving room for other mechanisms to influence plume area. Perhaps this interannual Amazon impact is not surprising since rainfall over the continent (e.g. Ropelewski & Halpert, 1987), and thus Amazon discharge, undergoes interannual and decadal changes including a 10% increase of discharge during La Niña years (e.g. Amarasekera, Lee, Williams, & Eltahir, 1997). In fact 2011 was the second year of the 2010–11 La Niña followed by weak El Nino in early 2012, and thus one might have expected some increase in Amazon discharge in 2011 relative to 2012. But, in fact, the previous year's La Niña conditions resulted in greater upper Amazon flow in 2012 (Satyamurty, da Costa, Manzi, & Candido, 2013) indicating that water storage can cause significant phase lags in the rainfall–discharge relationship (Chen, Wilson, & Tapley, 2010).
Amazon rainfall is also affected by the tropical Atlantic SST (e.g. Xie & Carton, 2004). In particular, one of the strongest Amazon discharge events in recent years, the 2009 flooding (Foltz, McPhaden, & Lumpkin, 2012), was triggered by anomalous cooling of SST in the equatorial north Atlantic and resulted in an anomalous southward shift of the ITCZ, severe flooding in northeast Brazil, and above normal Amazon discharge. Although relative changes in the Amazon discharge are not large (some 10%), their magnitude may well exceed the total discharge by any other Atlantic river.
Because the fresh Amazon/Orinoco plume covers a highly variable area that extends seasonally over 106 km2, the best way to monitor its variability is provided by satellite sensors as has been demonstrated using ocean color data (e.g. Hu et al., 2004). Recently two instruments, the soil moisture and ocean salinity (SMOS) (Boutin, Martin, Reverdin, Yin, & Gaillard, 2013) and the US/Argentina Aquarius/SACD (Lagerloef et al., 2012) have begun providing sea surface salinity (SSS) measurements from space, offering a more direct approach to track the plume characteristics (Tzortzi, Josey, Srokosz, & Gommenginger, 2013). Here we use observations from the Aquarius mission together with ancillary observations to identify a dramatic salinification of the plume from the summer–fall of 2011 to the summer–fall of 2012 and explore the potential causes of this change.
Section snippets
Data
The main SSS data set used in this study is the daily level 3 version 2.0 Aquarius SSS beginning 25 August, 2011, obtained from the NASA Jet Propulsion Laboratory Physical Oceanography Distributed Active Archive Center on a 1° × 1° grid. We compare Aquarius SSS with the ESA soil moisture and ocean salinity (SMOS) SSS derived by LOCEAN/IPSL for 2010–2012 (LOCEAN_v2013, reprocessed, Boutin et al., 2013, Yin et al., 2012). The observations are known to contain seasonally dependent errors which may
Results
Near-surface salinity along the ship of opportunity transects extending northeastward from French Guiana shows the seasonal growth of the low salinity plume northward from the coast into the region of high salinity subtropical water as the calendar year progresses (Fig. 2a). This northward extension develops coincident with the seasonal increase in freshwater discharge by the Amazon and surrounding rivers and with the relaxation of onshore winds at the Amazon mouth (Fig. 3b). The multi-year
Summary
New satellite SSS observations from the Aquarius/SACD satellite and in situ thermosalinograph measurements from commercial ships reveal significant year-to-year variations in the salinity of the Amazon/Orinoco plume, which was about 1 psu saltier in late summer–early fall of 2012 relative to 2011. The variability seen by Aquarius is also apparent in observations from the SMOS satellite (not shown). This increase in salinity is surprising because continental discharge was stronger in 2012 than in
Acknowledgments
This research was supported by NASA (NNX12AF68G, NNX09AF34G, and NNX10AO99G). GR was supported by CNRS with grants from CNES for SMOS validation studies. VC was supported by NSF (OCE-0933975) and the Gordon and Betty Moore Foundation. The continental discharge data are provided by the HYBAM observatory and the Brazil Water Agency. The GOSUD data used here are from the MN Colibri and Toucan, two vessels equipped and maintained by ORE SSS. We are grateful to Denis Diverres (US IMAGO, IRD) for
References (30)
- et al.
ENSO and the natural variability in the flow of tropical rivers
Journal of Hydrology
(1997) - et al.
The sea surface salinity in the tropical Atlantic between 10°S and 30°N — Seasonal and interannual variations (1977–1989)
Deep Sea Research I
(1994) - et al.
The dispersal of the Amazon and Orinoco River water in the tropical Atlantic and Caribbean Sea: Observation from space and S-PALACE floats
Deep Sea Research II
(2004) - et al.
Numerical modeling of the Amazon River plume
Continental Shelf Research
(2007) - et al.
The version-2 global precipitation climatology project (GPCP) monthly precipitation analysis (1979–present)
Journal of Hydrometeorology
(2003) - et al.
Ocean barrier layers' effect on tropical cyclone intensification
PNAS
(2012) - et al.
Gridded surface wind fields from Metop/ASCAT measurements
International Journal of Remote Sensing
(2012) - et al.
The Pirata Program: History, accomplishments, and future directions
Bulletin of the American Meteorological Society
(2008) - et al.
Sea surface freshening inferred from SMOS and ARGO salinity: Impact of rain
Ocean Science
(2013) - et al.
[Subsurface salinity] Global oceans [in State of the Climate in 2011]
Bulletin of the American Meteorological Society
(2012)
Estimates of the zonal slope and seasonal transport of the Atlantic North Equatorial Countercurrent
Journal of Geophysical Research
The 2009 exceptional Amazon flood and interannual terrestrial water storage change observed by GRACE
Water Resources Research
Amazon and Orinoco River plumes and NBC rings: Bystanders or participants in hurricane events?
Journal of Climate
Seasonal mixed layer salinity balance of the tropical North Atlantic Ocean
Journal of Geophysical Research
A strong Atlantic Meridional Mode event in 2009: The role of mixed layer dynamics
Journal of Climate
Cited by (72)
A tropical Atlantic dynamics analysis by combining machine learning and satellite data
2021, Advances in Space ResearchThermodynamic uptake of atmospheric CO<inf>2</inf> in the oligotrophic and semiarid São Francisco estuary (NE Brazil)
2021, Marine ChemistryCitation Excerpt :In large tropical estuaries and plumes such as the Amazon and the Congo, most of the salinity gradient occurs offshore where wind speed is higher. Gas exchange in these plumes can significantly alter the C-pCO2 mixing curves, and the impact of stronger wind and longer residence time in the mixing zone might be partially offset by a deeper mixed layer in the range of 5–30 m (Grodsky et al., 2014; Phillipson and Toumi, 2020). In August 2014, in the main channel of the São Francisco estuary, heating and tidal pumping were almost negligible.
CO<inf>2</inf> partial pressure and fluxes in the Amazon River plume using in situ and remote sensing data
2021, Continental Shelf ResearchOcean climatology at Brazilian Equatorial Margin: A numerical approach
2020, Journal of Computational ScienceCitation Excerpt :During the boreal Autumn and Winter, the lower salinity waters are located close to the coastline, mainly from December to February. The ARP confinement in the Winter was also reported by other authors and has been associated with the role of onshore winds [59,58]. During the boreal Spring, the plume is transported northwestward along the shelf and shelf break (Fig. 8b), following the NBC pathway [59,58].
Sea surface salinity estimates from spaceborne L-band radiometers: An overview of the first decade of observation (2010–2019)
2020, Remote Sensing of EnvironmentAmazon River plume influence on Western Tropical Atlantic dynamic variability
2019, Dynamics of Atmospheres and Oceans