Interacting Loop Current variability and Mississippi River discharge over the past 400 kyr

https://doi.org/10.1016/j.epsl.2008.04.051Get rights and content

Abstract

The Loop Current mediating the oceanic heat and salt flux from the Caribbean Sea into the Atlantic Ocean and its interference with the Mississippi River discharge are critical for both the regional climate in the Gulf of Mexico area and the water vapor transport towards high northern latitudes. We present a 400-kyr record of sea surface temperature and local surface salinity from the northeastern Gulf of Mexico (IMAGES core MD02-2575) approximated from combined planktonic foraminiferal δ18O and Mg/Ca, which reflects the temporal dynamics of the Loop Current and its relationship to both varying Mississippi discharge and evolution of the Western Hemisphere Warm pool. The reconstructed sea surface temperature and salinity reveal glacial/interglacial amplitudes that are significantly larger than in the Western Hemisphere Warm pool. Sea surface freshening is observed during the extreme cool periods of Marine Isotope Stages 2, 8, and 10, caused by the strengthened Mississippi discharge which spread widely across the Gulf favored by the less established Loop Current. Interglacial and interstadial sea-surface conditions, instead, point to a strengthened, northward flowing Loop Current in line with the northward position of the Intertropical Convergence Zone, allowing northeastern Gulf of Mexico surface hydrographic conditions to approach those of the Caribbean. At these times, the Mississippi discharge was low and deflected westward, promoted by the extended Loop Current. Previously described deglacial megadischarge events further to the west did not affect the northeastern Gulf of Mexico hydrography, implying that meltwater routing from the Laurentide Ice Sheet via the Mississippi River is unlikely to have affected Atlantic Meridional Overturning Circulation.

Introduction

The tropical Western Hemisphere warm pool (WHWP) is an important heat and moisture source for climate in the North Atlantic region and hence, acts as a key regulator of the subpolar North Atlantic oceanography and climate in NW-Europe. Marginal to the WHWP is the Gulf of Mexico (GOM), whose thermal variability affects the amount of water vapor exported and the intensity of storm tracks in the region (Oglesby et al., 1989). Long-term hydrographic changes in the GOM may thus have had a profound effect on both regional climate and moisture transport towards high northern latitudes ranging from orbital to sub-millennial time scales.

The Loop Current (LC) is the most prominent surface circulation feature in the GOM, clearly expressed in sea-surface height (e.g., Sturges and Evans, 1983). Although causation is not well understood yet, coherencies between LC retraction and extension, seasonal migrations of the Intertropical Convergence Zone (ITCZ), the related wind system, and changes in the thermohaline circulation are corroborated by a model simulation forced by a seasonal wind climatology and a meridional overturning cell (Johns et al., 2002), which denotes both a purely wind-driven Caribbean inflow through the Caribbean Passages and an overturning-related inflow. During fall/winter, the LC is at a minimum (~ 26 Sv; 1 Sv = 106m3s- 1; Molinari et al., 1990, Johns et al., 2002) and describes a direct flow from the Yucatan Channel to the Florida Straits, thus leaving the northern GOM unaffected by warm tropical surface water from the Caribbean.

A characteristic feature of the LC is the shedding of anticyclonic eddies, which are generated aperiodically every 3 to 21 months, with an average shedding time of 9.5 months (Zavala-Hidalgo et al., 2006). The detachment of an eddy from the Yucatan Strait throughflow may take weeks, and often a developing eddy reattaches to the LC (Sturges and Leben, 2000). Although the process of eddy shedding can be simulated by ocean models (e.g. Romanou et al., 2004, Zavala-Hidalgo et al., 2006), the forcing is hardly understood and appears to be related to a suite of oceanographic forcing fields such as the Yucatan Channel inflow (Ezer et al., 2003, Oey et al., 2003, Oey, 2004), the Florida Current and North Brazil Current variability, as well as synoptic meteorological forcing variability (Romanou et al., 2004). In spite of the complexity of eddy shedding, the spatial extension of the LC and its northward propagation appear to broadly reflect an annual cycle. Sturges and Evans (1983) mentioned annual variations of the LC, although they found as much power at periods near 30 months as at periods near 12 months. Lee and Mellor (2003) and Zavala-Hidalgo et al. (2006) showing timeseries of the areal extent and maximum latitude of the LC intrusion imply seasonality although overprinted by other cycles. Even if the complex eddy shedding process takes place on a non-seasonal cycle, eddies being generated during summer will be warmer than those being generated during winter. The net heat flux towards the northern Gulf of Mexico during summer – even at a lower number of eddies – most presumably exceeds that of the winter season.

Indeed, the average seasonal signal in near-surface temperature and horizontal velocity measured by near-surface drifters clearly shows that the eastern gulf area – an area which is characterized by highest surface current velocities related to the northward propagating LC – starts to warm during spring (http://oceancurrents.rsmas.miami.edu). Later in season, the warmth spreads from the (north)eastern areas towards the west. Cooling, in contrast, starts in the western gulf during fall and propagates eastward, in close relationship to the retraction of the LC. The thermal evolution of the (north)eastern Gulf of Mexico is hence not simply related to surface heat flux in line with the seasonal development of the WHWP, but is largely affected by LC dynamics.

Apart from the distinct role of the LC, fluvial freshwater discharge also influences the GOM surface hydrography. In particular, the ~ 3700 km long Mississippi draining an area from Lake Itasca in the northern USA down to Florida plays the major role with an average annual discharge at Baton Rouge (Louisiana) of ~ 13.500 m3/s (Morey et al., 2003). Together, LC flow variability and Mississippi River discharge combine to influence sea surface temperature (SST) and salinity (SSS) patterns in the GOM. Today, the annual mean SST in the DeSoto Canyon area is ~ 24.5 °C exhibiting a seasonal amplitude of ~ 4 °C. The average SSS is ~ 34.7 (Levitus and Boyer, 1994), hence slightly lower than in the western GOM (~ 35.5). It's seasonal variability from ~ 32.6–35.6 is instead larger than in the western GOM (~ 33.9-36.2), most likely due to the varying influence of the LC, precipitation in the eastern part, and Mississippi River discharge (Meade, 1995). In fact, the preferential westward routing of Mississippi freshwater during summer may be re-directed to the east by prevailing westerly winds leading to lowered salinities in the (north)eastern GOM (Morey et al., 2003).

Our paleoceanographic study focuses on SST and the stable oxygen isotope composition of seawater (δ18OSW, approximating SSS) reconstructions over the last 400 kyr, based on sediment core MD02-2575 from the DeSoto Canyon in the northeastern GOM. The observed temporal variations in these ocean properties reveal how LC and Mississippi discharge interacted, how closely Gulf of Mexico and central Caribbean surface hydrographies were related, and how LC dynamics relate to meridional overturning circulation and ITCZ migration.

Section snippets

Core selection and chronostratigraphy

IMAGES sediment core MD02-2575 was recovered from the northeastern GOM (DeSoto Canyon, 29°00.10′N 87°07.13′W) close to ODP Site 625 (Joyce et al., 1993) and core GS7109-2 (Emiliani, 1975) (Fig. 1). The DeSoto Canyon is an erosional valley cutting through the broad continental shelf of the northern Gulf. Sedimentation on the western Florida shelf is dominated by foraminifer/coccolith ooze accumulation (Blake and Doyle, 1983). The sediment core lies well within the present day Antarctic

Gulf of Mexico vs. Caribbean sea surface hydrographic changes

The δ18O values for the surface-dwelling foraminifer G. ruber18OG.ruber) range from ~ 1.1 to ~− 2.0‰ within core MD02-2575, showing pronounced glacial/interglacial changes (Fig. 2). When comparing the temporal δ18OG.ruber variations to δ18OG.ruber records from Orca Basin (Flower et al., 2004), the Louisiana slope (Aharon, 2003) and the Columbia Basin (western Caribbean) (Schmidt et al., 2004, Schmidt et al., 2006), spatial and temporal differences, in particular during the last deglaciation,

Conclusions

Our proxy records of core MD02-2575 reveal that the northeastern GOM is an ideal place to decipher the dynamic evolution of the LC and its relationship to Mississippi River discharge on orbital timescales. The SSTMg/Ca and Δδ18Oivf–sw variability in the northeastern GOM is much larger than in the WHWP and different from Orca Basin. During extreme cool periods, our data suggest fresher sea surface conditions, most likely resulting from a less established LC and a strengthened Mississippi River

Acknowledgements

Funding of this research was provided by the German Science Foundation (DFG) within the authors' projects Nu60/8 and Ti240/12. We thank the Shipboard Scientific Party and crew of RV Marion Dufrèsne cruise MD-127 (Page) in 2002 for their kind support in the framework of IMAGES. We are grateful for scientific and technical support by Carsten Rühlemann, Joachim Schönfeld, Jeroen Groeneveld, Dieter Garbe-Schönberg, Nadine Gehre, and Lulzim Haxhiaj. Critical comments of three reviewers are

References (84)

  • AharonP.

    Meltwater flooding events in the Gulf of Mexico revisited: implications for rapid climate changes during the last deglaciation

    Paleoceanography

    (2003)
  • AnandP. et al.

    Calibration of Mg/Ca thermometry in planktonic foraminifera from a sediment trap time series

    Paleoceanography

    (2003)
  • AndersonD.M. et al.

    Glacial-interglacial stability of ocean pH inferred from foraminifer dissolution rates

    Nature

    (2002)
  • ArcherD.E.

    An atlas of the distribution of calcium carbonate in sediments of the deep sea

    Glob. Biogeochem. Cycles

    (1996)
  • BardE.

    Correction of accelerator mass spectrometry 14C ages measured in planktonic foraminifera: paleoceanographic implications

    Paleoceanography

    (1988)
  • BarkerS. et al.

    Foraminiferal calcification response to glacial-interglacial changes in atmospheric CO2

    Science

    (2002)
  • BarkerS. et al.

    A study of cleaning procedures used for foraminiferal Mg/Ca paleothermometry

    Geochem. Geophys. Geosyst.

    (2003)
  • BergerW.H.

    Increase of carbon dioxide in the atmosphere during deglaciation: the coral reef hypothesis

    Naturwissenschaften

    (1982)
  • Blake, N.J., Doyle, L.J., 1983. Infaunal-sediment relationships at the shelf-slope break: the shelfbreak: critical...
  • BroeckerW.S. et al.

    Carbonate ion concentration in glacial-age deep waters of the Caribbean Sea

    Geochem. Geophys. Geosyst.

    (2002)
  • ClarkP.U. et al.

    Freshwater forcing of abrupt climate change during the last deglaciation

    Science

    (2001)
  • CurryW.B. et al.

    Synchronous high-frequency oscillations in tropical sea surface temperatures and North Atlantic deep water production during the last glacial cycle

    Paleoceanography

    (1997)
  • DekensP.S. et al.

    Core top calibration of Mg/Ca in tropical foraminifera: refining paleotemperature estimation

    Geochem. Geophys. Geosyst.

    (2002)
  • DroxlerA.W. et al.

    The marine carbonate system during oxygen isotope stage 11 (423–362 ka): a case of basin-to-shelf and/or basin-to-basin carbonate fractionation?

    EOS, Trans. Am. Geophys. Union

    (1997)
  • DroxlerA.W. et al.

    Youthful belize Barrier Reef: Strengthening the model for a mid-Brunhes global establishment of modern barrier reefs

    EOS, Trans. Am. Geophys. Union

    (2002)
  • EmilianiC.

    Paleoclimatological analysis of Late Quaternary cores from the northeastern Gulf of Mexico

    Science

    (1975)
  • EzerT. et al.

    The variability of currents in the Yucatan Channel: analysis of results from a numerical ocean model

    J. Geophys. Res.

    (2003)
  • FairbanksR.G.

    A 17,000-year glacio-eustatic sea level record: influence of glacial melting rates on the Younger Dryas event and deep-ocean circulation

    Nature

    (1989)
  • FlowerB.P. et al.

    Phasing of deglacial warming and Laurentide Ice Sheet meltwater in the Gulf of Mexico

    Geology

    (2004)
  • GardulskiA.F. et al.

    Carbonate mineral cycles in ramp slope sediment: Eastern Gulf of Mexico

    Paleoceanography

    (1986)
  • GonthierG.J. et al.

    Groundwater sources and flow patterns derived from stable isotopes and elemental chemistry in the southeast Louisiana freshwater aquifers

    Bull. Am. Assoc. Petrol. Geol.

    (1990)
  • GustavssonN. et al.

    Geochemical landscapes of the conterminous United States — New map presentations for 22 elements

  • HaddadG.A. et al.

    Metastable CaCO3 dissolution at intermediate water depths of the Caribbean and western North Atlantic: implications form intermediate water circulation during the past 200,000 years

    Paleoceanography

    (1996)
  • HughenK.A. et al.

    Deglacial changes in ocean circulation from an extended radiocarbon calibration

    Nature

    (1998)
  • HughenK.A. et al.

    Marine04 marine radiocarbon age calibration, 0–26 cal kyr BP

    Radiocarbon

    (2004)
  • HughenK.A. et al.

    C-14 activity and global carbon cycle changes over the past 50,000 years

    Science

    (2004)
  • JoyceJ.E. et al.

    North American glacial meltwater history for the past 2.3 m.y.: oxygen isotope evidence from the Gulf of Mexico

    Geology

    (1993)
  • KimS.-J. et al.

    A coupled climate model simulation of the Last Glacial Maximum, Part 2: approach to equilibrium

    Clim. Dyn.

    (2003)
  • LeaD.W. et al.

    Climate impact of Late Quaternary equatorial Pacific sea surface temperature variations

    Science

    (2000)
  • LeeH.-C. et al.

    Numerical simulation of the Gulf Stream System: the Loop Current and the deep circulation

    J. Geophys. Res.

    (2003)
  • Levitus, S., Boyer, T.P., 1994. World Ocean Atlas 1994: US-Department of Commerce, NOAA Atlas NESDIS, National Oceanic...
  • LinH.L. et al.

    Late Quaternary climate change from δ18O records of multiple species of planktonic foraminifera: high resolution records from the anoxic Cariaco Basin, Venezuela

    Paleoceanography

    (1997)
  • Cited by (82)

    View all citing articles on Scopus
    View full text