q 2004 Geological Society of America. For permission to copy, contact Copyright Permissions, GSA, or editing@geosociety.org. Geology; July 2004; v. 32; no. 7; p. 597–600; doi: 10.1130/G20604.1; 3 figures; Data Repository item 2004095. 597 Phasing of deglacial warming and Laurentide Ice Sheet meltwater in the Gulf of Mexico Benjamin P. Flower College of Marine Science, University of South Florida, St. Petersburg, Florida 33701, USA David W. Hastings Collegium of Natural Sciences, Eckerd College, St. Petersburg, Florida 33711, USA Heather W. Hill Terrence M. Quinn    College of Marine Science, University of South Florida, St. Petersburg, Florida 33701, USA ABSTRACT Evidence is emerging that the tropical climate system played a major role in global climate change during the last deglaciation. However, existing studies show that deglacial warming was asyn- chronous across the tropical band, complicating the identification of causal mechanisms. The Orca Basin in the northern Gulf of Mexico is ideally located to record subtropical Atlantic sea-surface temperature (SST) warming in relation to meltwater input from the Laurentide Ice Sheet. Paired d 18 O and Mg/Ca data on the planktonic foraminifer Globigerinoides ruber from core EN32-PC6 are used to separate deglacial changes in SST and d 18 O of sea- water. SST as calculated from Mg/Ca data increased by .3 8C from ca. 17.2 to 15.5 ka in association with Heinrich event 1 and was not in phase with Greenland air temperature. Subtracting tem- perature effects from d 18 O values in G. ruber reveals two excur- sions representing Laurentide meltwater input to the Gulf of Mex- ico, one of .1.5‰ from ca. 16.1 to 15.6 ka and a second major spike of .2.5‰ from ca. 15.2 to 13.0 ka that encompassed melt- water pulse 1A and peaked ca. 13.8 ka during the Bølling-A ˚ llerød. Conversion to salinity through the use of a Laurentide meltwater end member of 225‰ indicates that near-surface salinity de- creased by 2‰–4‰ during these spikes. These results suggest that Gulf of Mexico SST warming preceded peak Laurentide Ice Sheet decay and the Bølling-A ˚ llerød interval by .2 k.y. and that heat was retained in the subtropical Atlantic during Heinrich event 1, consistent with modulation of deglacial climate by thermohaline circulation. Keywords: Gulf of Mexico, Orca Basin, Laurentide Ice Sheet, degla- ciation, meltwater floods, freshwater switching. INTRODUCTION Evidence is emerging that the low-latitude climate system may have been an important driver of deglacial climate change, through its influence on poleward heat and vapor transport (e.g., Cane and Clem- ent, 1999). One key question is, did tropical sea-surface temperature (SST) warming lead or coincide with the onset of Bølling warming recorded in Greenland ice-core records? In the western tropical Atlantic (Fig. 1), the SST increase appears to have coincided with Bølling warming, on the basis of Cariaco Basin sediment records (Lea et al., 2003) and Barbados coral records (Guilderson et al., 2001). In contrast, the SST increase seems to have led Bølling warming in the Tobago Basin (Ruhlemann et al., 1999; Huls and Zahn, 2000). Indeed, SST may have been out of phase with Greenland air temperature throughout the deglaciation in this area. This relationship has been compared to the antiphasing of Greenland and Antarctic temperature (Blunier and Brook, 2001) and attributed to the ‘‘bipolar seesaw’’ model of deglacial climate change (Broecker, 1998). Therefore, substantial disagreement exists on the timing and possible causes of tropical Atlantic SST chang- es during deglaciation. Resolving the temporal and spatial pattern of deglacial SST variability has important implications for modulation of climate by thermohaline circulation and greenhouse gas levels. The timing and magnitude of Laurentide meltwater input to the Gulf of Mexico and/or meltwater diversion to the North Atlantic via the St. Lawrence River and other eastern outlets may have influenced North Atlantic Deep Water (NADW) formation and hence regional to global climate. In particular, low-density meltwater may have inhibited deep convection in the North Atlantic and reduced northward heat transport, leading to regional cooling during the Younger Dryas ca. 12.8–11.2 ka (e.g., Broecker et al., 1989; Keigwin et al., 1991). This is a topic of renewed interest because of evidence linking eastward meltwater discharge to NADW reduction since ca. 22 ka (Licciardi et al., 1999; Clark et al., 2001). During the deglacial interval, meltwater flow may have switched rapidly between the Mississippi River drain- age and the eastern drainage systems, triggering episodic cooling and warming in the North Atlantic region (Clark et al., 2001). Orca Basin is a small (400 km 2 ) anoxic basin located ;300 km south of Louisiana (Fig. 1) that is ideally located to record deglacial meltwater input through the Mississippi River system. A 200-m-thick hypersaline brine (salinity .250‰) limits oxygen mixing and diffusion and allows fine laminations to be preserved. Abundant aragonitic ptero- pods, planktonic foraminifera with intact spines, and fossil seaweed (Sargassum) with attached calcareous polychaetes attest to minimal carbonate dissolution. The entire Gulf of Mexico is within the Western Hemisphere Warm Pool (WHWP) of .28.5 8C water that develops during the Northern Hemisphere late summer and that represents the warmest water in the low-latitude Atlantic and eastern North Pacific (Fig. 1; Wang and Enfield, 2001). Oxygen isotope data on Globigerinoides ruber from Orca Basin sediment cores have previously been used to document Laurentide meltwater input during the last deglaciation (Leventer et al., 1982; Flower and Kennett, 1990). Attempts to separate Gulf of Mexico SSTs and d 18 O values of seawater (d 18 O sw ) have been hampered by the lack of an independent SST record. Here we describe new Mg/Ca data gen- erated on G. ruber (white variety predominantly) from core EN32-PC6 (26856.89N; 91820.09W) that provide such an SST record. The Mg/Ca and d 18 O data on G. ruber from the same samples (Fig. 2) as well as global d 18 O sw constraints from sea-level records (Fairbanks, 1989; Bard et al., 1990) help document the phasing of SST and d 18 O sw var- iations in the Gulf of Mexico. METHODS Picked planktonic foraminiferal specimens were cleaned for Mg/ Ca analysis following established procedures (Barker et al., 2003; also see Appendix 1 ). Foraminiferal Mg and Ca were analyzed on a Perkin Elmer Optima 4300 dual view inductively coupled plasma–optical 1 GSA Data Repository item 2004095, Appendix, Table DR1 (radiocarbon dates for EN32-PC6), and Table DR2 (Mg/Ca and d 18 O data), is available online at www.geosociety.org/pubs/ft2004.htm, or on request from editing@ geosociety.org or Documents Secretary, GSA, P.O. Box 9140, Boulder, CO 80301-9140, USA. Supporting data are also available electronically at World Data Center-A for Paleoclimatology, NOAA/NGDC, 325 Broadway, Boulder, CO 80305, USA, phone: (303) 457-6513; e-mail: paleo@mail.ngdc.noaa.gov; URL: http://www.ngdc.noaa.gov/paleo. on September 18, 2015 geology.gsapubs.org Downloaded from