Auxiliary Text 
S.1. Core Location and Hydrography Piston core HU87033-017 (54.62 degree N, 56.18 degree W, 514 m water depth) was raised from the Cartwright Saddle in the path of the Labrador Current (Figure 1). The proximity of HU87033-017 to the Hudson Strait outlet and Labrador coast should result in a robust record of early Holocene Laurentide Ice Sheet discharge into the Labrador Sea [Andrews et al., 1999]. Modern water mass properties measured below ~50 m water depth show little variability in density, salinity and temperature throughout the year, placing the pycnocline at ~50 m water depth (Figure S1) [Hillaire-Marcel et al., 1992; Khatiwala et al., 1999; Antonov et al., 2010; Locarnini et al., 2010]. Monthly World Ocean Atlas 2009 grid data show the warmest waters proximal to HU87033-017 in the upper 50 m forming during the late summer, with surface values in July and August reaching 5-7 degree C (Figure S1a). Surface temperatures then cool through November, approaching ~0 degree C. Sea-ice cover from December to February lowers surface temperatures to a minimum of approximately -2°C. Surface temperatures remain close to 0°C through the early spring until summer warming. Below 50 m water depth, monthly water temperatures range from  -2 to 1°C [Locarnini et al., 2010].  Upper water column (<50 m) salinity reaches a minimum of ~30 practical salinity units (psu) during the late summer (Figure S1b). Surface salinity increases during autumn by ~1 psu and reaches maxima of ~33 psu during the winter months. Below 50 m water depth, salinity ranges between 32 and 33 psu with minimal seasonal variation [Antonov et al., 2010]. Calculated upper water column density is at a minimum (~1024.4 kg m^-3 at the surface, ~1026 kg m^-3 at 50 m water depth) and stratification at maximum during the late summer (Figure S1c). Upper water column density gradually increases into autumn with cooling water temperature and increasing salinity, reaching a maximum (~1026.5 kg m^-3) from December to February [Antonov et al., 2010; Locarnini et al., 2010].   S.2. Neogloboquadrina pachyderma (sinistral) Habitat Neogloboquadrina pachyderma (s) is the dominant planktonic foraminifera species in subpolar to polar latitudes [Tolderlund and Bé, 1971; Bé et al., 1977; Wu and Hillaire-Marcel, 1994; Bauch et al., 1999; Jonkers et al., 2010]. In the northwest Atlantic, N. pachyderma (s) reaches production peaks during seasonal phytoplankton blooms and maximum water column temperature stratification (late summer) with minimal test flux during the winter months [Wu and Hillaire-Marcel, 1994; Jonkers et al., 2010]. In the Labrador Sea, N. pachyderma (s) is observed inhabiting the upper 200 m of the water column and generally calcifies at the pycnocline boundary in waters ≤10 degree C [Wu and Hillaire-Marcel, 1994; Jonkers et al., 2010] making it a reliable recorder of near-surface water temperature and delta 18O variability [Wu and Hillaire-Marcel, 1994; Bauch et al., 1999; Jonkers et al., 2010]. In waters warmer than 8-10 degree C, N. pachyderma (s) may change its depth habitat and thus not record temperature changes above 10 degree C [Wu and Hillaire-Marcel, 1994; Jonkers et al., 2010]. At the core site of HU87033-017, the modern pycnocline is at ~50 m water depth and the overlying water mass has a seasonal range of -2 to 7 degree C, with the warmest temperatures occurring during July and August (Figure S1a) [Locarnini et al., 2010]. N. pachyderma (s) therefore likely calcifies and records late summer water temperature and δ18O of seawater of <50 m water depth [Wu and Hillaire-Marcel, 1994; Jonkers et al., 2010].  S.3. Core Age Model Our age model for HU87033-017 is derived from existing 14C dates from the core of Andrews et al. [1999]. We applied a reservoir age of 450 years to the 14C dates and recalibrated the ages using the INTCAL09 calibration (Table S1), following Andrews et al. [1999]. As discussed in the text, an additional reservoir age of 310 years was applied to the two 14C dates from the Lake Agassiz drainage layer to account for the influence of Paleozoic carbonate on Lake Agassiz runoff 14C/12C [Barber et al., 1999]. The age model was developed by linear interpolation between 14C-dated horizons and shown in Figure 2a. Because of the low concentration of planktonic foraminifera, the Andrews et al. [1999] 14C dates were obtained upon benthic foraminifera tests, which may be subject to older reservoir ages that would shift our chronology younger. However, the shallow core site precludes North Atlantic deep waters from influencing the 14C/12C ratios of the benthic tests [Robinson et al., 2005; Thornalley et al., 2011]. We also apply a slightly greater reservoir age of 450 years [Andrews et al., 1999] than the standard 400 years for the surface North Atlantic [Thornalley et al., 2011] to account for any potential surface-514 m water depth offset in 14C/12C. Furthermore, the main results of our study do not directly depend on the age model or choice of reservoir age because the core stratigraphy (see text) clearly identifies the Lake Agassiz drainage layer, in which we document a significant decrease in N. pachyderma (s) Mg/Ca and in the δ18O of seawater (Figure 2).  S.4. Mg/Ca Methodology We sampled HU87033-017 at ~20 cm intervals in 2 cm sections for elemental analysis (Table S2 provides mid-point depths of samples). The silt/clay fraction (<63 microns) was disaggregated from the sand fraction by sieving in a 5% sodium hexametaphosphate solution. The 150-250 micron fraction was isolated from the sand fraction and 5 to 30 N. pachyderma (s) tests were picked from each interval based on test abundance. Elemental analysis was conducted at the W.M. Keck Laboratory for Plasma Spectrometry at Oregon State University. The flow-through time-resolved analysis (FT-TRA) technique [Haley and Klinkhammer, 2002; Benway et al., 2003; Klinkhammer et al., 2004] employed to produce our Mg/Ca-derived calcification temperature (CT) record reduces Mg/Ca uncertainty introduced by clay impurities, partial dissolution and secondary calcification of specimen tests. Remnant clay deposits and other contaminant detritus remaining in foraminifer tests from insufficient cleaning can lead to higher (up to 1,000×) Mg/Ca ratios relative to biogenic calcite [Benway et al., 2003; Klinkhammer et al., 2004]. FT-TRA allows for the identification and rejection of these contaminant phases through high temporal resolution analyses [Benway et al., 2003; Klinkhammer et al., 2004]. We use the BeaverDAMM program to isolate the pure biogenic portion of the tests for calculating Mg/Ca ratios [Benway et al., 2003; Klinkhammer et al., 2004]. Mg/Ca ratios were converted to CT following the empirical N. pachyderma (s) species calibration developed by Kozdon et al. [2009] from Nordic Seas core top samples [Nürnberg, 1995]: 
Mg/Ca=0.13(+/-0.037)CT+0.35(+/-0.17)		(1) 
where Mg/Ca is in mmol/mol and CT is in degree C (Table S2).  
S.5. Calculation of Seawater δ18O To calculate seawater delta 18O (delta 18Osw), we first remove the effect of continental ice volume on the foraminiferal delta 18O of foraminiferal calcite (delta 18Oc) record of Andrews et al. [1999] following Fairbanks [1989] and Clark and Mix [2002]. We then remove the temperature effect on foraminiferal delta 18Oc using the cold-water foraminifera calibration of Shackleton [1974]: 
delta 18Osw=delta 18Oc+0.27-(4.38-sqrt(4.382-0.4(16.9-CT)))/0.2		(2) 
wwhere delta 18Oc is the existing record from Andrews et al. [1999] converted from VPDB to standard mean ocean water (SMOW) scale and CT is calcification temperature in degree C. We note that equation (2) gives the same delta 18Osw values as other calibrations within the uncertainties of delta 18Osw [Wu and Hillaire-Marcel, 1994; Bemis et al., 1998]. Although vital effects on delta 18Oc have been observed for some foraminifera taxa [Rohling and Cooke, 2002], N. pachyderma (s) tests from the North Atlantic core tops and plankton traps have been shown to reliably record delta 18Oc in equilibrium with delta 18Osw and temperature, suggesting that vital effects have a minimal impact on this species [Wu and Hillaire-Marcel, 1994; Jonkers et al., 2010]. The sampling resolution of the Andrews et al. [1999] delta 18Oc record is at a higher resolution than our Mg/Ca record. In addition, several of our Mg/Ca samples were collected from HU87033-017 adjacent to the Andrews et al. [1999] delta 18Oc samples because of limited remaining core material. To account for this difference in resolution, we thus sampled the ice-volume corrected delta 18Oc and Mg/Ca records at 100-year resolution (the average sampling resolution of the Mg/Ca record) between 6.9 and 11.5 ka and calculated delta 18Osw with equation (2) (Table S3), which is plotted in Figures 2 and 3.  S.6. Propagating Uncertainty FT-TRA measurement of foraminifera test Mg/Ca has an average Mg/Ca uncertainty of +/-0.08 mmol/mol from varying Mg/Ca during analysis [Klinkhammer et al., 2004], resulting in CT analytical uncertainty of +/-0.6 degree C. Including the calibration uncertainty of +/-1.1 degree C in equation (1) [Kozdon et al., 2009] results in a propagated CT uncertainty of +/- 1.3 degree C. With the delta 18Oc uncertainty of the Andrews et al. [1999] record of +/-0.05 per mil, the propagated uncertainty in delta 18Osw is +/-0.3 per mil.  S.7. Near Surface Samples Modern N. pachyderma (s) Mg/Ca should record near-surface temperatures at Cartwright Saddle. Unfortunately, the core top for HU87033-017 was not retrieved (Table S1) [Andrews et al., 1999]. Our youngest (~0.9 ka, 49 cm core depth) Mg/Ca-derived CT of ~5 degree C is in good agreement with the observed late summer temperature of 4-6 degree C at ≤50 m water depth (Figure S1a), which also agrees with the timing of peak N. pachyderma (s) production [Tolderlund and Be, 1971; Jonkers et al., 2010]. Our next youngest sample from ~3.1 ka (101.5 cm core depth) has a CT of ~5 degree C, also agreeing with our inferred season and habitat for N. pachyderma (s) at the core site. In addition, the delta 18Oc from the same depths [Andrews et al., 1999] agrees with the density-predicted depth profile of delta 18Oc from precipitation in isotopic equilibrium [Hillaire-Marcel et al., 2001], further confirming that N. pachyderma (s) is living in the late summer at water depth of <50 m (Figure S1d). Thus, the agreement between our late Holocene CTs, the inferred seasonal temperature that the test is recording in the modern, and the timing of modern N. pachyderma (s) blooms supports our use of Mg/Ca as a paleo-temperature proxy in the western Labrador Sea. It also validates the use of the Kozdon et al. [2009] Mg/Ca-CT calibration in the western Labrador Sea.   S.8. Salinity and Carbonate-Ion Effects on Foraminiferal Mg/Ca Calibration experiments specific to the planktonic foraminifers Globigerinoides sacculifer, Globigerinoides ruber and Globigerinoides bulloides have determined that salinity and carbonate ion concentration affect test Mg/Ca but are secondary to the primary influence of temperature on test Mg uptake, with little effect on tests calcified in lower salinity waters like our study site [Lea et al., 1999; Russell et al., 2004; Bentov and Erez, 2006; Kisakürek et al., 2008; Hoogakker et al., 2009]. In the modern Labrador Sea, the minimal salinity variability in near-surface waters near HU87033-017 [Antonov et al., 2010] (Figure S1b) suggests that salinity should have an equally minimal effect on N. pachyderma (s) Mg/Ca. During the early Holocene, Labrador Sea salinity near the Laurentide Ice Sheet was likely lower than present [Hillaire-Marcel et al., 2001, 2008; Carlson et al., 2008], so we may rule out a salinity increase affecting Mg/Ca. Early Holocene decreases in salinity, however, could have affected the Mg/Ca record. For our most extreme Mg/Ca decrease observed during Lake Agassiz drainage, salinity would have had to have reduced by >8 psu [Lea et al., 1999], which is impossible [Bé et al., 1977; Jonkers et al., 2010] given that N. pachyderma (s) tests are still found during the event, and means a CT decrease is the main cause of the Mg/Ca decrease. We also note that the effects of salinity are only significant in environments where salinity is ≥36 psu [Lea et al., 1999; Kisakürek et al., 2008; Hoogakker et al., 2009], which is not the case of the HU87033-017 core site (Figure S1b). Thus, our record of Mg/Ca is mainly recording changes in CT rather than salinity.   Increased carbonate ion concentration results in the decreased uptake of Mg into foraminifer tests due to an enhancement of the calcium carbonate saturation state, producing anomalously low Mg/Ca ratios [Lea et al., 1999; Lea, 2003; Russell et al., 2004; Barker et al., 2005; Kisakürek et al., 2008]. The Andrews et al. [1999] sedimentary record identifies a peak of detrital carbonate deposition coincident with the drainage of Lake Agassiz, which could result in an increase in carbonate ion concentration near HU87033-017. However, Lake Agassiz drained subglacially, with the detrital carbonate transported to Cartwright Saddle likely in a hyperpycnal flow [Andrews et al., 1995, 1999; Kerwin et al., 1996; Lajeunesse and St. Onge, 2008; Clarke et al., 2009; Roy et al., 2011], which would not affect the carbonate ion concentration of pycnocline waters. The effect of carbonate ion concentration was also shown to have a minimal impact on N. pachyderma (s) test geochemistry in Hudson Strait [Carlson et al., 2009]. The Lake Agassiz runoff derived from the underlying Paleozoic carbonate terranes could also have a higher Mg/Ca ratio to the Precambrian shield adjacent to Cartwright Saddle, which could increase N. pachyderma (s) test Mg/Ca ratios [Carlson et al., 2007]. Because our Mg/Ca ratios decrease during the Lake Agassiz drainage event, this effect would only act to reduce our calculated CT cooling, thus muting the δ18Osw signal and dampening our inferred salinity decrease.   
Auxiliary References Andrews, J.T., L. Keigwin, F. Hall, and A.E. Jennings (1999), Abrupt deglaciation events and Holocene palaeoceanography from high-resolution cores, Cartwright Saddle, Labrador Shelf, Canada, J. Quat. Sci., 14, 383-397. Antonov, J. I., D. Seidov, T. P. Boyer, R. A. Locarnini, A. V. Mishonov, H. E. Garcia, O. K. Baranova, M. M. Zweng, and D. R. Johnson (2010), World Ocean Atlas 2009, Volume 2: Salinity, S. Levitus, Ed. NOAA Atlas NESDIS 69, U.S. Government Printing Office, Washington, D.C.  Barber, D.C., A. Dyke, C. Hillaire-Marcel, A.E. Jennings, J.T. Andrews, M.W. Kerwin, G. Bilodeau, R. McNeely, J. Southon, M.D. Morehead, and J.-M. Gagnon (1999), Forcing of the cold event of 8,200 years ago by catastrophic drainage of Laurentide lakes, Nature, 400, 344-348. Barker, S., I. Cacho, H.M. Benway, and K. Tachikawa (2005), Planktonic foraminiferal Mg/Ca as a proxy for past oceanic temperatures: A methodological review and data compilation for the Last Glacial Maximum, Quat. Sci. Rev., 24, 821-834.  Bauch, H.A., H. Erlenkeuser, K. Fahl, R.F. Spielhagen, M.S. Weinelt, H. Andruleit, and R. Henrich (1999), Evidence for a steeper Eemian than Holocene sea surface temperature gradient between Artic and sub-Artic regions, Palaeogeogr. Palaeoclimatol. Palaeoecol., 145, 95-117.  Bé, A.W.H., M. Hemleben, M. Spindler, J. Hacunda, and S. Tuntivate-Choy (1977), Laboratory and Field Observations of Living Planktonic Foraminifera, Micropaleontology, 23, 155-179. Bemis, B.E., H.J. Spero, J. Bijma, and D.W. Lea (1998), Reevaluation of the oxygen isotopic composition of planktonic foraminifera: Experimental results and revised paleotemperature equations, Paleoceanography, 13, 150-160.   Bentov, S., and J. Erez (2006), Impact of biomineralization processes on the Mg content of foraminiferal shells: A biological perspective, Geochem. Geophys. Geosys., 7, doi: 10.1029/2005GC001015.  Benway, H.M., B.A. Haley, G.P. Klinkhammer, and A.C. Mix (2003), Adaptation of a flow-through leaching procedure for Mg/Ca paleothermometry, Geochem. Geophys. Geosys., 4, doi: 10.1029/2002GC000312. Came, R. E., D. W. Oppo, and J. F. McManus (2007), Amplitude and timing of temperature and salinity variability in the subpolar North Atlantic over the last 10000 years, Geology, 35, 315-318. Carlson, A.E., A.N. LeGrande, D.W. Oppo, R.E. Came, G.A. Schmidt, F.S. Anslow, J.M. Licciardi, and Obbink, E.A. (2008), Rapid early Holocene deglaciation of the Laurentide ice sheet, Nature Geosci., 1, 620-624. Carlson, A.E., P.U. Clark, B.A. Haley, G.P. Klinkhammer, K. Simmons, E.J. Brook, and K.J. Meissner (2007), Geochemical proxies of North American freshwater routing during the Younger Dryas cold event, Proceed. Nat. Acad. Sci., 104, 6556 -6561. Carlson, A.E., P.U. Clark, B.A. Haley, and G.P. Klinkhammer (2009), Routing of western Canadian Plains runoff during the 8.2 ka cold event, Geophys. Res. Lett., 36, doi: 10.1029/2009GL038778.  Clark, P.U., and A.C. Mix (2002), Ice sheets and sea level of the Last Glacial Maximum, Quat. Sci. Rev., 21, 1-7. Clarke, G. K. C., A. B. G. Bush, and J. W. M. Bush (2009), Freshwater discharge, sediment transport, and modeled climate impacts of the final drainage of glacial Lake Agassiz, J. Climate, 22, 2161-2180. Ellison, C.R.W., M.R. Chapman, and I.R. Hall, Surface and Deep Ocean Interactions During the Cold Climate Event 8200 Years Ago, Science, 312, 1929-1932.  Fairbanks, R.G. (1989), A 17,000-year glacio-eustatic sea level record: influence of glacial melting rates on the Younger Dryas event and deep-ocean circulation, Nature 342, 637-642.  Haley, B.A., and G.P. Klinkhammer (2002), Development of a flow-through system for cleaning and dissolving foraminiferal tests, Chem. Geol. 185, 51-69. Hillaire-Marcel, C., A. de Vernal, S. Vallibres, and on-board participants (1992), Cruise report and on-board studies, CSS-Hudson 91-045, the Labrador Sea, the Irminger and Iceland basins, Geological Survey of Canada, Open File Report, Bedford Institute of Oceanography, Dartmouth (NS), Canada. Hillaire-Marcel, C., A. de Vernal, G. Bilodeau, and A.J. Weaver (2001), Absence of deep water formation in the Labrador Sea during the last interglacial period, Nature, 410, 1073-1077. Hoogakker, B.A.A., G.P. Klinkhammer, H. Elderfield, E.J. Rohling, and C. Hayward (2009), Mg/Ca paleothermometry in high salinity environments, Earth Planet. Sci. Lett., 284, 583-589. Jonkers, L., G.-J. A. Brummer, F.J.C. Peeters, H.M. van Aken, and M.F. De Jong (2010), Seasonal stratification, shell flux and oxygen isotope dynamics of left-coiling N. pachyderma and T. quinqueloba in the western sub-polar North Atlantic,  Paleoceanography, 25, doi: 10/1029/2009PA001849.  Kerwin, M.W. (1996), A Regional Stratigraphic Isochron (ca. 8000 14C yr B.P.) from Final Deglaciation of Hudson Strait, Quat. Res., 46, 89-103. Khatiwala, S., R.G. Fairbanks, and R. Houghton (1999), Freshwater sources to the coastal ocean off northeastern North America, J. Geophys. Res., 104, 18241-18255.  Kisakurek, B., A. Eisenhauer, F. Böhm, D. Garbe-Schönberg, and J. Erez (2008), Controls on shell Mg/Ca and Sr/Ca in cultured planktonic foraminiferan, Globigerinoides ruber (white), Earth Planet. Sci. Lett., 273, 260-269. Klinkhammer, G.P., B.A. Haley, A.C. Mix, H.M. Benway, and M. Cheseby (2004), Evaluation of automated flow-through time-resolved analysis of foraminifera for Mg/Ca paleothermometry, Paleoceanography, 19, doi: 10.1029/2004PA001050. Kozdon, R., A. Eisenhauer, M. Weinelt, M. Y. Meland, and D. Nürnberg (2009), Reassessing Mg/Ca temperature calibrations of Neogloboquadrina pachyderma (sinistral) using paired δ44/40Ca and Mg/Ca measurements, Geochem. Geophys. Geosyst., 10, Q03005, doi:10.1029/2008GC002169. Lajeunesse, P., and G. St-Onge (2008), The subglacial origin of the Lake Agassiz–Ojibway final outburst flood, Nature Geosci., 1, 184-188. Lea, D.W. (2003), Elemental and Isotopic Proxies of Past Ocean Temperatures, in Treatise on Geochemistry, vol. 6.14, edited by H.D. Holland and K.K. Turekian, pp. 365-390, Elsevier-Pergamon, Oxford, U.K.  Lea, D.W., T.A. Mashiotta, and H.J. Spero (1999), Controls on magnesium and strontium uptake in planktonic foraminifera determined by live culturing, Geochim. Cosmochim. Acta, 63, 2369-2379. Locarnini, R. A., A. V. Mishonov, J. I. Antonov, T. P. Boyer, H. E. Garcia, O. K. Baranova, M. M. Zweng, and D. R. Johnson (2010), World Ocean Atlas 2009, Volume 1: Temperature, S. Levitus, Ed., NOAA Atlas NESDIS, 68, U.S. Government Printing Office, Washington, D.C. Robinson, L.F., J.F. Adkins, L.D. Keigwin, J. Southon, D.P. Fernandez, S.-L. Wang, and D.S. Scheirer (2005), Radiocarbon Variability in the Western North Atlantic During the Last Deglaciation, Science 310, 1469-1473.  Rohling, E.J., and S. Cooke (2002), Stable oxygen and carbon isotopes in foraminiferal carbonate shells, in Modern Foraminifera, Gupta, B.S. (Ed.), 239-277, Kluwer Academic Publishers, Netherlands. Roy, M., F. Dell’Oste, J.J. Veillette, A. de Vernal, J.-F. Hélie, and M. Parent (2011), Insights on the events surrounding the final drainage of Lake Ojibway based on James Bay stratigraphic sequences, Quat. Sci. Rev., 30, 682-692. Russell, A.D., B. Honisch, H.J. Spero, and D.W. Lea (2004), Effects of seawater carbonate ion concentration and temperature on shell U, Mg, and Sr in cultured planktonic foraminifera, Geochim. Cosmochim. Acta, 68, 4347-4361.  Shackleton, N.J. (1974), Attainment of isotopic equilibrium between ocean water and the benthic foraminifera genus Uvigerina: isotopic changes in the ocean during the last glacial, Collo. Intern. C.N.R.S., 219, 203-209.  Simstich, J., M. Sarnthein, and H. Erlenkeuser (2003), Paired δ18O signals of Neogloboquadrina pachyderma (s) and Turborotalia quinqueloba show terminal stratification structure in Nordic Seas, Mar. Micropaleontol., 48, 107-125.  Thornalley, D.J.R., H. Elderfield, and I.N. McCave (2009), Holocene oscillations in temperature and salinity of the surface subpolar North Atlantic, Nature, 457, 711-714.  Thornalley, D.J.R., S. Barker, W.S. Broecker, H. Elderfield, and I.N. McCave (2011), The Deglacial Evolution of North Atlantic Deep Convection, Science, 331, 202-205.  Tolderlund, D.S., and A.W.H. Bé (1971), Seasonal distribution of planktonic foraminifera in the western North Atlantic, Micropaleontology, 17, 297-329. Wu, G., and C. Hillaire-Marcel (1994), Oxygen isotope compositions of sinistral Neogloboquadrina pachyderma tests in surface sediments: North Atlantic Ocean, Geochim. Cosmochim. Acta, 58, 1303-1312.