Deep-water Carbonate Ion Concentrations In The Western Tropical Pacific Since The Middle Pleistocene: Implications For Oceanic Carbon Storage And Atmospheric PCO₂ Change

The shell weights of planktonic foraminifera,which undergoes changes with calcite dissolution,can be used as a proxy to reconstruct deep-water carbonate ion concentrations([CO₃^2-])in the past and,therefore,to provide clues regarding the ocean carbon cycle and global climate influence.In this study,we provide an improved protocol for cleaning of planktonic foraminifera for size-normalized weight?SNW?measurement.Then we present new SNW-carbonate ion concentration saturation levels(?[CO₃^2-])core-top calibrations for three planktonic foraminiferal species Globigerinoides sacculifer,Neogloboquadrina dutertrei and Pulleniatina obliquiloculata and assess their reliability as a paleo-alkalinity proxy.Based on these calibrations,we reconstruct SNW-based deep-water [CO₃^2-] for core WP7 and MD06-3047 b from the western tropical Pacific since the middle Pleistocene.Our findings provide new insights into the role of the deep Pacific in the global carbon cycle and the causes of changing atmospheric CO2.Finally,we discuss the factors controlling the initial size-normalized weight?ISNW?of P.obliquiloculata.Foraminiferal shells must be cleaned prior to weight measurements in order to remove detritus trapped within the shell chambers.The standard cleaning procedure,which involves an eight-second sonication step in water,causes greater breakage of thin-walled shells relative to thick-walled shells,thereby biasing weight measurements.Furthermore,the extent of shell breakage is greater for samples that have undergone significant in-situ partial dissolution.Because the sonic cleaning efficiency of sodium hexametaphosphate is much greater than that of water,we recommend a cleaning procedure based on a four-second sonication step in a 2% sodium hexametaphosphate solution.This procedural change provides a greater detritus cleaning effect and minimizes shell breakage,making shell weight measurements more reliable.New core-top SNW-?[CO₃^2-] calibrations of the planktonic foraminifera G.sacculifer,N.dutertrei,and P.obliquiloculata in the western tropical Pacific were established by using our improved cleaning procedure?i.e.,soaking plus a four-second sonication step in 2 % sodium hexametaphosphate solution?.These calibrations are valid over the water depth range of 1.6 to 3.9 km in the western Pacific.For the purpose of testing the reliability of SNW proxy,we determined SNWs for the planktonic foraminiferal species G.sacculifer,N.dutertrei,and P.obliquiloculata from cores KX97322-4 and WP7 since the LGM.The results indicate that variation in respiratory dissolution or initial shell weight was not a major factor controlling shell weight of G.sacculifer and N.dutertrei in the sediment record from the Pacific Ocean.However,the SNWs of P.obliquiloculata were affected by environmental factors in addition to deep-water ?[CO₃^2-].We reconstruct SNW-based deep-water [CO₃^2-] for core WP7 from the western tropical Pacific warm pool since 250 ka.Secular variation in the SNW proxy documents little change in deep Pacific [CO₃^2-] between the Last Glacial Maximum?LGM?and the Holocene.Further back in time,deep-water[CO₃^2-] shows long-term increases from marine isotope stage?MIS?5e to MIS 3 and from early MIS 7 to late MIS 6,consistent with the “coral reef hypothesis” that the deep Pacific Ocean carbonate system responded to declining shelf carbonate production during these two intervals.During deglaciations,we have evidence of [CO₃^2-] peaks coincident with Terminations 2 and 3,which suggests that a breakdown of oceanic vertical stratification drove a net transfer of CO2 from the ocean to the atmosphere,causing spikes in carbonate preservation?i.e.,the “deglacial ventilation hypothesis”?.During MIS 4,a transient decline in SNW-based [CO₃^2-],along with other reported [CO₃^2-] and/or dissolution records,implies that increased deep-ocean carbon storage resulted in a global carbonate dissolution event.We also reconstruct SNW-based deep-water [CO₃^2-] for core MD06-3047 b from the margin of the western tropical Pacific warm pool since 700 ka.[CO₃^2-] is higher during glacials and deglacials,and lower during interglacials.This pattern is consistent with the Pacific-style deep-sea sedimentary carbonate cycles.Deep Pacific Ocean carbonate system responded to glacial-interglacial changes in shelf carbonate production since 700 ka.In addition,[CO₃^2-] reached minimum of ⁴0 ?mol kg-1 during MIS 11.Increased carbonate dissolution during the Mid-Brunhes Dissolution Interval may be caused by a global increase in pelagic coccolith carbonate production.Furthermore,the long-term cyclicity of 400-500 kyr observed in ?¹³C records around the globel ocean is not obsearved in our [CO₃^2-] record.Further work will be required to determine the causes for this discrepancy.P.obliquiloculata ISNWs for core WP7 and MD06-3047 b are calculated using P.obliquiloculata SNWs and N.dutertrei SNW-based deep-water ?[CO₃^2-].P.obliquiloculata ISNW seems to be related mainly to subsurface-water [CO₃^2-].P.obliquiloculata ISNW for core WP7 is heavier during interglacials and deglacials,and lighter during glacials.This pattern of ISNW variation is related to changes of thermocline gradient associated with El Ni?o/LaNi?a-Southern Oscillation?ENSO?.However,long-term changes in P.obliquiloculata ISNW for core MD06-3047 b seem to be related to changes of thermocline gradient associated with East Asian Summer Monsoon?EASM?.Enhanced wind-driven circulation caused by strengthened ESAM would introduce a shoaling of the thermocline.As a result,a drawdown of subsurface-water [CO₃^2-] would have caused the P.obliquiloculata ISNW to decrease.