PaleoclimateTesting of the Tropical Greenhouse Forcing Hypothesis (supported by NSF ESH)

When we made the first determinations of glacial SSTs in the tropical Pacific using foraminiferal Mg/Ca in 2000, we were surprised to observe that changes in SST preceded changes in continental ice volume by ~3000 years. This result was unexpected and demanded an explanation. A growing appreciation of the fundamental role greenhouse forcing plays in determining the temperature of tropical surface waters lead us to consider the timing of greenhouse gas changes as documented in Antarctic Ice cores. Because the timing of carbon dioxide changes occurs early in the climate cycle, we explored the relationship between the expected greenhouse forcing and observed SST changes. This exploration is documented in a recent (JCLI, 2004) paper, in which it is demonstrated that SST changes observed in the tropics can be explained as a response to changes in greenhouse forcing (pdf).

Following on this work, we began to pursue a longer record of western tropical Pacific SST. This new work, lead by graduate student Martin Medina-Elizalde, demonstrated that SST changes showed strong glacial - interglacial oscillations back to at least 1.3 million y B. P. More fundamentally, these observations demonstrate that the periodicity of SST changes shifted ~950,000 y B. P., at a time termed the mid-Pleistocene transition (MPT). At this same transition records of continental ice volume also shift their periodicity from 41,000 (obliquity) to 100,000 years. This shift has traditionally been explained as a consequence of changes in the behavior of the ice sheets. Our observations, however, indicate that the tropical SST changes took place several thousand years before the ice sheet changes. In this case, the SST changes cannot be a consequence of changes in the behavior of the ice sheets. We suggest instead that the tropics were responding to a shift in the periodicity of radiative forcing by atmospheric carbon dioxide that occurred at the MPT. We speculate that carbon dioxide, acting as an internal self-sustained oscillator, was paced by obliquity changes during the early Pleistocene; this response shifted to the eccentricity envelope of precession after the MPT.

The extended Pacific warm pool SST record was published in the October 13th, 2005 issue of Science Express: pdf file.

Fig. 4 from the 2005 Science paper. Blowup of the mid-Pleistocene transition as seen in the Hole 806B SST record (A) and the Hole 677 benthic foraminiferal oxygen isotope record (B). Right panels show a blowup over the MPT of the spectrograms shown in figures 2 and 3. The interglacial peak centered on 950 kyr B.P. is marine isotope stage 25. Note the presence of the positive shift in benthic oxygen isotopes at ~900 kyr and the absence of a similar shift in SST at that time. The transition between the 41 kyr and 100 kyr variability occurs between 950 and 1000 ky B.P. in both records. Point-to-point comparisons between the two signals suggest that SSTs lead benthic oxyen isotopes by ~3 ky over the MPT.

Rapid Climate Change in the Glacial Tropics (supported by NSF ESH)

We are using the trace element and isotopic composition of fossil calcite shells of planktonic and benthic foraminifera from deep-sea cores to document changes in the temperature, hydrology and physical circulation of the tropical oceans.

Our first important discovery in this area, published as the lead research article in Science in September 2000 (link to ms. or .pdf file), is that sea surface temperature and hydrological (i.e., salinity) shifts in the equatorial Pacific are directly related to the growth and decline of continental ice sheets. A notable finding was that tropical sea surface temperature (SST) variations preceded ice volume changes by several thousand years. The results from this study provide the strongest evidence yet for the climate impact of the tropics in glacial - interglacial climate change. A recent focus of my research has been evaluating a new hypothesis that emerged from this data: that tropical warm pool SST variation is driven by changes in atmospheric carbon dioxide levels. The figure at the bottom of this web page illustrates the correspondence between tropical Pacific temperatures, Antarctic temperatures and atmospheric carbon dioxide levels.

I have also been collaborating with Sandy Tudhope (Edinburgh University, UK) for a number of years on the use of records from reef corals as paleo-climate archives. Sandy led a major study of the uplifted reefs on the Huon Peninsula, Papua New Guinea, to gauge the intensity of the El Niño/Southern Oscillation (ENSO) phenomenon over the last 130,000 years. The results from this study were published in a research article in Science in January 2001 (link to ms. or .pdf file ). One of the important conclusions of this study is that the intensity of ENSO in the last century appears to exceed that of the other time slices evaluated. In addition, ENSO intensity during cold glacial periods appears to have been about 50% weaker than during warmer periods. Coupling observations from fossil corals, which record interannual variations, with observations from deep-sea cores, which record average conditions, provides powerful new constraints for understanding the role of the tropics in global climate change.

Our most recent effort is focused on high resolution climate records of the last twenty-five thousand years from the Cariaco Basin, on the northern shelf of Venezuela (in collaboration with Larry Peterson, RSMAS and Konrad Hughen, WHOI). This extraordinary silled basin is anoxic below ~300m, and the lack of bioturbation results in bottom sediments that are laminated throughout most of the last fifteen thousand years of accumulation. In addition, sedimentation rates are very high (~0.35 mm/yr) and carbonate microfossil preservation is exceptional. Together, these factors elevate Cariaco sediments to arguably the best tropical climate record in the world. We are focusing on applying the Mg/Ca paleothermometer, in combination with oxygen isotopes, to Cariaco Basin sediments.

We have completed a 25,000 year long record of Mg/Ca-based SST variations at an average resolution of ~135 y, but with resolutions of 50 y in key intervals like the Younger Dryas (see figure below). This record indicates that the Cariaco region experienced three rapid (century or less) shifts in SST that correspond within age uncertainty (<100 y) with the rapid temperature shifts inferred from Greenland ice core oxygen isotope records. Notably, the timing of the Cariaco SST shifts agrees with similar shifts that have been observed at low latitude sites such as Hulu Cave in sub-tropical China. These observations leads to the hypothesis that these low latitude shifts might reflect large (hundreds of km) shifts in the position of the intertropical convergence zone (ITCZ), which mediates rainfall and climate in the tropics. A key question that emerges for future research is whether such ITCZ shifts were a passive response to high latitude changes or whether the tropics acted as a trigger, amplifier or even cause of high latitude change. We hope to evaluate this question by studying a new set of high quality piston cores from the Cariaco Basin that span the last 130,000 years.

The G. ruber Mg/Ca-based SST record from Cariaco core PL07-39PC was published in the September 5th, 2003 issue of Science magazine: pdf file. Supplementary Online Material.

Fig. 1 from the 2003 Science paper. The dramatic dip in G. ruber Mg/Ca and sediment grayscale reflectance data centered on ~450 cm depth marks the Younger Dryas chronozone. Note that the YD is much less obvious in the G. ruber oxygen isotope record, most likely because that record also reflects slowly changing ice volume as well as local salinity shifts.

Calibrating the Foraminiferal Recording System (supported by NSF, 1992-2001)

Glacial Ocean Chemistry (supported by NSF, 1990-1998)

Chemical Signals in Otoliths and Statoliths (supported by the Packard Foundation)

I am working with a number of biologists at UCSB to utilize the chemical signals encoded in otoliths (aragonite concretions in fish) and gastropod shells as a means of tracing patterns of larval recruitment (with Dr. Robert Warner and Dr. Steven Gaines, Ecology, Evolution and Marine Biology Department, UCSB). We have found that otoliths from different oceanic regions are marked by distinct chemical signatures. In gastropods, we hope to use seasonal cycles in the chemical signatures of the shells to establish their ages. Danielle Zacherl, Julie Standish and Ben Ruttenberg, all EEMB students, are pursuing their Ph.D. dissertations in this area. A recent EEMB Ph.D., Stephen Swearer, had his work in this groundbreaking area featured on the cover of Nature magazine! link to .pdf file. Stephen is now a lecturer at the University of Melbourne, Australia.

Deep-Sea Corals (supported by LLNL-CAMS and LANL-CULAR, 1994-1999)

We are exploring the geochemistry of solitary deep-sea corals, a new archive of ocean-climate change. These corals grow to depths of several thousand meters, are found at virtually all latitudes, and grow slowly enough to record hundreds of years of paleoceanographic history. We find that relict specimens of glacial age are preserved on the Antarctic seafloor and can be age dated by uranium series disequilibrium and radiocarbon decay. These independent age estimates can be combined to establish ventilation or mixing rates for the glacial oceans. Our first set of results from a glacial-age coral from the Drake Passage at 1100m indicates that Southern Ocean mixing rates were up to ~50% slower during the last ice age (Goldstein, Lea and others, 2001, EPSL: link to ms.). This work is being done collaboratively with Dr. Steven Goldstein, Los Alamos National Laboratory, and has been supported by a CULAR grant from LANL.