The Role of the Tropical Oceans on Global Climate During a Warm Period and a Major Climate Transition

of 10

Please download to get full document.

View again

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
10 pages
0 downs
The Role of the Tropical Oceans on Global Climate During a Warm Period and a Major Climate Transition BY ANA CHRISTINA RAVELO AND MICHAEL WILLIAM WARA Pliocene global warmth; steeper temperature gradients,
The Role of the Tropical Oceans on Global Climate During a Warm Period and a Major Climate Transition BY ANA CHRISTINA RAVELO AND MICHAEL WILLIAM WARA Pliocene global warmth; steeper temperature gradients, more typical of the modern ocean, were established during the cool, ice-age climatic state by ~1.5 Ma. What caused the end of El Niño-like conditions and the onset of the ice ages? We show that changes in the oxygen isotope gradient between the Indian and Pacific Oceans occurred between ~3.0 and 1.5 Ma, indicating that gradual tectonic influences on flow through the Indonesian seaways may have caused changes in tropical sea surface temperature patterns that forced NHG. The marked increase in the gradient between the Pacific and Atlantic Oceans, possibly related to restriction of flow through the Panamanian seaway, occurred ~4.2 Ma, too early to be responsible for the onset of NHG. Other sources of gradual global climate cooling through the Pliocene are also discussed. Paleoceanographic records extracted from a global array of sediment cores obtained by the Ocean Drilling Program (ODP) can be used to elucidate differences between oceanographic conditions during the early Pliocene warm period (~4.5 to 3.0 million years ago [Ma]) and the late Pliocene and Pleistocene cool ice age period (3.0 Ma to present). Oxygen isotope gradients derived by laboratory analysis of calcareous microfossil shells from low-latitude sites are used to reconstruct tropical surface hydrographic (i.e., temperature ) gradients and to examine the role of tropical oceans on global climate over the last 5 million years, including the factors that caused the warm to cold climate transition, commonly referred to as the onset of significant Northern Hemisphere Glaciation (NHG). We find that a small west-east temperature gradient across the Pacific Ocean, similar to El Niño conditions, accompanied and perhaps played a critical role in determining early 32 Oceanography Vol.17, No.3, Sept This article has been published in Oceanography, Volume 17, Number 3, a quarterly journal of The Oceanography Society. Copyright 2003 by The Oceanography Society. All rights reserved. Reproduction of any portion of this article by photocopy machine, reposting, or other means without prior authorization of The Oceanography Society is strictly prohibited. Send all correspondence to: or 5912 LeMay Road, Rockville, MD , USA. The Ocean Drilling Program (ODP) drillship JOIDES Resolution sailing through the Panama Canal. Photo courtesy of ODP. Oceanography Vol.17, No.3, Sept BACKGROUND One of the main objectives of the ODP was to gain insight into fundamental processes regulating Earth s ocean and climate systems. Although the instrumental record provides direct measurements of seasonal to interannual climate cycles, paleoclimatological proxy-climate data derived from ocean sediment samples can be used to map oceanic and climatic variability that occurs on time scales longer than a few decades. Paleoclimatological studies are relevant to future global warming predictions because they provide evidence for the response of global climate to high levels of greenhouse gases, the relationships between high-frequency variability and mean climate states, and the causes of abrupt, nonlinear climate responses to perturbations and persistent forcings. Paleoclimate records are also critical for the study of climate components that respond on relatively long time scales (e.g., ice sheets and deep ocean circulation). To this end, the ODP focussed much of it drilling efforts on recovering ocean sediments (from a range of latitudes and water depths in the Pacific, Atlantic, and Indian Oceans) that represent past extreme warm and cold periods, past abrupt changes and transitions, and past variability or cyclicity of climate under different global boundary conditions. In particular, a unique opportunity exists to study climate change over the last 5 million years using ODP cores. Earth s climate regime was dramatically differ- Ana Christina Ravelo is Associate Professor, Ocean Sciences Department, University of California, Santa Cruz, CA. Michael William Wara, Ph.D., is currently a student at Stanford Law School, Stanford, CA cold period warm period 2.5 Less ice More ice Pleistocene Pliocene Figure 1. Benthic foraminifera δ 18 O record of high-latitude climate (primarily ice volume) from ODP Site 677 (Shackleton et al., 1990) and ODP Site 1085 (Andreasen, 2001). The end of the Pliocene warm period and the onset of Northern Hemisphere glaciation occurred at ~3.0 Ma. ent in the early Pliocene warm period (5 to 3 Ma) as compared to the Pleistocene ice age (~1.5 million years to present) (Figure 1), and it is relatively easy to obtain pristine material from this time period. As outlined below, the first goal of our study was to understand the role of tropical oceans in determining global warmth during the early Pliocene (~5 to 3 Ma). The second goal was to understand the role of the tropical oceans in the transition to cooler climatic conditions (from ~3 to ~1.5 Ma). Warm early Pliocene. The Pliocene warm period was the most recent period of sustained global warmth relative to today with ~3 C higher global surface temperature (Sloan et al., 1996; Dowsett et al., 1996; Haywood et al., 2000). Northern Hemisphere ice sheets were small (Jansen and Sjøholm, 1991), and atmospheric carbon dioxide concentration (pco 2 ) was probably slightly higher (Van der Burgh et al., 1993; Raymo et al., 1996) compared to today. Initial compilations of sea surface temperatures (SSTs) from the Pliocene warm period indicated that high-latitude oceans, but not tropical oceans, were significantly Age (Ma) warmer than today (Dowsett et al., 1996). This observation led paleoclimatologists to speculate that the Pliocene warm period could be explained by enhanced meridional heat advection, which would cause high-latitude warming without affecting tropical SSTs significantly (Sloan et al., 1996; Chandler et al., 1994). The difficulty in understanding the role for enhanced advection versus greenhouse-gas forcing was summarized by Crowley (1991, 1996), who highlighted the need for additional low-latitude SST data in order to resolve the causes of Pliocene warmth. Contrary to this earlier work, a recent modeling study predicts that in the Pliocene warm period tropical and sub-tropical, SSTs were warmer than modern SSTs due to differences in the cryosphere and cloud feedbacks (Haywood and Valdes, 2004). This new and different modeling result is independently supported by observations based on alkenone-derived SSTs (Haywood et al., in review) and suggests that the Pliocene warm period could actually be related to higher-than-modern pco 2 concentrations (Crowley, 1991; Crowley, 1996). More importantly, the model sug- 34 Oceanography Vol.17, No.3, Sept. 2004 gests that the causes of Pliocene warmth need to be understood in the context of warmer tropical and sub-tropical oceans. Thus, to advance understanding of the causes of early Pliocene global warmth, our first goal was to further characterize major differences in tropical climate conditions during the warm period compared to today. The transition to colder climate. A comprehensive view of high-latitude climate changes demonstrates that the end of the Pliocene warm period at ~3.0 Ma and the onset of significant NHG were generally not abrupt (Raymo, 1994), although in some locations in the North Pacific and North Atlantic, the transition appears to have occurred specifically at 2.75 Ma (with errors of ~50 ka) (Haug et al., 1999) once specific regional thresholds were reached. The cause or causes of NHG has been the subject of much debate. NHG may have been a result of long-term global cooling caused by a small decrease in carbon dioxide concentration coupled with positive feedbacks (e.g., Saltzman and Verbitsky, 1993; Berger and Wefer, 1996; Crowley, 1996; Raymo, 1998), as suggested for the onset of significant Antarctic glaciation around 34 Ma (DeConto, 2003). Alternatively, other studies propose that a highlatitude threshold, such as atmospheric cooling from repeated volcanic eruptions (Prueher and Rea, 2001), or diversion of Arctic air masses to mid-latitudes due to mountain building (Ruddiman and Raymo, 1988; Ruddiman and Kutzbach, 1990), was reached by ~3.0 Ma, which allowed for the onset of significant NHG and subsequent global cooling. As a third alternative, recent studies have proposed that low-latitude tectonic events, such as the uplift of the Isthmus of Panama (Driscoll and Haug, 1998) or the uplift and movement of islands within the Indonesian seaways (Cane and Molnar, 2001), may have forced tropical climate reorganization, which in turn may have caused high-latitude cooling and NHG. These hypotheses for the end of the warm period can be tested by examining the timing of the transition to cooler Pleistocene climate in different regions using ODP records. A recent paper (Ravelo et al., 2004) showed that major climate events around the globe occurred at significantly different times, implying that no abrupt regional event followed by a cascade of climate changes could have led to the Pleistocene ice ages. This result implies that the end of the warm period was caused by a gradual and persistent forcing, with different regions responding to the (unidentified) forcing at different times due to regionally specific thresholds. Thus, the second goal of our study was to investigate the potential role of the tropical oceans in gradually putting an end to the Pliocene warm period. Reorganization of tropical SST patterns can strongly influence global climate, as occurs interannually with the El Niño-Southern Oscillation phenomenon (Cane and Evans, 2000; Philander and Fedorov, 2003). Even small changes in tropical SST patterns can profoundly affect extratropical climate on geological time scales (Yin and Battisti, 2001). Although it has been proposed that lowlatitude tectonic events (uplift of the Isthmus of Panama or restriction of flow through the Indonesian seaway) may have changed the distribution of heat between basins, causing reorganization of global climate patterns, tropical ODP records have never been systematically synthesized and interpreted in this light. This article will present ODP records of sea surface changes from the tropical Indian, Pacific, and Atlantic Oceans with the intention of characterizing tropical conditions during the warm Pliocene, resolving the timing of tropical SST changes at the end of the warm period, and exploring the implications of this timing on proposed mechanisms for the Pliocene-Pleistocene climate transition. MODERN VERSUS EARLY PLIOCENE SST DISTRIBUTION The modern tropical SST distribution in the Indian, Pacific, and Atlantic Oceans is predominantly determined by the trade winds and the structure and depth of the subsurface thermocline. Although wind-driven equatorial upwelling occurs across the three basins, the thermocline is sufficiently deep in the Indian Ocean and in the western Pacific and Atlantic Oceans that SSTs remain relatively warm in those regions. Cool SSTs are found where upwelling-favorable winds and a shallow thermocline coincide, mainly in eastern Pacific and Atlantic tropical oceans (Figure 2). Thus, the tropical Pacific, whose temperature distribution is thought to influence climate patterns on a global scale, is characterized by strong west-east gradients in temperature and pressure. The easterly trade winds, strengthened by these gradients, further reinforce cool upwelling in the east, thereby augmenting the temperature and pressure gradients. The strong zonal, or Walker, circulation is maintained by these positive air -sea feedbacks (Bjerknes, 1969). The average conditions in the tropical Pacific today include strong zonal gradients and Walker circulation. However, the same air-sea feedbacks that maintain strong Walker circulation also amplify small perturbations that weaken Walker circulation, causing the thermocline to deepen in the Oceanography Vol.17, No.3, Sept 758 * * * * * 925 * 30 E 60 E 90 E 120 E 150 E W 120 W 90 W 60 W 30 W 0 18 C 19 C 20 C 21 C 22 C 23 C 24 C 25 C 26 C 27 C 28 C 29 C 30 C 31 C 32 C 33 C 34 C Sea Surface Temperature Figure 2. Sea surface temperature (SST) map of the world s tropical oceans (Reynolds and Smith, 1994). Change in the strong SST gradient across the equatorial Pacific is monitored using ODP Sites 806, 847, and 851. Changes in the strong eastern Pacific to Caribbean and Caribbean to western Atlantic gradients (not shown) are monitored using ODP Sites 851 and 999, and ODP Sites 999 and 925, respectively. Changes in the weak SST gradient from the western Pacific to the eastern Indian Ocean are monitored using ODP Sites 806 and 758. east and El Niño conditions to develop every few years. There is strong evidence that tropical Pacific conditions in the early Pliocene resembled a permanent El Niño. Specifically, the west-east SST gradient was greatly reduced, and the thermocline in the eastern tropical Pacific was deep compared with modern normal conditions (Cannariato and Ravelo, 1997; Chaisson and Ravelo, 2000; Wara, 2003; see discussion of Figure 6 below). Furthermore, extratropical climate anomalies relative to today were similar to those manifested during a modern El Niño (Molnar and Cane, 2002). The reasons for an El Niño-like pattern in the early Pliocene are unclear, but could be related to differences in the source of thermocline waters compared to today. The thermocline in the tropical Pacific is influenced by conditions in the regions where it is ventilated at mid-latitudes (Gu and Philander, 1997; Harper, 2000) by the amount of exchange with the Indian Ocean through Indonesian passages (Rodgers et al., 2000; Cane and Molnar, 2001), and by whole-ocean stratification (Philander and Fedorov, 2003). Changes in any of these factors could have potentially caused the thermocline to be deeper warmer in the warm Pliocene compared to today, thereby resulting in warmer SSTs in upwelling regions, reduced west-east SST gradient in the Pacific, and weaker Walker circulation. ESTABLISHMENT OF MODERN SST GRADIENTS How and when did the modern tropical SST patterns become established? How did the restriction of flow through the Indonesian and Panama seaways affect tropical ocean conditions at the end of the warm Pliocene, and specifically, at the onset of significant NHG? These questions can be answered using records from tropical ODP sites that monitor changes in hydrographic gradients (Figure 2, Table 1). Although relatively new paleo-proxy measurements, such as magnesium to calcium ratios (Mg/Ca) and alkenone unsaturation ratios (U k 37) are being widely applied to reconstruct SSTs in Pleistocene and recent times, there are no published records spanning the last 5 million years using these techniques at tropical locations. There are, however, oxygen isotope (δ 18 O) records from a number of tropical locations (Figure 2). The records presented in this article were all generated by measuring δ 18 O of the calcite shells of Globigerinoides sacculifer, a planktonic (surface-dwelling) foraminifer species that has been widely applied in studies of surface paleoceanography. The δ 18 O composition of a foraminiferal shell primarily reflects changes in the δ 18 O of seawater, and the temperature of seawater, in which the shell calcified. Evaporation/precipitation processes are responsible for changes in both the whole-ocean global and the local δ 18 O of seawater. When evapora- Table 1. Site Locations Site Latitude Longitude Water Depth (m) 999 (Caribbean Sea) 13 N 79 W (Western Tropical Atlantic Ocean) 4 N 43 W (Western Tropical Pacific Ocean) E (Eastern Tropical Pacific Ocean) 3 N 111 W (Eastern Tropical Pacific Ocean) 0 95 W (Eastern Tropical Indian Ocean) 5 N 90 E Oceanography Vol.17, No.3, Sept. 2004 tion occurs, water vapor is enriched in the lighter isotope ( 16 O), and therefore precipitation (rainfall or snow) has low δ 18 O compared to seawater. Ice sheets act as a reservoir of water (precipitation) with low δ 18 O values, and thus, during cold periods when large ice sheets store abundant water, the δ 18 O of the whole ocean is relatively high. Much of the long-term variability in a δ 18 O record is due to changes in the amount of ice stored on land. Over the last 5 million years, the long-term trend to higher δ 18 O values (Figure 1) is mostly due to the effect of increasing Northern Hemisphere ice-sheet size on the whole-ocean δ 18 O composition. Evaporation and precipitation also affects the local δ 18 O composition of surface seawater, just as it does the of surface seawater. Because rainfall has low δ 18 O values compared to seawater, regions with high rainfall relative to evaporation will have lower seawater δ 18 O values than regions with lower rainfall relative to evaporation. As a result, the δ 18 O and of seawater are highly correlated, and past local variations in can sometimes be derived using the δ 18 O measured on planktonic foraminifera. In addition, there is a strong temperature-dependent oxygen isotopic fractionation during the precipitation of calcite with more 18 O being incorporated into calcite as temperature decreases; thus, local temperature changes also influence the δ 18 O composition of foraminiferal calcite shells. Overall, changes in foraminiferal δ 18 O at any one locality reflects three factors: wholeocean global changes in the δ 18 O of seawater due to changes in ice-sheet size, local changes in the δ 18 O of seawater due to changes in the hydrologic cycle, and local changes in seawater temperature. Because changes in the whole-ocean δ 18 O are embedded in all the tropical δ 18 O records of G. sacculifer, it is difficult to use single records to predict absolute SST in past times. Rather, we can use the differences among the tropical records to reconstruct temperature gradients between sites. In other words, the shared variability (trends and cycles) in δ 18 O records from different locations are primarily due to global ice-volume changes, but variations in one record relative to another reflect changes in local surface-water properties. The records can be examined to resolve the timing of the evolution from sustained El-Niño-like to modern-like hydrographic gradients. Restriction of the Panamanian Seaway. Today, hydrographic conditions on either side of the Isthmus of Panama are different, with the eastern Pacific having lower temperature and than the western Atlantic (Caribbean Sea). If the uplift of the Isthmus of Panama, E. Eq. Pacific (851) warmer lower Age (Ma) which restricted flow between the two basins and reorganized the surface circulation in the Atlantic, played a key role in the end of the Pliocene warm period (Driscoll and Haug, 1998), then the δ 18 O records on either side of the Isthmus should provide evidence for this reorganization at ~3 Ma. However, Haug et al. (2001) demonstrate that there is a clear divergence between δ 18 O records from the eastern Pacific and the Caribbean at ~4.2 Ma, and attribute the establishment of the modern gradient to the effective restriction of the Panama seaway at that time (Figure 3). Models show that the Panama seaway restriction could have had a significant influence on Atlantic circulation (Maier- Reimer et al., 1990; Mikolajewicz and Crowley, 1997), including enhanced meridional ocean heat advection to high latitudes and thermohaline circulation. However, the idea that the increase in Caribbean (999) Figure 3. Planktonic (Globigerinoides sacculifer without sac-like final chamber) foraminifera δ 18 O record from ODP Sites 851 in the eastern equatorial Pacific Ocean (Cannariato and Ravelo, 1997) (blue) and ODP Site 999 in the Caribbean Sea (Haug et al., 2001) (orange), from both sides of the Panama seaway. Only the smoothed (using weighted squared error method) record through the higher (~4 ky
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks