PP21C-01 INVITED 08:00h
The Pliocene Paradox
The early Pliocene, roughly 5 to 3 million years ago (Ma), was paradoxical in being both similar to, and also very different from the world of today. The distribution of continents and the atmospheric concentration of carbon dioxide were essentially what they are today, but globally averaged surface temperatures were approximately 3$^{o}$C higher. A key contributor to those warmer conditions was a perennial El Ni\~{n}o with surface waters of the equatorial Pacific as warm in the east as in the west. Other factors included the absence of major coastal upwelling regions - off the coasts of California and South-West Africa. Such conditions were possible because, presumably, the thermocline was so deep that the winds were unable to bring cold water to the surface. Studies of the oceanic circulation indicate that the maintenance of such a deep tropical thermocline requires surface waters in low latitudes to be as dense as in high latitudes. A small or zero equator-to-pole density gradient implies a weaker than today meridional temperature gradient but stronger meridional salinity gradient. This is favored by rainfall that is low in the subtropics, and high in subpolar regions, causing surface salinities to decrease with latitude. Thus, permanent El Ni\~{n}o conditions could prevail in the early Pliocene because salinity and temperature had the same (but opposite) effect on density. The subsequent gradual global cooling led to an increase in the meridional density gradient and resulted in a shallower thermocline. This terminated the perennial El Ni\~{n}o, introduced La Ni\~{n}a, and caused a reorganization of the wind-driven oceanic circulation that can account for various climate changes at that time. This study interprets the Pliocene paradox as an indicator that, at present, the global climate is sensitive to modest changes in latitudinal sea surface density gradients.
http://www.geology.yale.edu/~avf5/publications
PP21C-02 08:15h
Simulating the Atmospheric Climate of 3Ma
Recently, paleoclimate reconstruction have shown that permanent El Nino conditions prevailed in the tropical Pacific 3Ma. What was the role of the tropics at that time? Can the tropical conditions be at least partially reponsible for the warm climate during the early Pliocene? Here we present a first step toward addressing this issue in which atmospheric general circulation models are forced with different boundary conditions in the tropics to determine the response and their ability to reproduce key aspects of the atmospheric circulation 3Ma. We compare simulated surface temperature, hydrological cycle, wind stress and albedo, with those of today's climate, and look into the possibility that changes in surface atmospheric fluxes (of freshwater, heat and momentum) may in turn force the ocean generating a response that feeds back onto the atmosphere consistent with the boundary conditions, maintaining the climate.
PP21C-03 08:30h
Tropical-Extratropical Connections in Response to Obliquity Variations
An increase in obliquity causes the intensity of sunlight to increase in high latitudes and to decrease in the tropics. Recent observational evidence that equatorial sea surface temperatures increase when obliquity increases provides a test for theories concerning the role of tropical-extratropical atmospheric and oceanic connections in climate fluctuations. This presentation briefly reviews the observations and theories.
PP21C-04 INVITED 08:45h
Evidence for High Latitude Influence On Tropical Oceans Over the Pliocene and Pleistocene Epochs
Whether tropical or high latitude climate feedbacks dominate the signal of global climate change is an important, but difficult problem to solve. In the orbital frequency band, the presence of 41 kyr (obliquity) variations in low latitude records provides one unequivocal line of evidence that, in at least some important aspects, the high latitudes drive much of the climatic variance of the last 5 Ma. The effects of obliquity on either seasonal or mean annual insolation are very small in the tropics; furthermore, the effects on mean annual insolation in the tropics are in the opposite sense to those at latitudes $ > $ 45 $^{o}$. Here we compare spectral signatures from a variety of surface water proxy records from tropical locations of Pliocene and Pleistocene age. The proxies can be related to high latitude glaciation by comparing their response to benthic isotopic data gathered from the same cores. We find many periods of time in which obliquity variations, in phase with high latitude glaciation, dominated marine tropical surface records. The data imply a tight coupling between feedbacks to insolation originating in the high latitudes and their transmission to low latitude parts of the globe.
PP21C-05 INVITED 09:00h
Tropical Atlantic Upwelling and SST Changes associated with the mid-Pliocene onset of glacial cycles
Dynamical models examining the sensitivity of coupled tropical ocean-atmosphere processes to imposed boundary condition changes have proposed that late-Pliocene onset of glacial cycles near 3 Ma may have resulted from cooling and shoaling of the mean depth of the tropical thermocline. To test these ideas we measured proxies for upwelling, photic zone thermal gradients, and SSTs at eastern equatorial Atlantic Site 662 and compared these data to benthic d18O measured on the same samples. Our results show that the tropical Atlantic thermocline shoaled, cooled, and became responsive to orbital obliquity forcing after 2.9 Ma, in agreement with these hypotheses. Upwelling cycles led ice volume changes by 6-8 ka and maximum upwelling was in-phase with maximum annual equatorial insolation. Mg/Ca meaasurements of G. ruber indicate that SSTs also cooled and became more variable after 2.9 Ma. In agreement with theory, the mid-Pliocene warm period (ca. 3.0-3.3 Ma) was characterized by a deep, warm thermocline in the tropical Atlantic and Pacific. These results support recent observations that the tropics may have contributed to the late Pliocene onset of globally cooler and more variable climates. The onset of strong precessional variability which characterizes late Pleistocene upwelling records appears to have commenced after 2 Ma and it appears to reflect further shoaling and cooling of the tropical thermocline.
PP21C-06 INVITED 09:15h
Warm pool dynamics and the tropical glacial climate
One of the most prominent features of the tropical oceans is the existence of warm pools, vast areas of water with relatively homogenous temperatures. The extent to which the tropical warm pools were different in temperature or geographical distribution at the Last Glacial Maximum has been the subject of debate in recent years. Paleoceanographic evidence can provide some guidance on this issue. However, there is little understanding of the processes that control the features of the warm pool, how changes in the warm pool might come about, and how such changes would impact the tropical climate as a whole. In this paper, an array of modeling results is used to advance a simplified framework for understanding the oceanic and atmospheric processes that determine the geographical distribution of the warm pool. It is shown that cloud feedbacks and ocean dynamics are the key processes. Ocean general circulation model results are then used to demonstrate how altered ocean heat transports in the glacial climate can bring about significant changes in the warm pool. Finally, atmospheric general circulation model results are presented which show that changes in the size of the warm pool impact the mean temperature of the tropics via controls on the overall greenhouse effect of the tropical atmosphere.
PP21C-07 09:30h
Oceanic Tidal Mixing as a Contributor to Milankovitch-scale Climate Change
We propose that changes in the magnitude of oceanic tidal mixing on long time scales is an important, but previously unrecognized, contributor to global climate change. It is well known that Earth's orbital and rotational state changes significantly on 10$^{4}$-10$^{5}$ year time scales, and that this influences the spatial and temporal pattern of incident radiation. It is widely supposed that climatic variations on these same time scales are, in large part, a response of the ocean-atmosphere-cryosphere system to this radiative forcing. Our proposal is that variations in the luni-solar tidal potential, induced by these same orbital and rotational variations, influences oceanic mixing and thus modulates meridional heat transport, by amounts which are competitive with the radiative forcing. There are some obvious differences between tidal potential and insolation. First is that the Sun and Moon both contribute to tides, whereas the radiation is entirely of solar origin. Second is that the Earth is transparent to gravity but opaque to radiation. Clipping associated with this opacity makes the radiation pattern temporal spectrum rather more complex than the tidal spectrum. A third point is that solar radiation directly delivers energy to Earth's surface whereas tidal mixing will only expedite lateral transport of heat in association with oceanic thermo-haline circulation. The diurnal average insolation pattern is best parameterized via a Fourier series in time of year and Legendre polynomials in sine of latitude. Our present focus will be on the annual average terms. The Legendre degree n=0 term describes the global average insolation, and is nearly constant. The degree n=1 term describes differences between northern and southern hemispheres, and the annual mean is zero. The degree n=2 term is the main contributor to the equator to pole variations, and varies with obliquity and orbital eccentricity, with the obliquity variation dominating. The lowest order decomposition of the tidal potential recognizes 3 constituents: semi-diurnal, diurnal, and long period, with solar and lunar contributions to each. Our present focus will be on long term variations in the mean square amplitude of the semi-diurnal constituent, with averaging over all the short period variations. For the solar tide that includes the day and year. For the lunar tide it includes the day, month, year, and the apsidal (8.85 year) and nodal (18.6 year) periods. We present calculations of the variations in radiative and tidal forcing for the past 3 million years. The two signals are quite similar. Both vary by ~1% of their respective mean values, are dominated by obliquity variations, and exhibit only secondary influence from orbital eccentricity.
PP21C-08 09:45h
100 kyr-Period in Tropical Insolation
Since the paper by Hays et al. (1976), spectral analyses of climate proxy records provide substantial evidence that a fraction of the climatic variance is driven by insolation changes in the frequency ranges of obliquity and precession variations. However it is the variance components centered near 100 kyr which dominate most Upper Pleistocene Climatic records, although the amount of insolation perturbation at the eccentricity driven 100-kyr period is much too small to cause a climate change of ice-age amplitude. This direct influence of eccentricity on daily irradiation is actually through the variations of the so-called solar constant which amounts up to a few per mil at the maximum (Berger, 1977). Many attempts to find an explanation to this 100-kyr cycle in climatic records have been made over the last decades. Here we show that the double maximum which characterizes the daily irradiation received in tropical latitudes over the course of the year is at the origin of a 100-kyr and a 5.5-kyr periods related respectively to eccentricity and precession.