Search

Search Results

• 1. [Article] Late Cenozoic Climate, Ice-sheet and Earth-surface Evolution Derived from Terrestrial and Marine Sedimentary Archives

  The goals of this dissertation are centered on understanding changes in Earth surface and climate systems through the use of geologic proxies as records of past changes in these systems. Specifically, ...

  Citation × Citation Citation Abstract Copy Title: Late Cenozoic Climate, Ice-sheet and Earth-surface Evolution Derived from Terrestrial and Marine Sedimentary Archives Author: Bill, Nicholas S. The goals of this dissertation are centered on understanding changes in Earth surface and climate systems through the use of geologic proxies as records of past changes in these systems. Specifically, this dissertation (1) establishes a new chronology for retreat of the Ross Sea sector of the West Antarctic Ice during the last deglaciation, (2) investigates the changes in the global climate system during the mid- Pleistocene transition, and (3) constrains the timing of surface uplift in Alaska, and the control that topography has on regional climate and the hydrologic cycle over the tectonic timescale. This dissertation establishes the timing of the last deglaciation of the Ross Sea Sector of the West Antarctic Ice Sheet (WAIS), addressing the question of whether grounded ice in the Ross Embayment deglaciated entirely in the Holocene or earlier. ¹⁰Be surface exposure dating of granitic glacial erratic boulders indicates that the onset of sustained retreat from the local Last Glacial Maximum started at 18.6 ± 1.1 ka, and that sustained retreat occurred into the middle Holocene. We attribute most of this retreat to temperature and radiative forcing of the ablation zone in McMurdo Sound with only the final stages of retreat possibly influenced by Holocene grounding line retreat. The mid-Pleistocene transition (MPT) represents a transition from predominantly 40-kyr climate cycles to predominantly 100-kyr cycles in the absence of any change in orbital forcing. Here I combine all existing records of SST (n=7) and δ¹³C (n=17) that span the entirety of the last 2 Ma and use principal component analysis to detect the shared global signal of these records across the MPT. I also develop stacks of ice volume, ocean basin-scale δ¹³C gradients and CO₂ reconstructions in an attempt to characterize the interaction between deep ocean circulation and climate change across the MPT. I find that the characteristic change in cyclicity of the MPT appears in SST, δ¹³C and ice-volume reconstructions. I interpret marine isotope stage (MIS) 23 (~900 ka) as a skipped interglacial that led to MIS 25- 21 as being the first 100-kyr period, potentially initiating the 100-kyr cycle during the rest of the Pleistocene. I also find that the largest global negative δ¹³C excursion in the Pleistocene occurred during the glacial periods MIS 24 and MIS 22. This excursion is likely related to a mean shift to a reduced glacial-period AMOC across the MPT that is observed in multiple AMOC strength reconstructions, and that I interpret as a key component of the MPT. The topography of southern Alaska has been shown to have likely experienced rapid exhumation during the early Pliocene stating ~5-6 Ma. Using δD measurements on OH- groups of clay minerals, I constrain the change in isotopic composition of paleo-meteoric water from the Miocene-Holocene in the interior of Alaska as a proxy for the surface uplift history of Alaskan topography via its control on the δD signal of surface water. My results suggest that there was rapid surface uplift of the several mountain ranges encompassing southern Alaska during the period from ~6.5-3 Ma. Title: Late Cenozoic Climate, Ice-sheet and Earth-surface Evolution Derived from Terrestrial and Marine Sedimentary Archives Author: Bill, Nicholas S. The goals of this dissertation are centered on understanding changes in Earth surface and climate systems through the use of geologic proxies as records of past changes in these systems. Specifically, this dissertation (1) establishes a new chronology for retreat of the Ross Sea sector of the West Antarctic Ice during the last deglaciation, (2) investigates the changes in the global climate system during the mid- Pleistocene transition, and (3) constrains the timing of surface uplift in Alaska, and the control that topography has on regional climate and the hydrologic cycle over the tectonic timescale. This dissertation establishes the timing of the last deglaciation of the Ross Sea Sector of the West Antarctic Ice Sheet (WAIS), addressing the question of whether grounded ice in the Ross Embayment deglaciated entirely in the Holocene or earlier. ¹⁰Be surface exposure dating of granitic glacial erratic boulders indicates that the onset of sustained retreat from the local Last Glacial Maximum started at 18.6 ± 1.1 ka, and that sustained retreat occurred into the middle Holocene. We attribute most of this retreat to temperature and radiative forcing of the ablation zone in McMurdo Sound with only the final stages of retreat possibly influenced by Holocene grounding line retreat. The mid-Pleistocene transition (MPT) represents a transition from predominantly 40-kyr climate cycles to predominantly 100-kyr cycles in the absence of any change in orbital forcing. Here I combine all existing records of SST (n=7) and δ¹³C (n=17) that span the entirety of the last 2 Ma and use principal component analysis to detect the shared global signal of these records across the MPT. I also develop stacks of ice volume, ocean basin-scale δ¹³C gradients and CO₂ reconstructions in an attempt to characterize the interaction between deep ocean circulation and climate change across the MPT. I find that the characteristic change in cyclicity of the MPT appears in SST, δ¹³C and ice-volume reconstructions. I interpret marine isotope stage (MIS) 23 (~900 ka) as a skipped interglacial that led to MIS 25- 21 as being the first 100-kyr period, potentially initiating the 100-kyr cycle during the rest of the Pleistocene. I also find that the largest global negative δ¹³C excursion in the Pleistocene occurred during the glacial periods MIS 24 and MIS 22. This excursion is likely related to a mean shift to a reduced glacial-period AMOC across the MPT that is observed in multiple AMOC strength reconstructions, and that I interpret as a key component of the MPT. The topography of southern Alaska has been shown to have likely experienced rapid exhumation during the early Pliocene stating ~5-6 Ma. Using δD measurements on OH- groups of clay minerals, I constrain the change in isotopic composition of paleo-meteoric water from the Miocene-Holocene in the interior of Alaska as a proxy for the surface uplift history of Alaskan topography via its control on the δD signal of surface water. My results suggest that there was rapid surface uplift of the several mountain ranges encompassing southern Alaska during the period from ~6.5-3 Ma. Close

• 2. [Article] Glacial greenhouse-gas fluctuations controlled by ocean circulation changes

  Earth's climate and the concentrations of the atmospheric greenhouse gases carbon dioxide (CO2) and nitrous oxide (N2O) varied strongly on millennial timescales during past glacial periods. Large and rapid ...

  Citation × Citation Citation Abstract Copy Title: Glacial greenhouse-gas fluctuations controlled by ocean circulation changes Author: Schmittner, Andreas, Galbraith, Eric D. Earth's climate and the concentrations of the atmospheric greenhouse gases carbon dioxide (CO2) and nitrous oxide (N2O) varied strongly on millennial timescales during past glacial periods. Large and rapid warming events in Greenland and the North Atlantic were followed by more gradual cooling, and are highly correlated with fluctuations of N2O as recorded in ice cores. Antarctic temperature variations, on the other hand, were smaller and more gradual, showed warming during the Greenland cold phase and cooling while the North Atlantic was warm, and were highly correlated with fluctuations in CO2. Abrupt changes in the Atlantic meridional overturning circulation (AMOC) have often been invoked to explain the physical characteristics of these Dansgaard–Oeschger climate oscillations, but the mechanisms for the greenhouse-gas variations and their linkage to the AMOC have remained unclear. Here we present simulations with a coupled model of glacial climate and biogeochemical cycles, forced only with changes in the AMOC. The model simultaneously reproduces characteristic features of the Dansgaard–Oeschger temperature, as well as CO2 and N2O fluctuations. Despite significant changes in the land carbon inventory, CO2 variations on millennial timescales are dominated by slow changes in the deep ocean inventory of biologically sequestered carbon and are correlated with Antarctic temperature and Southern Ocean stratification. In contrast, N2O co-varies more rapidly with Greenland temperatures owing to fast adjustments of the thermocline oxygen budget. These results suggest that ocean circulation changes were the primary mechanism that drove glacial CO2 and N2O fluctuations on millennial timescales. Title: Glacial greenhouse-gas fluctuations controlled by ocean circulation changes Author: Schmittner, Andreas, Galbraith, Eric D. Earth's climate and the concentrations of the atmospheric greenhouse gases carbon dioxide (CO2) and nitrous oxide (N2O) varied strongly on millennial timescales during past glacial periods. Large and rapid warming events in Greenland and the North Atlantic were followed by more gradual cooling, and are highly correlated with fluctuations of N2O as recorded in ice cores. Antarctic temperature variations, on the other hand, were smaller and more gradual, showed warming during the Greenland cold phase and cooling while the North Atlantic was warm, and were highly correlated with fluctuations in CO2. Abrupt changes in the Atlantic meridional overturning circulation (AMOC) have often been invoked to explain the physical characteristics of these Dansgaard–Oeschger climate oscillations, but the mechanisms for the greenhouse-gas variations and their linkage to the AMOC have remained unclear. Here we present simulations with a coupled model of glacial climate and biogeochemical cycles, forced only with changes in the AMOC. The model simultaneously reproduces characteristic features of the Dansgaard–Oeschger temperature, as well as CO2 and N2O fluctuations. Despite significant changes in the land carbon inventory, CO2 variations on millennial timescales are dominated by slow changes in the deep ocean inventory of biologically sequestered carbon and are correlated with Antarctic temperature and Southern Ocean stratification. In contrast, N2O co-varies more rapidly with Greenland temperatures owing to fast adjustments of the thermocline oxygen budget. These results suggest that ocean circulation changes were the primary mechanism that drove glacial CO2 and N2O fluctuations on millennial timescales. Close

• 3. [Article] Augmenting and interpreting ice core greenhouse gas records

  The three studies that comprise this dissertation seek to answer significant questions in paleoclimatology through unconventional applications of ice core greenhouse gas data. These studies involve different ...

  Citation × Citation Citation Abstract Copy Title: Augmenting and interpreting ice core greenhouse gas records Author: Rosen, Julia L The three studies that comprise this dissertation seek to answer significant questions in paleoclimatology through unconventional applications of ice core greenhouse gas data. These studies involve different gases and span the interval of time between the Last Glacial Maximum and the Industrial Revolution, but are united by their nontraditional use of greenhouse gases and their attempt to realize the potential for greenhouse gases to reveal important information about Earth’s climate. Ever since their discovery, the abrupt climate changes of the last glacial period known as Dansgaard-Oeschger (D-O) events have proved challenging to explain. The dominant hypothesis involves periodic freshwater discharges into the North Atlantic, which may regulate the strength of the Meridional Overturning Circulation (MOC) and its role in transporting heat to high latitudes. These events were not restricted to the North Atlantic, and can also be recognized in paleoclimate archives around the world. However, numerous uncertainties surrounding the mechanism behind D-O events remain, including how they are communicated to low latitudes and whether other hypotheses can be definitively ruled out. To constrain the mechanism behind abrupt climate changes, we investigate the phasing of climate changes in high- and low-latitude regions at the Bølling Transition, the penultimate abrupt warming event of the last glacial period. We use methane and the ¹⁵N/¹⁴N ratio of N₂ from the North Greenland Eemian (NEEM) ice core, which serve as proxies for tropical climate and Greenland temperature, respectively. We find that these gases change synchronously in the ice core record, and use a firn air model together with a Monte Carlo approach to constrain the phase lag to within several decades. Our results indicate that the mechanism behind the Bølling Transition was capable of rapidly transmitting the climate signal across the planet in a matter of years, and must therefore involve components of the climate system that are suitably reactive. The glacial-interglacial change in atmospheric methane concentrations revealed in ice core records has spurred a decade of debate about its cause. The most likely explanations involve dramatic changes in methane emissions, which originate from both high- and low-latitude wetlands. One method of investigating the changing latitudinal distribution of methane sources is to quantify the difference in methane concentrations between Greenland and Antarctica, which changes in proportion to the fraction of methane produced at high northern latitudes. Previous attempts to determine the methane interpolar difference (IPD) abound, but many have been hampered by complications in synchronizing bipolar ice core records and analytical uncertainties. We present the first continuous estimate of the methane IPD across the termination using high-resolution methane data from the NEEM and West Antarctic Ice Sheet (WAIS) Divide ice cores. Our results reveal the dominant role of tropical sources in driving abrupt changes in atmospheric methane concentrations, and show that boreal methane sources were surprisingly insensitive to dramatic climate changes. We hypothesize that changes in Northern Hemisphere snow and ice cover exerted strong control over tropical methane emissions, while gradually increasing solar insolation and land area allowed boreal sources to grow during the termination. We also investigate the IPD across the major climate transitions of the termination, and during four centennial-scale methane variations, and find opposing trends in boreal and tropical source strengths during these transient events. We propose that temporary decoupling of the locus of interhemispheric mixing, the position of the Intertropical Convergence Zone, and tropical precipitation may explain these results. Atmospheric concentrations of nitrous oxide (N₂O) have risen by ~20% from preindustrial to modern times, but the cause of this increase is not fully understood. The change has been previously attributed to various agricultural activities which perturb microbial processes in soils, but exactly how remains an outstanding question with important implications for future mitigation of N₂O emissions. We present the first measurements of the isotopomers of tropospheric N₂O over the interval from 1450 to 1920 CE. Our results confirm that the preindustrial atmosphere was enriched in all isotopes relative to the modern atmosphere. Furthermore, we estimate that the net anthropogenic source of nitrous oxide must be depleted in all heavy isotopes and have a strong site preference, consistent with a strong role for agricultural emissions and characteristic of N₂O derived from nitrification. We also find a large oscillation in the site preference of the ¹⁵N in N₂O during the Little Ice Age between 1500 and 1700 CE. We hypothesize that this excursion may be due to changing climate conditions that led to an increase in the amount of N₂O produced by nitrification vs. denitrification. Title: Augmenting and interpreting ice core greenhouse gas records Author: Rosen, Julia L The three studies that comprise this dissertation seek to answer significant questions in paleoclimatology through unconventional applications of ice core greenhouse gas data. These studies involve different gases and span the interval of time between the Last Glacial Maximum and the Industrial Revolution, but are united by their nontraditional use of greenhouse gases and their attempt to realize the potential for greenhouse gases to reveal important information about Earth’s climate. Ever since their discovery, the abrupt climate changes of the last glacial period known as Dansgaard-Oeschger (D-O) events have proved challenging to explain. The dominant hypothesis involves periodic freshwater discharges into the North Atlantic, which may regulate the strength of the Meridional Overturning Circulation (MOC) and its role in transporting heat to high latitudes. These events were not restricted to the North Atlantic, and can also be recognized in paleoclimate archives around the world. However, numerous uncertainties surrounding the mechanism behind D-O events remain, including how they are communicated to low latitudes and whether other hypotheses can be definitively ruled out. To constrain the mechanism behind abrupt climate changes, we investigate the phasing of climate changes in high- and low-latitude regions at the Bølling Transition, the penultimate abrupt warming event of the last glacial period. We use methane and the ¹⁵N/¹⁴N ratio of N₂ from the North Greenland Eemian (NEEM) ice core, which serve as proxies for tropical climate and Greenland temperature, respectively. We find that these gases change synchronously in the ice core record, and use a firn air model together with a Monte Carlo approach to constrain the phase lag to within several decades. Our results indicate that the mechanism behind the Bølling Transition was capable of rapidly transmitting the climate signal across the planet in a matter of years, and must therefore involve components of the climate system that are suitably reactive. The glacial-interglacial change in atmospheric methane concentrations revealed in ice core records has spurred a decade of debate about its cause. The most likely explanations involve dramatic changes in methane emissions, which originate from both high- and low-latitude wetlands. One method of investigating the changing latitudinal distribution of methane sources is to quantify the difference in methane concentrations between Greenland and Antarctica, which changes in proportion to the fraction of methane produced at high northern latitudes. Previous attempts to determine the methane interpolar difference (IPD) abound, but many have been hampered by complications in synchronizing bipolar ice core records and analytical uncertainties. We present the first continuous estimate of the methane IPD across the termination using high-resolution methane data from the NEEM and West Antarctic Ice Sheet (WAIS) Divide ice cores. Our results reveal the dominant role of tropical sources in driving abrupt changes in atmospheric methane concentrations, and show that boreal methane sources were surprisingly insensitive to dramatic climate changes. We hypothesize that changes in Northern Hemisphere snow and ice cover exerted strong control over tropical methane emissions, while gradually increasing solar insolation and land area allowed boreal sources to grow during the termination. We also investigate the IPD across the major climate transitions of the termination, and during four centennial-scale methane variations, and find opposing trends in boreal and tropical source strengths during these transient events. We propose that temporary decoupling of the locus of interhemispheric mixing, the position of the Intertropical Convergence Zone, and tropical precipitation may explain these results. Atmospheric concentrations of nitrous oxide (N₂O) have risen by ~20% from preindustrial to modern times, but the cause of this increase is not fully understood. The change has been previously attributed to various agricultural activities which perturb microbial processes in soils, but exactly how remains an outstanding question with important implications for future mitigation of N₂O emissions. We present the first measurements of the isotopomers of tropospheric N₂O over the interval from 1450 to 1920 CE. Our results confirm that the preindustrial atmosphere was enriched in all isotopes relative to the modern atmosphere. Furthermore, we estimate that the net anthropogenic source of nitrous oxide must be depleted in all heavy isotopes and have a strong site preference, consistent with a strong role for agricultural emissions and characteristic of N₂O derived from nitrification. We also find a large oscillation in the site preference of the ¹⁵N in N₂O during the Little Ice Age between 1500 and 1700 CE. We hypothesize that this excursion may be due to changing climate conditions that led to an increase in the amount of N₂O produced by nitrification vs. denitrification. Close

• 4. [Article] Simulating past, present, and future changes in ENSO : a model evaluation and data-model comparison

  This thesis presents the results of a formal evaluation of a new AOGCM, GENMOM, demonstrating its ability to simulate present-day climate and ENSO dynamics. The model is applied to simulate climate for ...

  Citation × Citation Citation Abstract Copy Title: Simulating past, present, and future changes in ENSO : a model evaluation and data-model comparison Author: Alder, Jay R. This thesis presents the results of a formal evaluation of a new AOGCM, GENMOM, demonstrating its ability to simulate present-day climate and ENSO dynamics. The model is applied to simulate climate for the Last Glacial Maximum, deglacial, and Holocene time periods. The model output is evaluated against the best available proxy reconstructions in a detailed data-model comparison. ENSO strength is analyzed in seven paleo simulations and compared to coral and laminated lake sediment proxy records to provide an understanding of how ENSO related mechanisms varied in the past and how they vary under increased atmospheric CO₂ forced global warming. The GENMOM simulated present-day is found to be on par with three models used in the IPCC AR4 assessment and is comparable with reanalysis products (e.g, NCEP2). Atmospheric features such as the jet stream structure and major semipermanent sea level pressure centers are well simulated as is the mean planetary-scale wind structure that is needed to produce the correct position of stormtracks. Most ocean surface currents are reproduced except where they are not resolvable at T31 resolution. Overall, GENMOM captures the observed gradients and spatial distributions of annual surface temperature and precipitation and the simulations are on par with other AOGCMs. Deficiencies in the GENMOM present-day simulation include a warm bias in the surface temperature over the southern oceans, a split in the ITCZ and weaker-thanobserved overturning circulation. GENMOM produces a global temperature bias of 0.6 °C. GENMOM is demonstrated to capture ENSO dynamics similar to eight AOGCMs that were evaluated in the IPCC AR4. The Niño 3 - 4 indices have a standard deviation within 0.3 °C of the observations, indicating GENMOM is producing variability in the tropical Pacifc that is comparable to observations. GENMOM produces present-day ENSO events with an average period of 5.6 years, which is within the 2 – 7 range exhibited in the observed historical record. The mid-Holocene (6ka) and Last Glacial Maximum (LGM, 21ka) simulations are compared to the best available proxy reconstructions for sea surface temperature, precipitation and net moisture to ensure the simulations are plausible. This thesis finds that the model is in good agreement over broad spatial scales, with regional discrepancies between the model and proxy data. Coral and laminated lake sediment proxy records indicate mid-Holocene ENSO strength was reduced by 15 - 60%, offering a scenario in which ENSO-related components can be tested in climates different than present-day, thereby providing context for future changes in ENSO. The mid-Holocene simulations exhibit a 20% reduction of ENSO strength, caused by a precession forced enhancement of the Indian summer monsoon, which strengthened ENSO-related Bjerknes feedbacks. ENSO strength in the LGM is weakened by ~25%, which is not found to be caused by changes in equatorial Pacific dynamics but rather mean state cooling that weakens the tropical thermocline. The 2x and 4x simulations have strongly enhanced and more frequent ENSO events caused by disproportionate warming of the eastern Pacific relative to the western Pacific, which weakens the east-west Pacific surface temperature gradient, allowing larger anomalies, and hence ENSO events, to develop. Title: Simulating past, present, and future changes in ENSO : a model evaluation and data-model comparison Author: Alder, Jay R. This thesis presents the results of a formal evaluation of a new AOGCM, GENMOM, demonstrating its ability to simulate present-day climate and ENSO dynamics. The model is applied to simulate climate for the Last Glacial Maximum, deglacial, and Holocene time periods. The model output is evaluated against the best available proxy reconstructions in a detailed data-model comparison. ENSO strength is analyzed in seven paleo simulations and compared to coral and laminated lake sediment proxy records to provide an understanding of how ENSO related mechanisms varied in the past and how they vary under increased atmospheric CO₂ forced global warming. The GENMOM simulated present-day is found to be on par with three models used in the IPCC AR4 assessment and is comparable with reanalysis products (e.g, NCEP2). Atmospheric features such as the jet stream structure and major semipermanent sea level pressure centers are well simulated as is the mean planetary-scale wind structure that is needed to produce the correct position of stormtracks. Most ocean surface currents are reproduced except where they are not resolvable at T31 resolution. Overall, GENMOM captures the observed gradients and spatial distributions of annual surface temperature and precipitation and the simulations are on par with other AOGCMs. Deficiencies in the GENMOM present-day simulation include a warm bias in the surface temperature over the southern oceans, a split in the ITCZ and weaker-thanobserved overturning circulation. GENMOM produces a global temperature bias of 0.6 °C. GENMOM is demonstrated to capture ENSO dynamics similar to eight AOGCMs that were evaluated in the IPCC AR4. The Niño 3 - 4 indices have a standard deviation within 0.3 °C of the observations, indicating GENMOM is producing variability in the tropical Pacifc that is comparable to observations. GENMOM produces present-day ENSO events with an average period of 5.6 years, which is within the 2 – 7 range exhibited in the observed historical record. The mid-Holocene (6ka) and Last Glacial Maximum (LGM, 21ka) simulations are compared to the best available proxy reconstructions for sea surface temperature, precipitation and net moisture to ensure the simulations are plausible. This thesis finds that the model is in good agreement over broad spatial scales, with regional discrepancies between the model and proxy data. Coral and laminated lake sediment proxy records indicate mid-Holocene ENSO strength was reduced by 15 - 60%, offering a scenario in which ENSO-related components can be tested in climates different than present-day, thereby providing context for future changes in ENSO. The mid-Holocene simulations exhibit a 20% reduction of ENSO strength, caused by a precession forced enhancement of the Indian summer monsoon, which strengthened ENSO-related Bjerknes feedbacks. ENSO strength in the LGM is weakened by ~25%, which is not found to be caused by changes in equatorial Pacific dynamics but rather mean state cooling that weakens the tropical thermocline. The 2x and 4x simulations have strongly enhanced and more frequent ENSO events caused by disproportionate warming of the eastern Pacific relative to the western Pacific, which weakens the east-west Pacific surface temperature gradient, allowing larger anomalies, and hence ENSO events, to develop. Close