Description: Coinciding with global warming, Arctic sea ice has rapidly decreased during the last four decades and climate scenarios suggest that sea ice may completely disappear during summer within the next about 50-100 years. Here we produce Arctic sea ice biomarker proxy records for the penultimate glacial (Marine Isotope Stage 6) and the subsequent last interglacial (Marine Isotope Stage 5e). The latter is a time interval when the high latitudes were significantly warmer than today. We document that even under such warmer climate conditions, sea ice existed in the central Arctic Ocean during summer, whereas sea ice was significantly reduced along the Barents Sea continental margin influenced by Atlantic Water inflow. Our proxy reconstruction of the last interglacial sea ice cover is supported by climate simulations, although some proxy data/model inconsistencies still exist. During late Marine Isotope Stage 6, polynya-type conditions occurred off the major ice sheets along the northern Barents and East Siberian continental margins, contradicting a giant Marine Isotope Stage 6 ice shelf that covered the entire Arctic Ocean.

Description: Herein, we publish the simulated global annual mean surface air temperatures (tsurf), zonal (UKO) and meridional (VKE) velocities, temperature (THO), salinity (SAO) and horizontal barotropic streamfunction (PSIUWE) over a time period of 100 years retrieved from equilibrium climate simulations for testing the sensitivity of the Greenland-Scotland Ridge and utilized in the publication by Stärz et al. (2017). The climate data has been produced with COSMOS (ECHAM5/JSBACH/MPIOM/OASIS3), utilized at a resolution of T31 in the atmosphere (19 hybrid sigma-pressure levels) and a resolution of GR30 (bipolar orthogonal curvilinear grid, formal resolution of ~3.0°x1.8°) in the ocean (40 z-coordinate levels). The model setup refers to boundary conditions (incl. changes in orography, bathymetry, physical land surface characteristics, ice sheets, atmospheric CO2) representative for the Miocene. Further information on the model setup and the model scenarios, including identifiers, is given in the Supplementary Table 1 of Stärz et al. (2017).

Description: North Atlantic Deep Water (NADW) currently redistributes heat and salt between Earth's ocean basins, and plays a vital role in the ocean-atmosphere CO2 exchange. Despite its crucial role in today's climate system, vigorous debate remains as to when deep-water formation in the North Atlantic started. Here, we present datasets from carbonate-rich middle Eocene sediments from the Newfoundland Ridge, revealing a unique archive of paleoceanographic change from the progressively cooling climate of the middle Eocene. Well-defined lithologic alternations between calcareous ooze and clay-rich intervals occur at the ~41-kyr beat of axial obliquity. Hence, we identify obliquity as the driver of middle Eocene (43.5-46 Ma) Northern Component Water (NCW, the predecessor of modern NADW) variability. High-resolution benthic foraminiferal d18O and d13C suggest that obliquity minima correspond to cold, nutrient-depleted, western North Atlantic deep waters. We thus link stronger NCW formation with obliquity minima. In contrast, during obliquity maxima, Deep Western Boundary Currents were weaker and warmer, while abyssal nutrients were more abundant. These aspects reflect a more sluggish NCW formation. This obliquity-paced paleoceanographic regime is in excellent agreement with results from an Earth system model, in which obliquity minima configurations enhance NCW formation.

Description: The climate of the Sahara and Arabian deserts during the Little Ice Age is not well known, due to a lack of annually resolved natural and documentary archives. We present an annual reconstruction of temperature and aridity derived from Sr/Ca and oxygen isotopes in a coral of the desert-surrounded northern Red Sea. Our data indicate that the eastern Sahara and Arabian Desert did not experience pronounced cooling during the late Little Ice Age (~1750-1850), but suggest an even more arid mean climate than in the following ~150 years. The mild temperatures are broadly in line with predominantly negative phases of the North Atlantic Oscillation during the Little Ice Age. The more arid climate is best explained by meridional advection of dry continental air from Eurasia. We find evidence for an abrupt termination of the more arid climate after 1850, coincident with a reorganization of the atmospheric circulation over Europe.

Description: Reduced surface-deep ocean exchange and enhanced nutrient consumption by phytoplankton in the Southern Ocean have been linked to lower glacial atmospheric CO2. However, identification of the biological and physical conditions involved and the related processes remains incomplete. Here we specify Southern Ocean surface-subsurface contrasts using a new tool, the combined oxygen and silicon isotope measurement of diatom and radiolarian opal, in combination with numerical simulations. Our data do not indicate a permanent glacial halocline related to melt water from icebergs. Corroborated by numerical simulations, we find that glacial surface stratification was variable and linked to seasonal sea-ice changes. During glacial spring-summer, the mixed layer was relatively shallow, while deeper mixing occurred during fall-winter, allowing for surface-ocean refueling with nutrients from the deep reservoir, which was potentially richer in nutrients than today. This generated specific carbon and opal export regimes turning the glacial seasonal sea-ice zone into a carbon sink.

Description: New sclerochronological data suggest that a variability comparable to the North Atlantic Oscillation (NAO) was already present during the middle Oligocene, about 20 Myr earlier than formerly assumed. Annual increment width data of long-lived marine bivalves of Oligocene (30-25 Ma) strata from Central Europe revealed a distinct quasi-decadal climate variability modulated on 2-12 (mainly 3-7) year cycles. As in many other modern bivalves, these periodic changes in shell growth were most likely related to changes in primary productivity, which in turn, were coupled to atmospheric circulation patterns. Stable carbon isotope values of the shells (d13Cshell) further corroborated the link between shell growth and food availability. Sub-decal oscillations in the 3-7 year band in other annually resolved fossil archives were often interpreted as El Niño-Southern Oscillation (ENSO) cycles. This possibility is discussed in the present study. However, combined shell-derived proxy and numerical climate model data lend support to the interpretation of a NAO-like variability. According to numerical climate models, winter sea-level pressure (wSLP) and precipitation rate (wPR) across Central Europe during the Oligocene exhibited a pattern similar to the modern NAO. The simulated NAO index for the Oligocene shows periodicities coherent with those revealed by the proxy data (2.5-6 years), yet, on shorter wavelengths than the modern NAO (biennial and 6-10 year cycles). Likely, the different paleogeography and elevated atmospheric CO2 concentrations not only influenced the sea-level pressure pattern, but also the temporal variability of the NAO precursor. The present study represents the first attempt to characterize the inter-annual climate variability in Central Europe during the Oligocene and sets the basis for future studies on the early phase of the Cenozoic icehouse climate state.

Description: Here we publish simulated Late Cretaceous (~70 Ma) annual mean surface temperatures (tsurf) with different (280-1680 ppm) CO2 levels, annual mean surface temperatures with different gateway configurations between the Arctic Ocean and North proto-Atlantic basin, as well as summer (JJA) and winter (DJF) mean surface temperatures with 840 ppm CO2 level. All data were averaged over the period of 100 years. The climate data has been produced with COSMOS (ECHAM5/MPIOM/OASIS3). The atmosphere component ECHAM5 was run in the resolution of T31/L19, while the ocean model MPIOM in the resolution of GR30 (bipolar orthogonal curvilinear grid with a formal resolution of ~3.0°x1.8°). OASIS3 is responsible for coupling between the atmosphere and ocean components. The Late Cretaceous setup is based on the Markwick and Valdes (2004, doi:10.1016/j.palaeo.2004.06.015) paleogeography. More details about a model setup can be found in Niezgodzki et al. (2017, doi:10.1002/2016PA003055). Additionally, we have prepared the .xlsx file where we provide the details of the locations of data points as well as reconstructed temperatures as used in a model-data comparison. These data are the modified data from Upchurch et al. (2015, doi:10.1130/G36802.1). For more details about the procedure used to modify the original data set, see Niezgodzki et al. (2017, doi:10.1002/2016PA003055).

Description: Model output of the intermediate complexity climate model CLIMBER-2 over the past 5 million years. The simulations were forced with insolation data (O), insolation and land ice data (OI), insolation and carbon dioxide data (OC) and with insolation, land ice and carbon dioxide data (OIC). Sheet 1 contains the main results: northern hemispheric (30-90 deg N), southern hemispheric (30-90 deg S) and global temperatures. Sheet 2 contains the land ice and carbon dioxide forcing in terms of globally averaged radiative forcing. Details are given in the publication. More information or data can be obtained by contacting L.B. Stap (lennert.stap@awi.de).

Description: We present ice thickness and bedrock height from simulations of the Miocene Antarctic ice sheet (AIS), using the 3D thermodynamical Parallel Ice Sheet Model (PISM) version 0.7.3. The applied climate forcing consists of temperature and precipitation anomalies with respect to a preindustrial reference simulation, obtained from simulations using the atmosphere-ocean general circulation model COSMOS. These anomalies are added to an ERA-40/WOD-09 base climate. COSMOS was run using preindustrial settings, and Miocene settings with CO2 levels of 278 ppm (low), 450 ppm (medium), and 600 ppm (high). The steady state simulations using PISM are started with present-day bedrock conditions, with a present-day AIS (Bedmap2), and isostatically rebounded after removal of the ice. Additional simulations are started with the Wilson et al. (2012) late-Eocene bedrock topography reconstruction. The steady state simulations are conducted by applying the same climate forcing over 200 kyr. The transient simulations are performed by using an index method to interpolate between different forcing climate states. In these runs, the forcing climate state is gradually varied over quasi-orbital timescales of 400 kyr, 40 kyr, 1600 kyr, or 200 kyr. Details are given in the accompanying publication. For more information or data, please contact L.B. Stap mailto:lstap@awi.de. Dataset 2019_Stap_steadystate_startfrom_presentdayAIS.nc contains the data to plot Fig. 1. Dataset 2019_Stap_PISM_transient.xlsx contains the data to plot Fig. 2 and 3. Datasets 2019_Stap_steadystate_startfrom_noice.nc and 2019_Stap_steadystate_startfrom_Wilson_noice.nc contain additional data plotted in Fig. 2 and 3.