Arctic Paleoclimates

Arctic Paleoclimates The geologic time scale [from the Geological Society of America, product code CTS004, compiled by A.R. Palmer and J. Geissman, ...
Author: Caitlin Powell
7 downloads 2 Views 5MB Size
Arctic Paleoclimates

The geologic time scale [from the Geological Society of America, product code CTS004, compiled by A.R. Palmer and J. Geissman, by permission of Geological Society of America].

How can paleoclimate research be helpful? • The paleoclimate record puts our present climate into perspective. • Past climates reveal what the system is capable of doing. • Paleoclimate “proxy” data (tree rings, ice cores, pollen, etc.) give spatial and temporal details of past climate behavior. • Paleoclimate data allow can guide model experiments to address provide plausible physical mechanisms of past climate behavior.

Climate forcing mechanisms, processes, and feedbacks • • • • • • • • •

Geographic changes (land-sea distribution, mountain uplift) Variations in solar output Variations in greenhouse gases Variations in Earth’s orbital geometry (Milankovitch cycles) Volcanic eruptions --------------------------Vegetation feedbacks Ice/snow-albedo feedbacks Changes in atmospheric and oceanic circulation ---------------------------Abrupt climate change

The changing configuration of continents - a major river of climate changes in the distant past

http://omegaearthscience.pbworks.com/w/page/6702410/Continental-Drift

Paleoclimate records for the Quaternary (1.8 Ma to present) •

• • • •

Ice cores from Antarctica and Greenland (GISP2, GRIP, EPICA); Greenland records go back to the Eemian, Antarctic records go back 450,000-750,000 years. Marine sediment cores from all over the world, covering the entire Quaternary and even longer Pollen, diatoms, plant and animal macrofossils preserved in lake seiments and peat bogs Loess deposits, tree rings (dendrochronology), speleothems (mineral formations in limestone caves) Geomorphic features, such as raised beaches, moraines and glacial erratics

Ice and ocean cores

http://earthobservatory.nasa.gov/Features/Paleocl imatology_IceCores/

http://www.global-greenhousewarming.com/climate-and-ocean-cores.html

Ice and Ocean Cores •

Ice cores contain records of temperature from oxygen isotopes ( 18O). Other factors equal, the higher the 18O concentration in the ice core, the higher the temperature of condensation of the water vapor that led to the precipitation, and hence the warmer the conditions. Records of accumulation can be obtained from the thickness of annual ice layers. Impurities in the ice, such as wind blown dust and soluble ions yield information on characteristics of the atmospheric circulation and land surface.



Marine (ocean) sediment cores comprise mixes of continentally-derived material and microscopic marine organisms that rain out, or live and die in the sediments and accumulate year after year in layers. Oxygen isotope records are preserved, coming from the remains of calcareous organisms (planktonic and benthic foraminifera) preserved in sediments. The higher the 18O concentration, the higher the global ice volume. Different species of planktonic and benthic foraminifera thrive in different environmental conditions so changes in the abundance of different types record environmental changes in the uppermost and deep oceans, respectively. Marine cores also contain information from inorganic markers, such as detritus carries and dropped by icebergs.

Foraminifera www.eeob.iastate.edu

bio1903.nicerweb.com

www-paoc.mit.edu

www.ucl.ac.uk www.treehugger.com

Radiocarbon dating: Based on the decay rate of radiocarbon (14C) which has a half life of 5730 years







Radiocarbon is produced in the upper atmosphere by neutron bombardment of nitrogen atoms. The neutrons are part of the cosmic ray flux. Plants and animals assimilate radiocarbon into their tissues via photosynthesis and respiration, with a radiocarbon content in equilibrium with that in the atmosphere. Equilibrium occurs as there is a constant exchange of 14C as old cells die and are replaced. When organisms die, exchange and replacement of radiocarbon ends, activating a radioactive clock For various reasons, including variations in the cosmic ray flux, 14C levels have varied over time, and this needs to be adjusted for if one is directly compare radiocarbon and calendar ages.

Calibration curve for the radiocarbon dating scale [from Wikipedia]

Dendrochronology: Dating and paleoclimate analysis using data from tree rings.

Limiting climate factors

Cross dating

Sources: http://web.utk.edu/~grissino/gallery.htm

Palynology: The study of pollen

Image of pollen grains from a variety of common plants: sunflower (Helianthus annuus), morning glory (Ipomoea purpurea), prairie hollyhock (Sidalcea malviflora), oriental lily (Lilium auratum), evening primrose (Oenothera fruticosa), and castor bean (Ricinus communis). Source: http://en.wikipedia.org/wiki/Pollen

A glacial erratic

U.S. Geological Survey photo by Bruce Molnia

Chronology of the Quaternary

The Vostok ice core record

The Vostok record shows at least 4 major global scale ice advances over the past 400,000 years. The inferred temperature time series from oxygen isotope records is highly correlated with the ice core record of atmospheric carbon dioxide concentration. Ocean core records reveal that over the past 2.5 million years, there have been about two dozen glacial/interglacial cycles.

Milankovitch forcings: Pacemaker of the ice ages

Variations in eccentricity, axial tilt and precession (the timing of the equinoxes) affect the solar flux striking the surface at different latitudes and at different times of the year. These forcings have “paced” the timing of the major ice ages and interglacials of the Quaternary. Source:http://www.homepage.montana.edu /~geol445/hyperglac/time1/milankov.htm

Milankovitch cycles and glaciations: Past 1 million years

http://www.globalextinction.org/Solar.html

Evidence of orbital forcing on climate

http://www.ncdc.noaa.gov/paleo/abrupt/data2.html Interglacial periods, (shaded in yellow) in the Dome Fuji ice core from Antarctica, tend to happen during times of more intense summer solar radiation in the Northern Hemisphere. Since the middle Quaternary, glacial-interglacial cycles have had a frequency of about 100,000 years. In the solar radiation time-series, cycles of this length (eccentricity) are present but are weaker than cycles lasting about 23,000 years (precession of the equinoxes). For reasons not yet fully understood, full interglacials occur only about every fifth peak in the precession cycle. Internal feedbacks must be important. Temperature variations in Antarctica are in phase with solar radiation changes in the high northern latitudes.

Temperature and atmospheric greenhouse gas concentrations have closely followed each other for hundreds of thousands of years. Rises and falls in temperature precede greenhouse gas changes. This tells us that greenhouse gases operate as a feedback, “globalizing” the effects of Milankovitch forcings. Today’s carbon dioxide concentration (about 390 ppm) is higher than anything seen in ice core records.

http://earthobservatory.nasa.gov/Features/Paleoclimatology_IceCores/

The last glacial cycle

The SPECMAP (Spectral Mapping Project) composite chronology for a set of seven stacked (superimposed) δ 18O records from different ocean basins of the world. Dating involves tuning of the marine isotope records by orbital forcing [from Martinson et al., 1987, by permission of Elsevier]. Negative excursions in the δ 18O record correspond to cold periods and positive excursions correspond to warm periods. The last interglacial, known as the Eemian (Isotope Stage 5.5), peaked about 125,000 years ago. The last glacial maximum (Isotope Stage 2.2) occurred 18,000-25,000 years ago depending on region. We are presently in another warm interglacial, and are scheduled to slide into another ice age in 10,000 years or so.

Modeled extent and elevation of the Greenland ice sheet during the Eemian interglacial (a-c) under three different temperature reconstructions (d) based on the GRIP ice core records [adapted from Cuffey and Marshall, 2000, by permission of Nature]. The ice sheet was much smaller during the Eemian and sea level was likely 4-6 m higher than today.

Summer surface air temperature anomaly over the Arctic (left) and extent and thickness for the Greenland ice sheet during the height of the Eemian interglacial from a multi-model and multi-proxy synthesis [from the IPCC-AR5, Working Group 1 Report, Figure 6.6]. The temperature plot is for the height of the Eemian minus pre-industrial.

Extent of Northern Hemisphere glacial ice during the Last Glacial Maximum (LGM) [from Denton and Hughes (eds.), 1981, by permission of John Wiley and Sons]. There were ice sheets over both North America and northwestern Eurasia. Global sea level was around 120 m lower than today.

The Bering Strait was closed – there was no Arctic-Pacific ocean connection when continental ice sheets were present. While altering the circulation and freshwater budget of the Arctic Ocean, the Bering land bridge allowing for the migration of humans into North America.

Ice extent in northwestern Eurasia during the LGM [adapted by Siegert et al., 2001 from Svendsen et al., 1999, by permission of John Wiley and Sons]. Glacial ice covered parts of the continental shelves.

http://www.ncdc.noaa.gov/paleo/abrupt/data3.html Two different types of climate changes, Heinrich (H) and Dansgaard-Oeschger (D-O) events, occurred throughout much of the last glacial cycle. Each of the 25 known D-O events (first reported in Greenland ice cores) consist of an abrupt warming (over a matter of decades) to near-interglacial conditions, followed by a gradual cooling. Heinrich events are associated with six of the coldest intervals between D-O events. They are recorded in North Atlantic marine sediments as layers with a large amount of coarse-grained sediments derived from land, evidence for both an increase in sediment-carrying icebergs discharged from the Laurentide ice sheet and a southward extension of cold, polar waters, allowing icebergs to travel farther south before melting. D-O events repeated every several thousand years on average, while ~10,000 years elapsed between Heinrich events.

What caused D-O cycles? • •



Most ideas invoke episodes of surface freshening in the North Atlantic that disrupt the thermohaline circulation. The are many different variations, including the concept of a “salt oscillator”, “binge-purge” behavior of the Laurentide ice sheet, and massive discharge of freshwater from glacial-dammed lakes. Other ideas involve more direct climate forcing (e,g., periodic changes in solar irradiance) and then subsequent climate responses to surface freshening.

The Younger Dryas Event About 14,500 years ago, global climate began to shift into interglacial state. Partway through this transition, Northern Hemisphere temperatures suddenly plunged to near-glacial conditions. This period of cold conditions is called the Younger Dryas (YD), after a flower (Dryas octopetala) that thrives in cold conditions and became common in Europe during this time. The end of the YD, about 11,500 years ago, was particularly abrupt. In Greenland, temperatures rose 10° C (18° F) in a decade. The YD is sometimes referred to as the last of the D-O cycles. The figure at left shows climate changes associated with the YD (highlighted by the light blue bar) including cooling and decreased snow accumulation in Greenland, cooling in the tropical Cariaco Basin, warming in Antarctica and the estimated flux of meltwater from the Laurentide Ice Sheet down the St. Lawrence River. http://www.ncdc.noaa.gov/paleo/abrupt/data4.html

What caused the Younger Dryas? Like D-O cycles, ideas focus on a massive freshwater flux to the North Atlantic reducing the strength of the thermohaline circulation. Just prior to the YD, meltwater fluxes into the North Atlantic through the St. Lawrence River increased dramatically. In addition, there was probably a short-lived period of particularly high freshwater flux about 13,000 years ago from a large discharge of freshwater from a glacial lake in North America. Why the warming seen in Antarctic ice cores? If the thermohaline circulation slowed, less heat would be transported from the South Atlantic, causing the South Atlantic to warm and the North Atlantic to cool. This “bi-polar seesaw” is supported by comparisons between the YD recorded at GISP2 (Greenland, cold conditions) and Dome C (Antarctica, warm conditions). Eventually, as the meltwater flux abated, the thermohaline circulation strengthened and the climate recovered. The figure at left shows the Laurentide Ice Sheet and the routing of overflow from the Lake Aggasiz basin (dashed line) to the Gulf of Mexico just before the YD (a) and routing of overflow from Lake Aggasiz through the Great Lakes to the St. Lawrence and northern North Atlantic during the YD (b) [from Broecker et al., 1989, by permission of Nature].

Calcium concentrations (ppb) covering the period 10-20 ka based on GISP2 ice core data. The sample resolution is approximately 2 years through the Holocene, a mean of 3.48 years within the YD (Younger Dryas) and BA (Bolling/Allerod), and 3-15 years during the OD (Older Dryas) [from Mayewskii et al., 1993, by permission of AAAS]. High calcium and dust concentrations during the YD point to an intensified atmospheric circulation over continental regions and increased aridity.

Inferred changes in Artemisia, sea surface temperature and sea ice cover in the North Sea during deglaciation [from Rochon et al., 1998, by permission of Elsevier]. The YD was associated with strong cooling in the North Atlantic.

Deglaciation

Notable Events

• •

• • •

The 8.2 ka cooling event – another thermohaline disruption The Holocene Thermal Maximum (HTM), which saw disappearance of the major continental ice sheets. Its onset is linked to perihelion in July and stronger axial tilt, leading to July solar radiation at 65 deg. N being at a maximum positive anomaly at around 9 ka. However, the onset and termination of the HTM varied strongly by region, in part reflecting proximity to the shrinking ice sheets. Late Holocene Cooling: Climatic deterioration after the HTM. The Medieval Warm Period (1000-1200 AD, North Atlantic Sector) and Little Ice Age (16th and 17th centuries, but variable) Warming over the past 100 years

The 8.2 ka cooling event Following the YD, the climate warmed, the continental ice sheets shrank and sea level rose. However, the GISP-2 ice core and other data sources reveal that around 8.2 ka, temperatures in the vicinity of Greenland cooled about 3.3°C over a period of about two decades. The cold event lasted about 150 years. European climate was affected, with temperatures dropping about 2°C. Some evidence exists from speleothems, ocean sediments and an ice core that parts of the tropics became drier. Shrinking of tropical wetlands in a drier climate might explain the 10-15% drop in atmospheric methane recorded in air bubbles of Greenland ice cores.

Current thinking again invokes Lake Aggasiz, which formed south of the Hudson Bay from meltwater, dammed to the north by a remnant of the Laurentide ice sheet (figure at left). The dam failed, releasing the waters to the Hudson Bay and downstream to the Labrador Sea, changing the density structure of the ocean, slowing deepwater formation and the thermohaline circulation.

Geography of the Hudson Bay region just prior to the 8.2 ka event. From Clarke et al. 2003, http://www.ncdc.noaa.gov/paleo/abrupt/data5.html

The Holocene Thermal Maximum (HTM)

Variations in the onset and termination of the Holocene Thermal Maximum (HTM) over northern North America (left) and some of the proxy records documenting the event (right) [Source: Kaufman et al., 2004].

Records of Northern Hemisphere temperature variations over the last 1300 years. Panels are (top) annual temperature over the instrumental record, (middle) reconstructions using various proxies, (bottom) overlap of all proxy records in middle panel with shading indicating level of agreement between the different reconstructions. The observed temperature record in the bottom panel is shown in black [Source; IPCC-AR-4, Working Group I Report, Figure 6.10]. The “Medieval Warm Period” from about 1000-1200 seems to have had it strongest expressions over the Northern North Atlantic sector. The “Little Ice Age” is dated anywhere from 1250-1920 to 1550-1850. The GSIP-II Greenland ice core records put the maximum cooling from 1579-1730.

Arctic summer temperature anomalies for the past 2000 years based on a variety of proxy sources. The blue line shows a reconstruction of summer Arctic land temperatures over the last 2000 years, based on a composite of 23 proxy records from lake sediments, ice cores, and tree rings relative to 1961-1990 reference period. The shaded area represents variability among different Arctic sites. The red line shows the recent Arctic warming based on instrumental observations [adapted from Kaufman et al. 2009].

Reconstructed Arctic temperatures over the past 400 years (each panel) along with estimated time series of (a) methane (b) atmospheric carbon dioxide concentration (c) total irradiance (d) sulfate. Asterisks indicate known large volcanic events [from Overpeck et al., 1997, by permission of AAAS]. Solar irradiance seems to have played a significant role in temperature variability up to about the middle of the 20th century.

Recent global warming

http://data.giss.nasa.gov/gistemp/graphs/