 |
| Titel |
Diffusion length history over the last 16 ka based on a high resolution δ¹⁸O record from NGRIP. Implications for glaciological and paleoclimatic studies. |
| VerfasserIn |
Vasileios Gkinis, Sebastian B. Simonsen, Susanne L. Buchardt, Bo M. Vinther, James W. C. White |
| Konferenz |
EGU General Assembly 2013
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 15 (2013) |
| Datensatznummer |
250071943
|
|
|
|
|
|
| Zusammenfassung |
| The Holocene epoch as seen in the water isotopic records of polar ice cores is described by
a relatively stable climate characterized by minimal fluctuations in temperature.
Arguably, the most commonly used proxy in ice core studies, the ratios of water’s
stable isotopes, provide an insight in past temperatures via a linear relationship with
temperature, commonly referred to as the isotope slope. However, the validity of this slope
has been extensively debated. Based on borehole thermometry and gas isotope
fractionation studies, it has been shown that temperature changes over the Bølling –
Allerød and Younger Dryas transitions as well as several interstadial events have been
underestimated by the water isotope slope. Additionally, isotopic artifacts related to ice sheet
elevation changes, apparent between 6 and 10 ka b2k, result in a poor or even absent
representation of the Holocene climatic optimum in the δ18O record from Greenland
ice cores, contrary to what other paleoclimatic records from Northern latitudes
indicate.
In this study we present ongoing work on the use of the firn isotopic diffusion lengths as a
high resolution proxy of the snow and firn temperature. Our reconstruction is based on the
high resolution δ18O dataset from NGRIP. Water isotope diffusion is a process that occurs
after deposition of the precipitation and takes place in the porous space of the firn until the
close off depth. Assuming a diffusivity parameterization and based on a densification and
strain rate history, it is possible to investigate the effects of temperature and accumulation on
the diffusion length.
By inverting the model we produce a temperature reconstruction for the last 15 ka. This
temperature signal is independent of factors like the water vapor source location and
temperature, the intensity of the atmospheric inversion over the deposition site and the
presence or not of clear sky precipitation. In order for the reconstruction to reproduce the
long term climate signal, a correction for the thinning function is required. Under the
assumption that the GICC05 chronology is the best available estimate for the age – depth
relationship in the ice, that would require about 10 – 15% lower accumulation rates at the
time of the climatic optimum.
The temperature reconstruction is able to infer a Younger Dryas warming signal very
close to what previous borehole thermometry and gas isotope fractionation studies indicate. A
strong 8.2 ky event can be seen in the record and seems to occur in a two stage fashion and
last longer than the raw δ18O signal indicates. Overall, the inferred temperature signal reveals
a significant variance with climatic events that are initially not reflected in the δ18O record.
Some of those events are supported by the findings of other northern hemispheric climatic or
historical records (Medieval and Roman warm periods). The most profound of those events
is a rapid warming occurring between 4 and 5 ky b2k, indicating a clear mid –
Holocene optimum and ending with a rapid cooling at approximately 4.2 ky b2k.
We will comment on the validity of those results as well as the feasibility of the
magnitude of the temperature shifts and propose ways to constrain the findings further. |
|
|
|
|
|
|