Estimating Antarctic climate variability of the last millennium
Climate variability is determined by the climate system’s internal variability as well as its response to external forcing. A quantitative understanding of past Antarctic climate variability is therefore essential if we are to attribute and to detect anthropogenic influences on the current and futur...
| Main Authors: | , |
|---|---|
| Format: | Conference Object |
| Language: | unknown |
| Published: |
2017
|
| Subjects: | |
| Online Access: | https://epic.awi.de/id/eprint/44824/ https://epic.awi.de/id/eprint/44824/1/poster_osm2017.pdf https://hdl.handle.net/10013/epic.51067 https://hdl.handle.net/10013/epic.51067.d001 |
| Summary: | Climate variability is determined by the climate system’s internal variability as well as its response to external forcing. A quantitative understanding of past Antarctic climate variability is therefore essential if we are to attribute and to detect anthropogenic influences on the current and future climate in Antarctica, and thus crucial for projecting the evolution of the Antarctic ice sheet. Analysis of stable water isotope data from ice cores in principle provides information on past temperature variability, but its quantitative interpretation is challenged by strong non-climate effects. So far, the magnitude and timescale dependency of both the climate signal and the noise in Antarctic isotope records remains largely unknown. Here, we present a new spectral method to separate climate signal and noise in a large collection of published and new annually-resolved ice core records from East Antarctic Dronning Maud Land and the West Antarctic Ice Sheet, spanning the last 200—1000 years. With this, we derive the first timescale-dependent estimate of Antarctic temperature variability and isotopic signal-to-noise ratio on decadal to centennial time scales. In contrast to the raw isotope data, we find a stronger increase in temperature variability on longer time scales, which is similar between the two study regions and to estimates from reanalysis and marine SST data. Spatial analysis of the estimated noise levels allows the separation of local stratigraphic noise from larger-scale noise due to precipitation intermittency. Signal-to-noise ratios only reach values above one for multi-centennial time scales. Our findings illustrate a consistent way of interpreting isotope records, but also highlight the remaining knowledge gaps in our understanding of Holocene climate and ice-core derived variability. We emphasize that our new method is applicable for distinguishing climate variability from local effects for any spatial, well-dated array of proxy temperature records. |
|---|