To what extent do water isotope records from low accumulation Alpine ice cores reproduce instrumental temperature series?

• Published in: Tellus B: Chemical and Physical Meteorology
• Main Authors: Pascal Bohleber, Dietmar Wagenbach, Wolfgang Schöner, Reinhard Böhm
• Format: Article in Journal/Newspaper
• Language: English
• Published: Stockholm University Press 2013
• Subjects: Alpine ice cores; stable water isotopes; temperature proxy record; Meteorology. Climatology; QC851-999; Monte Rosa; ice core
• Online Access: https://doi.org/10.3402/tellusb.v65i0.20148; https://doaj.org/article/cbd3c6af556b4733b2b9c3aadd18dc15

Among Alpine ice core drilling sites, the Colle Gnifetti glacier saddle situated in the Monte Rosa summit range is the only one whose net snow accumulation rate is low enough to offer climate records back to some 1000 yr. It is demonstrated that the strong snow erosion at this site particularly hampers the interpretation of stable water isotope records δ18O, δD in terms of atmospheric temperature changes. We evaluate the δ18O records from four Colle Gnifetti cores for their common variability to extract a composite isotope record that may be compared with the instrumental temperature evidence. Time series analyses over the last 120 yr reveal that the common δ18O signal is mainly reflected in the low frequency variability, starting at the decadal scale. Comparing the correspondingly smoothed composite record to the high-elevation temperature time series (specifically adjusted to the seasonality of the net snow accumulation) reveals the following findings: On the decadal scale, the isotope variability correlates with the temperature record at around R=0.65 but is interrupted by three, ca. 10-yr long mismatch periods. The multidecadal isotope signal closely reflects the strong overall 20th century temperature increase, thereby showing an up to three-fold higher isotope temperature sensitivity than commonly assumed. Over the entire instrumental period back to 1760, five more such mismatch periods are embedded in the generally coherent pattern of the δ18O and instrumental temperature records (including the strong overestimate of the temperature around 1850 by the isotope temperature proxy). For the early instrumental period (1890–1760) characterized by a comparably weak long-term temperature trend, the isotope signal generally suggests warmer conditions of about 0.4°C compared to instrumental data.