RECONSTRUCTING ANNUAL AND SEASONAL CLIMATIC RESPONSES FROM VOLCANIC EVENTS SINCE AD 1270 AS RECORDED IN THE DEUTERIUM SIGNAL FROM THE GREENLAND ICE-SHEET PROJECT 2 ICE CORE

Citation
De. White et al., RECONSTRUCTING ANNUAL AND SEASONAL CLIMATIC RESPONSES FROM VOLCANIC EVENTS SINCE AD 1270 AS RECORDED IN THE DEUTERIUM SIGNAL FROM THE GREENLAND ICE-SHEET PROJECT 2 ICE CORE, JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES, 102(D16), 1997, pp. 19683-19694
Citations number
39
Categorie Soggetti
Metereology & Atmospheric Sciences
Volume
102
Issue
D16
Year of publication
1997
Pages
19683 - 19694
Database
ISI
SICI code
Abstract
A 200 m ice core at the GISP 2 site (Summit, Greenland) provides a 720 year record of volcanic eruptions, identified by pronounced increases in SO42- in relation to the background concentration. Deuterium/hydro gen isotope ratios (delta D) measured in the same core provide a proxy climate record (temperature) used to examine the response of the clim ate system in Greenland to these eruptions. A group of 34 known volcan ic events and a subgroup of the six largest volcanic events are isolat ed and examined. For both groups, significant negative (cooler) isotop ic excursions occur beginning the year the volcanic event is detected in the ice core. Cooling is significant for 1 to 2 years. The maximum isotopic excursion averages 4 parts per thousand (similar to 0.7 degre es C) for all 34 events and 10 parts per thousand (similar to 1.8 degr ees C) for the subgroup of largest events. For individual events the r esponse in the isotope signal to volcanic events is highly variable bu t occurs at most 1 year before SO42- is detected in the core. Examinat ion of subannual isotope data suggests that peak winter isotope values are lowered more than peak summer values. We hypothesize that this re sult is flawed by postdepositional alteration of the isotopic profile by vapor diffusion. To test this hypothesis, we use numerical deconvol ution to account for the smoothing effects of vapor diffusion in the f un. Analysis of the deconvolved data shows greater summer than winter cooling, a result more consistent with instrumental temperature data.