Duration of Greenland Stadial 22 Andice-Gas ∆age from Counting of Annuallayers in Greenland NGRIP Ice Core
Climate of the Past http://dx.doi.org/10.5194/cpd-8-2583-2012
The NorthGRIP ice core chronology GICC05modelext is composed of the annual-layercounted GICC05 chronology to 60 kyr before 2000 AD (b2k), and an ice ﬂow model dating the deepest part of the ice core to 123 kyr b2k. Determination of annual strata in ice5 beyond 60 kyr b2k has been challenged by the thinning of annual layers to < 1 cm andthe appearance of microfolds in some early glacial strata. We report high-resolutionmeasurements of a 50 m section of the NorthGRIP ice core and corresponding annual layer thicknesses, constraining the duration of the Greenland Stadial (GS-22)between Greenland Interstadials (GIs) 21 and 22 which occurred between approxi-10 mately 89 (end of GI-22) and 83 kyr b2k (onset of GI-21) depending on the chronologyused. Multiple analytes (insoluble dust particles, electrolytic conductivity, ammoniumand sodium) were determined in annual layers of ice often thinner than 1 cm. Fromannual layer counting, we ﬁnd that GS-22 lasted 2894 ± 198 yr and was followed by aGI-21 pre-cursor event lasting 350 ± 19 yr. Our layer-based counting agrees with the15 duration of GS-22 determined from the NALPS speleothem record (3250 ± 526 yr) butnot with that of the GICC05modelext chronology (2620 yr). These results show thatGICC05modelext overestimates accumulation and/or underestimates thinning in thisearly part of the last glacial period. We also revise the NorthGRIP ice depth-gas depth(5.67 ± 0.18 m) and ice age-gas age (550 ± 52 yr) diﬀerences at the warming onset ofGI-21, observing that δ1520 N increases before CH4concentration by no more than a fewdecades.