Timescale Calculations for Ice Core Drilling Sites on the Temperate Ice Caps in Iceland

Modelling of age vs. depth profiles and annual-layer thickness changes with depth in ice sheets forms part of the investigations carried out prior to the selection of ice core drilling sites. The well known Nye model, which assumes a constant vertical strain rate with depth in an ice sheet of thickness H is generally applicable in the upper half of polar and temperate ice caps, but the assumption of a constant vertical strain rate is unrealistic near the bed in an ice sheet frozen to bedrock. Dansgaard-Johnsen (D-J) type models assume that the vertical strain rate is constant down to height h above bedrock and then decreases linearly with depth towards zero at the bed. The parameter h can be calibrated according to the way in which the horizontal velocity varies with depth. Here we introduce a new derivation of the D-J model that accounts for bottom melting due to the geothermal heat flux, which averages 200 mW/m2 in Iceland. The model is then applied to five different locations on the temperate ice caps in Iceland, with ice thicknesses varying between 220 m and 850 m and accumulation rates ranging between 2.0 and 3.6 m ice/year. Data from ice cores drilled at three of these sites are used to calibrate the model. For the summit location on the Hofsjokull ice cap (H = 300 m), we find that a D-J model with a relatively high h/H ratio reproduces the timescale from a 100 m ice core better than the Nye model. Results indicate that a continuous precipitation record covering the last 400-500 years could be retrieved at the Hofsjokull summit (1790 m a.s.l.), and the assumption of bottom melting has a large effect on the modelled timescale at this site, yielding 50% lower ages at 90% of the ice depth than model runs that neglect bottom melting. For deeper drillings in Iceland, the ice-filled caldera at Bardarbunga, NW-Vatnajokull (H = 850 m), where a 415 m core was drilled in 1972, is among the most promising sites. Selection of the h/H ratio in the D-J model for timescale calculation within the caldera rims is complicated by an unusual ice-flow pattern but results strongly indicate that a 700-800 m ice core could yield a record covering historical time in Iceland (870 AD - present). Model results predict that by 90% of ice depth, the annual layers have thinned to 17 cm at the Hofsjokull summit and 8 cm within the Bardarbunga caldera. Annual layers of this thickness are detectable with the methods used in pilot ice core drilling and processing efforts in Iceland in recent years.