Alan E. Taylor

RECORDS OF CLIMATIC CHANGE IN THE CANADIAN ARCTIC : TOWARDS CALIBRATING OXYGEN ISOTOPE DATA WITH GEOTHERMAL DATA

Borehole temperature-depth data have been combined with oxygen isotope data from an ice core at a nearby location in the Canadian Arctic to calibrate the oxygen isotope data to ground temperature histories. The relationship of oxygen isotope data to ground surface temperature history is δ18O = 0.42ΔTg and this reproduces the main features of the observed subsurface temperature profile measured at the borehole site. These results suggests that oxygen isotope data are a good representation of the regional, long term climatic variations at these latitudes. Discrepancies between the observed subsurface profile and the theoretical profile calculated from the calibrated oxygen isotope time series may be explained in terms of the different character of the records, in particular snow cover, and active layer processes.

Marine transgression, shoreline emergence: Evidence in seabed and terrestrial ground temperatures of changing relative sea levels, Arctic Canada

Precise temperatures to depths of several hundred meters have been measured in abandoned petroleum exploration wells in coastal and offshore areas of the Canadian Arctic. In these regions, changes in relative sea level during the Quaternary have left a strong thermal signature on deep ground temperatures. Surface temperature changes of the order of 10–20 K accompanied shoreline emergence due to uplift in some regions and marine transgression due to eustatic changes in sea levels in others. Classical, analytic techniques are used to demonstrate that these anomalous features are a direct consequence of recent changes in sea level. In coastal areas of the Queen Elizabeth Islands, modelling of the large, near-surface temperature-depth gradients and the distinct curvature in the measured profiles yield emergence times in general agreement with dates taken from emergence curves. The analysis suggests that surface temperatures of these coastal areas were 16–19 K higher than at present from the Late Wisconsinan until emergence in the Holocene, suggesting either marine conditions for this period or glacial ice cover with basal temperatures near the pressure melting point. In contrast, at two offshore wells on the Beaufort Shelf, analyses of the negative temperature-depth gradients below the seabed and profile curvature yield times of marine transgression generally consistent with the published sea level curve for the area. The analyses suggest that surface temperatures of some present offshore areas were 10–16 K lower than todays sea bottom temperatures from the Late Wisconsinan until marine transgression in the Holocene. The characteristic ground temperature profiles measured in such arctic areas provide independent evidence for the relative changes in sea level indicated by proxy data. It is especially valuable for these areas, where there is little historical record and from which datable material, such as shells, driftwood, and pumice fragments, may be scarce or ambiguous. These effects may not be observable in temperate areas because of the smaller contrast in sea bottom and terrestrial temperatures.

Terrestrial heat flow from project CESAR, Alpha Ridge, Arctic Ocean

During two months in spring, 1983, a multidisciplinary study, project CESAR, was undertaken from the sea ice across the eastern Alpha Ridge, Arctic Ocean. In the geothermal program, 10 gradiometer profiles were obtained; 63 determinations of in situ sediment thermal conductivity were obtained with the same probe, and 714 measurements of conductivity using the needle probe method were obtained on nearby core. Weighted means of the thermal conductivity of the sediment are 1.26 W/mK (in situ) and 1.34 W/mK (core), consistent with the compacted sediment encountered across the ridge and with the lithology. Calculated terrestrial heat flow values, corrected for the regional topography, range from 37 to 72 mWm−2; the average is 56+/−8 mWm−2. Some temperature and heat flow versus depth profiles exhibit non-linearities that can be explained by physically reasonable (but otherwise unsubstantiated) variations in bottom water temperatures preceding the measurements; models are hypothesized that reduce the curvatures. Two heat flow values considerably higher than others in the area may be explained by higher bottom water temperature over several years, while the low value is consistent with a recent deposition from a slump. This hypothetical modelling reduces the scatter of heat flows and reduces the average to 53+/−6 mWm−2. The CESAR heat flow is somewhat greater than expected for a purely continental fragment but is consistent with crust of oceanic origin. The heat flow is similar to values obtained in Cretaceous back-arc basins. Based on the oceanic heat flow-age relationship, the heat flow constrains the age of the ridge to 60–120 million years. The heat flow observed on other aseismic features in the worlds oceans suggests that the Alpha Ridge has experienced no significant tectono-thermal event in the last 100 million years.

Geothermal inversion of Canadian Arctic ground temperatures and effect of permafrost aggradation at emergent shorelines

We apply traditional geothermal spectrum inversion to precision temperature logs and thermal conductivity from 10 wells in the Canadian Arctic Archipelago (75° to 81°N). Sites lie beyond the Holocene marine limit, and no effect of deep permafrost dynamics is expected. Ground surface temperature (GST) changes correlate with the Little Ice Age and Little Climatic Optimum with average amplitudes relative to 1980 of −2.7 K and +1.6 K, respectively. Results correlate broadly with similar reconstructions for this area and Greenland ice cap holes GRIP and Dye-3 to the southeast. An offshore site in 244 m water yields a Little Ice Age seabed temperature amplitude of −0.7 K, suggesting a moderated climate impact on regional ocean temperatures. Nearshore boreholes where permafrost is aggrading owing to glacioisostatic emergence are excluded; we demonstrate that traditional inversion codes without latent heat of phase change predict the magnitude of the emergence signal but a timing far too recent.