HR: 1330h
AN: PP52A-0960    [PDF]
TI: Multi-century climate change determined from borehole temperatures: an examination of bias
AU: * Chapman, D S
EM: dchapman@mines.utah.edu
AF: University of Utah, Department of Geology and Geophysics, 135 S 1460 E, Salt Lake City, UT 84112 United States
AU: Harris, R N
EM: rnharris@mines.utah.edu
AF: University of Utah, Department of Geology and Geophysics, 135 S 1460 E, Salt Lake City, UT 84112 United States
AU: Bartlett, M G
EM: bartlett@mines.utah.edu
AF: University of Utah, Department of Geology and Geophysics, 135 S 1460 E, Salt Lake City, UT 84112 United States
AB: Northern hemisphere, continental surface temperature reconstructions from borehole temperature profiles suggest roughly twice the warming since preindustrial times as inferred from some proxy studies. Several possible causes of this discrepancy have been attributed to biases in borehole temperature records including: 1) inadequate spatial sampling; 2) noise in borehole temperatures; 3) different frequency sampling; 4) land-surface changes over time; and 5) seasonal snow cover influences. Sampling biases and the ability of surface ground temperature (SGT) to track changes in surface air temperature (SAT) at frequencies appropriate for climate change studies are presented elsewhere in this session. This paper focuses on seasonal snow cover influences and specifically a recent publication (Mann and Schmidt, GRL, 2003) erroneously concluding that "past SAT trends from borehole-based SGT reconstructions may therefore be substantially biased by seasonal influences and snow cover changes." There are several flaws in the Mann and Schmidt analysis. (1) Borehole temperatures respond to a continuous temperature signal at the surface, therefore any useful comparison of SGT and SAT tracking must be based on continuous (i.e. annual) signals rather than separated seasonal effects. When one makes the appropriate comparison for climate tracking, the correlation coefficient between annual SGT and SAT changes using their model results is extremely high (r=0.96). (2) Borehole temperatures integrate surface temperature fluctuations, providing a low-pass filter that is particularly useful in climate change studies. Thus while the Mann and Schmidt focus on explaining seasonal variance may be interesting from other aspects it has little relevance for tracking climate change. (3) The Mann and Schmidt result that snow can have both a warming and cooling effect on SGT relative to SAT agrees with our previous model studies and field observatory results. However, it is misleading for them to use end points in short, fluctuating time series, rather than statistical trends to conclude that "mean SGT increases are ${0.2\deg}$ C less than those in SAT." Slightly changing the time interval reverses this particular conclusion. In summary, by using the Mann and Schmidt model output and an appropriate analysis we find that SGT tracks SAT extremely well, and that their modeled seasonal snow cover changes are unimportant at the relevant frequencies for climate change studies. The borehole temperature record of climate change remains a robust indicator of warming and indicates greater climatic sensitivity than some hemispheric proxy reconstructions.
DE: 1600 GLOBAL CHANGE (New category)
DE: 1645 Solid Earth
DE: 3322 Land/atmosphere interactions
DE: 3344 Paleoclimatology
SC: Paleoceanography and Paleoclimatology [PP]
MN: 2003 Fall Meeting