LIMITATIONS OF HENDY TEST CRITERIA IN JUDGINGTHE PALEOCLIMATIC SUITABILITY OF SPELEOTHEMSAND THE NEED FOR REPLICATION

摘要

Carbon and oxygen isotopes in calcite speleothems are powerful proxies forunderstanding past climate change. For calcite deposited under isotopic equilibriumconditions, variations in δ18O values directly reflect changes in cave temperature and theisotopic composition of meteoric water. Speleothem 613C values have bedrock,atmospheric, and soil gas sources. Soil gases can be traced to the overlying vegetation,which is related to climate. Both δ13C and δ18O values are therefore potentially powerfultracers of climate change. Processes that could alter speleothem δ13C and/or δ18O values,and thereby mask primary environmental signals, fall in the categories of 1) kineticprocesses, including deposition of calcite out of isotopic equilibrium, and 2) vadoseprocesses, including evaporation of water at or near the land surface. In truth, there is noabsolute test for the absence of these kinetic/vadose-zone processes. However, the HendyTest is widely used for assessing whether isotopic equilibrium existed during the time ofcalcite deposition. Criterion (1) of the Hendy Test (i.e., that δ18O values remain constantalong a single growth layer) may not be a valid control of equilibrium conditions becauseisotopic equilibrium could theoretically occur in the center of the speleothem at the sametime that kinetic fractionation occurs at the flanks. Moreover, the concept of samplingalong a single growth layer is flawed in both theory and practice. Criterion (2) of theHendy Test (i.e., that there is no relationship between δ13C and δ18O) is based on theassumption that speleothem δ13C values are not linked to climate. However, speleothemδ13C values may well be linked to climate because provides a first-order controlon soil productivity and the type of vegetation. Therefore, Hendy Test criterion (2) is nota prerequisite to isotopic equilibrium in all cases. We propose instead the ReplicationTest (i.e., the demonstration of similar isotopic profiles among two or more speleothems)for evaluating the likelihood of calcite deposition under isotopic equilibrium conditions.Replication of isotopic profiles among two or more speleothems is possible only ifkinetic/vadose-zone processes are either: 1) absent or 2) have affected spatially separatedspeleothems in exactly the same way. Because the second scenario is highly unlikely, wepropose that the Replication Test is effectively sufficient in ruling out kinetic/vadose-zone overprinting processes. We further suggest that the Replication Test is far morerobust in testing for the absence of the wide range of processes described above than isthe traditional Hendy Test.