LAYER-BOUNDING SURFACES IN STALAGMITES AS KEYS TO BETTER CHRONOLOGIES AND PALEOCLIMATE RECORDS

Petrographic recognition of layer-bounding surfaces in stalagmites offers an important tool in constructing paleoclimate records. Previous petrographic efforts have examined thickness of layers (a possible proxy for annual rainfall) and alternation of layers in couplets (a possible indicator of seasonality). Layer-bounding surfaces, in contrast, delimit series of layers and represent periods of non-deposition, either because of exceptionally wet or exceptional dry conditions.

Two types of layer-bounding surfaces have been recognized. Type E layer-bounding surfaces are surfaces at which layers have been truncated or eroded at the crest of a stalagmite. Keys to their recognition include irregular termination of layers otherwise present on the stalagmite's flank, dissolutional cavities, and coatings of non-carbonate detrital materials. Type E surfaces are interpreted to represent wet periods during which drip water became so undersaturated as to dissolve pre-existing stalagmite layers, and thus they necessarily represent a hiatus in the stalagmite record.

Type L layer-bounding surfaces are surfaces below which layers become thinner upward and have lesser lateral extent upward, so that that the stalagmite's layer-specific width decreases. They are thus surfaces of lessened deposition and are interpreted to represent drier conditions in which drip rate slowed so much that little deposition occurred. A Type L surface may, but does not necessarily, represent a hiatus in deposition. Experience shows that Type L surfaces commonly represent significant hiatuses.

These surfaces are significant to paleoclimate research both for their implications regarding climate change (exceptionally wet or dry conditions) and in construction of chronologies in which other data, such as stable isotope ratios, are placed. Recognition of layer-bounding surfaces and thus hiatuses implies temporal gaps in chronologies, and it shifts stable-isotope data into discrete intervals in which those data are more closely spaced. Attention to changing thickness of annual layers and thus to accumulation rate can also refine a chronology. A chronology constructed with attention to layer-bounding surfaces and to changing layer thickness is much more accurate than a chronology in which hiatuses are not recognized at such surfaces.