RADIOCARBON DATING POLLEN RESIDUES IN A VOLCANIC SETTING

Accurate age-depth models, usually accomplished with radiocarbon dating, are required for the interpretation of paleoecological records in sediment cores. When lacking sufficient terrestrial plant macrofossils or charcoal particles for radiocarbon dating, age-depth models are often developed from a series of radiocarbon dates obtained from pollen residues. Here, we present a case study from two lakes in Yellowstone National Park, Yellowstone Lake (N 44.5393°, W 110.3892°, 2360 m) and Cub Creek Pond (N 44.5054°, W 110.2469°, 2500 m), which are currently under study to reconstruct the hydrothermal history of the Yellowstone Lake basin. Radiocarbon dates were obtained from a variety of materials including charcoal, terrestrial plant macrofossils, bulk sediments, chemically extracted pollen residues, and residues of manually picked Pinaceae pollen. Chemically extracted pollen residues from the Yellowstone Lake and Cub Creek Pond cores consistently produce ages that are ca. 2000 to 5000 cal yrs older than plant macrofossil and tephra ages at the same stratigraphic level. Moreover, manually picked pollen ages from Yellowstone Lake, taken at the sediment-water interface and stratigraphically adjacent to the Mazama Ash (independently aged at ca. 7630 cal yr BP), are up to 4340 cal yrs older than expected, similar to bulk sediment ages from the same level. The absence of superficial carbonate bedrock in either watershed helps to rule out carbonate contamination. However, Yellowstone Lake lies mostly within the Yellowstone Caldera, from which at least 45 ±16 kt of CO₂ emitted daily from a combination of magmatic and pre-Tertiary basement sources (Werner and Brantley 2003, Geochemistry, Geophysics, Geosystems 4, 7). We hypothesize that uptake of degassed CO₂ is the source of these erroneously old dates, similar to other studies (eg. Pasquier-Cardin et al. 1999, Journal of Volcanology and Geothermal Research 92, 195-207) who found old carbon taken up by living plants near active hydrothermal features. Efforts to establish a reliable age-depth model for these sites are ongoing.