EXAMINING LIFESPAN BIAS IN DEATH ASSEMBLAGES USING TWO BIVALVE SPECIES IN SEAGRASS BEDS IN NORTH CAROLINA

Lifespan bias – the overrepresentation of shorter-lived species in death assemblages compared to their actual abundance in a living community – is one possible source of bias in the fossil record. This project field tested the possibility of lifespan bias using the bivalve species Mercenaria mercenaria and Chione elevata collected from seagrass beds near Beaufort, North Carolina. Samples were collected from an intertidal seagrass bed over four days in June, 2014, February, 2015, and March, 2015. The ratio of shorter-lived C. elevata to longer-lived M. mercenaria increased from 1.125:1 in the live community to 2.560:1 in the death assemblage, which is consistent with lifespan bias. Calculations were performed to determine how much of the observed increase in C. elevata could be explained by lifespan bias. The calculations incorporated mortality rates from local, regional, and species-wide maximum ages, and mortality rates based on age-frequency distributions from the samples. Results indicate that mortality rates derived from age-frequency distributions are the worst predictor of the discordance in ratio between live and dead samples. This result may be due to alterations in the age-frequency distribution due to M. mercenaria harvest. Mortality rates based on species-wide published maximum ages (C. elevata t_max = 11; M. mercenaria t_max = 106) overestimated discordance by 74%. Lifespan bias best explained the observed discordance in the ratio of C. elevata to M. mercenaria in the seagrass bed when mortality rates were calculated from local and regional maximum ages (C. elevata t_max = 12 or 11, respectively; M. mercenaria t_max = 28 or 46, respectively). Predictions based on the regional t_max over-predicted discordance by 22%; predictions based on local t_max under-predicted discordance by 27%. These results will be compared to those for samples collected in August 2015 from a second, nearby seagrass bed, but preliminary data indicate that lifespan bias may be an important factor in explaining live-dead discordance of proportional abundances when comparing species with similar life modes and preservation potential.