Live and Live-Dead Intraspecific Morphometric Comparisons as Proxies for Seagrass Stability in Conservation Paleobiology

Abstract

Comparisons of life and death assemblages are commonly conducted to detect environmental change, including when historical records of live occurrences are unavailable. Most live-dead comparisons focus on assemblage composition, but morphology can also vary in species with environmental variables. Although live-dead morphologic comparisons are less explored, their data could be useful as a proxy in conservation paleobiology. We tested the potential for geometric morphometric data from live-and dead-articulated Stewartia floridana (Bivalvia: Lucinidae) to serve as proxies for seagrass occurrence and stability. The study area is at the northern end of Pine Island in Charlotte Harbor, FL, United States, an estuarine system with substantial seagrass loss in the 20th century and subsequent partial recovery. The area sampled has had relatively stable seagrass occurrences since at least the early 2000s. Live and dead-articulated S. floridana samples were collected from two transects through a patchy seagrass meadow, with sampled sites ranging from bare sand to 100% seagrass cover. Dead-articulated specimens were also collected from three adjacent transects. Live S. floridana shape covaried significantly with seagrass taxonomic composition and percent cover at the time of collection based on two-block partial least squares analysis, although shape differences between seagrass end members (100% Halodule wrightii and 100% Syringodium filiforme) were not significant by multivariate analysis of variance (MANOVA). Instead, specimens from 100% H. wrightii had significantly greater Procrustes variance. Live S. floridana shape data placed in categories describing seagrass stability over 6 years prior to sampling (and reflecting sclerochronologic estimates of maximum longevity) differed significantly based on MANOVA. For live and dead S. floridana from the same transects, shape differed significantly, but allometric trends did not. In addition, patterns of morphologic variation tied to seagrass stability were detected in dead-articulated valve shape. Dead shells from adjacent transects differed significantly in shape and allometric trend from both live and dead specimens collected together. We infer that morphometric differences recorded fine-scale spatial and temporal patterns possibly tied to environmental change. Therefore, geometric morphometrics may be a powerful tool that allows for death assemblages to track seagrass distributions through time prior to systematic monitoring, including in areas under high anthropogenic stress.