Dataset: Artificial seagrass faunal counts

Experiment #1: Deployments 1 and 2 were part of this experiment, which took place in central San Diego Bay, California, USA. The objective of this experiment on seagrass habitat fragmentation was to determine the interactive effects of habitat loss and structural complexity on epifaunal diversity, abundance, and biomass. Structural complexity was represented by shoot density, which varied across three levels (low SC (160 shoots m-2 = 10 shoots per ASU), intermediate SC (512 shoots m-2 = 32 shoots per ASU), and high SC (1024 shoots m-2 = 64 shoots per ASU). Habitat loss varied over 10 levels, from 0 to 90% loss (specifically, 0, 10, 20, 40, 60, 70, 75, 80, 85, or 90% habitat loss). These variables were fully crossed to allow a determination of how structural complexity influences the relationship between habitat loss and epifaunal community structure. For this experiment, epifauna are crustaceans, molluscs, and other taxa that are residents in seagrass habitat (size range ca. 0.5 mm - 2 cm in body length). The experiment was conducted by creating 30 plots of artificial seagrass (each 2 m x 2 m), each of which was modular such that small sections (known as ASUs, for “Artificial Seagrass Unit”) of each could be removed to simulate habitat loss. Each ASU consisted of a 25 x 25 cm square of black plastic mesh to which were tied 64 artificial eelgrass shoots. Each artificial seagrass shoot was a 1 m long piece of green polypropylene ribbon folded in half and tied to the mesh to form two 50 cm tall simulated eelgrass blades. Plots were laid out in the shallow subtidal zone of San Diego Bay, were allowed to be colonized by organisms for 45 d, and then were experimentally fragmented by removing the appropriate number of modules from each. One month later, samples were collected from each plot. Four ASUs were collected from each plot, rinsed into sieves, and animals preserved in 10% formalin.

Epifauna were sorted in the laboratory using dissecting microscopes. To make sorting more efficient, animals first were sorted into two sizes, those that were retained on a 2.8 mm sieve, and those that were retained on a 0.5 mm sieve. Animals were sorted to species, genus, or family (lowest category possible) using field guides such as the Light and Smith Guide to Intertidal Invertebrates. Training for taxon identification was aided by digital photographs taken for most taxa. Specimens of small, similar looking taxa (e.g. amphipods) were sent to other benthic ecology laboratories for confirmation of identification.

Experiment #2: Deployments 3 and 4 were part of this experiment, which took place near the mouth of San Diego Bay, California, USA, adjacent to Shelter Island. The objective of this experiment on seagrass habitat fragmentation was to determine the interactive effects of predator access and habitat patchiness on epifaunal diversity, abundance, and biomass. The general design of the experiment matched that of Experiment #1, except that all plots consisted of the same structural complexity, and that plots were not fragmented after being colonized (instead, plots were deployed pre-fragmented [10 levels] and were colonized in that state).  Instead of varying structural complexity, predator access to plots was varied by creating three treatments: caged ASUs (full wire mesh cage of selected ASUs to restrict predators, but allow colonization by epifauna), open ASUs (no cage), and cage-control ASUs (wire mesh cages with two sides removed to allow access by predators, but otherwise identical to full cages). Cages and cage-controls were placed over three separate AUSs in each plot for the duration of the soak. To sample plots, three ASUs were retrieved from each plot and handled identically to Experiment 1. For deployment 3, a shoot density of 1024 shoots m-2 was used in all plots. For deployment 4, a shoot density of 106 shoots m-2 was used in all plots.