The horseshoe crab (Limulus polyphemus) is an arthropod more closely related to spiders than other crabs and has a body form that evolved more than 200 million years ago. They are found from Nova Scotia to Mexico and live year-round in Long Island Sound. They are harvested as bait for other fisheries, but are not typically eaten for meat. They come on shore to spawn in May and June, with spawning tending to peak at night around the new and full moons.
Since horseshoe crabs rely on beaches and the shallow intertidal environment to produce their young, their abundance is an indicator for the health and productivity of this transitional environment. Horseshoe crab eggs are an essential food source for migrating shorebirds, and their larvae are consumed by many fish. They are most valuable to humans for their blood proteins which are extremely sensitive to bacteria. The extracted compound, known as LAL, is used to screen injected drugs and implanted biomedical devices for contamination.
Both the CT and NY indices show a stable or increasing abundance in western Long Island Sound through the 1990s, with high abundance in 2003, 2007, and 2011. Although abundance was equally strong in the eastern Sound and Peconic Bay during the early 1990s, it has declined since 1995, currently reaching among the lowest values in each of the time series.
CTDEEP’s survey samples mainly in the western and central Sound, while the Millstone Environmental Laboratory dataset in Niantic Bay reflects abundance in the eastern Sound near the Connecticut shoreline. The Manhasset and Little Neck Bays NYSDEC seine survey reflects conditions in western Long Island Sound near the New York shoreline, while the seine survey in the Peconics, which borders Long Island Sound, is indicative of conditions in eastern Long Island Sound.
The CTDEEP data is from the spring trawl survey.
The survey uses geometric mean instead of arithmetic mean to find the most frequently observed number of fish collected per tow. In a natural environment such as Long Island Sound fish have a “patchy” distribution, i.e. some areas will have a very high abundance of fish, and other areas very low abundance or no fish at all. An arithmetic mean can be easily biased by unusually high or low values so that it doesn’t reflect the true center of a data set. The geometric mean minimizes the effects of very high or low values using a log transformation and is a better average for this type of biological data.