The use of long-term biological data to generate testable hypotheses

Abstract

Long-term biological data sets are typically collected, analyzed for their periodicities, possibly correlated with some suspected forcing function(s), published and forgotten. To fully utilize such long-term data sets for management decisions and basic scientific inquiry, they should be used to generate new ideas and as the basis for testing new hypotheses. Rarely is this done! Ten plus years of monthly or fortnightly data on meiofauna abundance from a mud and a sand site in North Inlet, South Carolina, indicate very different seasonal patterns, and thus controlling mechanisms, at the two sites. The mud site faunal abundance appears predator controlled; the sand site fauna physically controlled. Using the 10+-year data set as the base, a series of experiments are proposed to test whether the mud meiofauna are predator controlled. Additionally, the copepod species composition at the sand site has changed over the sampling period; interstitial species have almost disappeared, while epibenthic species abundance has remained constant. The elimination of the interstitial forms is thought to be the result of increasing silt/clay at the site. Experiments are proposed to test the effect of changing silt/clay content or species composition in sandy habitats.