Comparison of Methods for Processing Sweep-Net Samples of Aquatic Invertebrates From Forested Wetlands

Abstract

Abstract Aquatic invertebrate data are useful for assessing wetland community structure, function, and water quality. Although collecting samples of aquatic invertebrates is relatively efficient and economical, processing these samples can be time consumptive and costly. Accordingly, researchers have devised methods to increase processing efficiency and effectiveness. For example, supersaturated solutions of solutes in various aqueous media have been used to separate invertebrates from aquatic media and organic matter. However, no study has evaluated this method for extracting invertebrates from sweep-net samples from flooded bottomland hardwood forests. We compared invertebrate recovery rates from samples processed using 1) tap water (control), 2) a supersaturated solution of sugar and water, and 3) a supersaturated solution of salt and water. We also evaluated a subsampling procedure by comparing taxonomic Order-level richness and Shannon diversity between sub- and whole samples. Numbers and dry biomass of invertebrates recovered were similar among the three aforementioned treatments. Use of supersaturated solutions did not reduce processing time, because invertebrates and leaf litter both floated instead of separating. Thus, we recommend using only tap water in processing sweep-net samples of invertebrates from forested wetlands that contain abundant leaf litter. Overall, we recovered 72.2% (95% CI = 3.0%) of all detected invertebrates and 48.0% (95% CI = 7.5%) of invertebrate biomass. Invertebrates that weighed less than others (e.g., mosquito larvae; Diptera) were more efficiently recovered than were heavier taxa (e.g., snails; Pulmonata). Our subsampling method underestimated Order-level richness and diversity indexes by 12% and 19%, respectively. However, processing subsamples was nearly two times faster than processing whole samples. Our method of using a sieve to subsample invertebrates is appropriate when numerical abundances are desired, because most (70%) invertebrates were detected and recovered.