USGS Home
Contact USGS
Search USGS

The chironomids we sampled in this study also were benthic, and fewer were estimated with the multiple tube sampler. Like the oligochaetes, over half of the chironomids in the pooled samples came from the water column sample. We could not judge the accuracy of each respective sampling method, but the higher estimates of the pooled sample, along with the large proportion of individuals caught in the water column sampler, seemed to suggest that this benthic organism was overestimated by the pooled sampling method.

Cladocera are pelagic organisms but may be associated with the benthic-pelagic interface at certain times of the day. The multiple tube sampler yielded density estimates higher than the pooled estimate, which suggested that pooled samples may underestimate this taxa. Twelve percent of the cladocera in the pooled sample were sampled by the benthic sampler. The benthic sampler caught cladocera as it descended through the water column before reaching the bottom of the wetland. As the device was forced into the sediments, water and pelagic organisms were forced through the unstoppered hole at the top of the sampler. It is possible that avoidance by cladocera contributed to the low densities estimated by our pooled sample; however, we feel it is more likely that an adequate seal was not obtained at the benthic-pelagic interface (i.e., sampler stopped short of the interface or tilted slightly so that a gap resulted), and cladocera were lost before the door was closed.

Cyclopoid copepods are associated with the benthic-pelagic interface, but we have observed them to migrate higher in the water column at times. As with cladocera, density estimates from the multiple tube sampler were significantly higher than the pooled estimate. Unlike the other invertebrate taxa, however, the 2 sampling methods were influenced by type of wetland. In the seasonally flooded wetland, both sampling methods gave similar results; whereas estimates from the semipermanent wetland were significantly higher with the multiple tube sampler. The seasonal wetland had approximately 8-10 cm of decomposing vegetation at the benthic-pelagic interface, whereas the semipermanent wetland was free of recalcitrant vegetation. Thus, pond architecture differed between wetlands. Because of low oxygen availability in decomposing vegetation, it is likely that copepods hovered above the decomposing vegetation rather than at the benthic-pelagic interface. As with cladocera, difficulties in achieving an adequate seal at the benthic-pelagic interface with the modified water column sampler would result in undersampling as our data suggest. Although 76% of the copepods in our pooled sample were obtained from the water column sampler, it is not clear whether they were suspended higher in the water column or whether their presence in the water column represented a sampling bias.