Short-term changes in channel form and macro invertebrate communities following low-head dam removal

Although >70 dams have been decommissioned in Wisconsin over the past 30 y, little is known about the physical and ecological effects of dam removal on riverine ecosystems. The purpose of our study was to document changes in channel form and macroinvertebrate assemblages following the removal of a low-head, run-of-river dam from the Baraboo River, Wisconsin, in January 2000. We surveyed cross sections and collected benthic macro invertebrate samples in 6 reaches (an upstream reference reach, reaches immediately above and below the dam that was to be removed, and sequential unimpounded and impounded reaches further downstream) in a multiple-dam system. Survevs were conducted in December 1999, before dam removal, in March 2000 immediately after dam removal, and in August 2000 following a flood. Benthic sediments were collected from selected sites in March and August to measure particle size shifts associated with the dam breach. Before dam removal, impounded reaches were characterized by relatively deep, wide channels, extensive deposits of loose sediments, and macroinvertebrate taxa characteristic of lentic or depositional habitats. Removal of the dam significantly decreased the cross-sectional area of the active channel in the former impoundment from 59 m(2) to 11 m(2), but did not alter channel form in downstream reaches. However, we found an increase in loose sediments and in the relative abundance of the sand fraction (0.5-2.0 mm) below the dam immediately after it had been removed. A flood in June increased cross-sectional area in the former impoundment by widening the channel. Sediments that had settled in the reaches below the dam in March were transported similar to3.5 km downstream, where they became trapped in another impoundment. The flood had little or no detectable effect on the other 5 study reaches. Within 1 y of dam removal, macroinvertebrate assemblages in formerly impounded reaches did not significantly differ from those in either the upstream reference site or in other unimpounded reaches below the dam site. All unimpounded sites were characterized by lotic taxa such as netspinning caddisflies and heptageniid mayflies regardless of their impoundment history. Thus, dam removal caused relatively small and transient geomorphic and ecological changes in downstream reaches, and apparently rapid channel development to an equilibrium form within the impoundment, associated with both dam removal and the subsequent June flood. These muted changes and rapid recovery likely result from the relatively large channel size and the small volume of stored sediment available for transport following dam removal.