Faunal response to fragmentation in seagrass habitats: implications for seagrass conservation

Fragmentation in seagrass systems results in changes to landscape features that may have implications for fauna. We examine published studies to identify whether faunal abundance shows any relationship with patch size of seagrass beds, suggesting preferential use of the edge or interior by seagrass associated taxa. In a series of studies in Tampa Bay, FL, we also examined: (1) the relationship between abundance of both fish and amphipod fauna and seagrass patch size in 24 seagrass (Halodule wrightii) beds (5-93 m(2)) in 1994 and 1995; (2) whether abundance of the infaunal polychaete, Kinbergonuphis simoni, was significantly different within the 1 m edge versus interior of two seagrass (Halodule wrightii) beds of similar size and age; and (3) compared the spatial distribution of the tube-building polychaete, Spirorbis spirillum, in Thalassia testudinum seagrass beds in two sites in Tampa Bay. Neither review of the published literature on fauna and seagrass patch size nor the data presented from our Tampa Bay studies suggest that habitat fragmentation has any consistent impact on fauna over the spatial scales that have been investigated. Likewise, little evidence exists that identifies any taxonomic group to be fragmentation sensitive in that they differentially utilized edge or core areas of seagrass patches. While we did detect a reduction of both infaunal and epibenthic polychaetes at the 1 m edges of seagrass beds relative to interior areas, the reduction was not similar along all edges. Events such as seagrass die off or a high incidence of boat propeller-damage over an extensive area may be required to detect effects of habitat fragmentation on fauna. Given that patch size alone does not appear to adequately account for variation in faunal abundance, we suggest that restoration efforts might best focus upon locating areas with similarity of landscape context or patch characteristics other than patch size. (C) 2001 Elsevier Science Ltd. All rights reserved.