Long-Term Stability of the Faunal Community of a Subtropical Estuary: Evaluating Disturbances in the Context of Interannual Variability

Estuarine and coastal marine ecosystems can have significant intra- and interannual variability in faunal community structure, complicating management response to disturbances as disturbance effects must be detectable within normal variability and on timescales relevant to management entities. We examined a long-term, multi-gear dataset on estuarine faunal (fish and select invertebrate) communities to determine if community structure changes due to disturbances could be detected on management-relevant timescales (e.g., years), given that these communities have known seasonal and interannual variability. Results from multivariate, community-based analyses, and several univariate diversity indices for fauna of Tampa Bay, Florida, USA suggested general community stability over the long term, with the exceptions of 2005-2006 (prolonged red tide event) and 2010 (extreme cold event in winter). Community structure was notably outside the typical variability for both time periods. In 2005, species richness decreased by half for one gear type and was the lowest on record for the other two gear types. In 2010, when the relative abundance of a top predator decreased by 50% from cold-related mortality, Simpson's diversity index was the lowest on record for one gear type. The red tide and extreme-cold events differed in duration and the number of taxa directly impacted (multispecies vs. single-species, respectively) but both multivariate and univariate analyses indicated significant deviation in the faunal community structure during these years. Within 1-3 years after these major events, however, the community seemed to have returned to a structure like that of previous years, suggesting long-term stability and resilience. Our results confirm that multivariate and univariate analyses can detect major disturbances to estuarine faunal communities, which gives management entities options for which analysis approach is appropriate for their needs. Since the Tampa Bay faunal communities recovered within 3 years, we suggest that where long-term monitoring exists, active restoration might be deferred while monitoring for signs of recovery following the removal of a perturbation. This is a fundamental part of sound adaptive management processes to promote sustainable ecosystems.