Resilience of benthic macroinvertebrates to extreme floods in a Catskill Mountain river, New York, USA: Implications for water quality monitoring and assessment

Abstract

Changes in the timing, magnitude, frequency, and duration of extreme hydrologic events are becoming apparent and could disrupt species assemblages and stream ecosystems across the Northeastern United States. Between August 28 and 29 of 2011, an average of 31 cm of rain from Tropical Storm Irene fell across Eastern New York State in less than 24 h and caused historic flooding in numerous streams of the Catskill Mountain Region. Peak discharges exceeded the 0.01 annual exceedance probability (> 100 year flood) in many Catskill Mountain streams. Approximately one week later, the remnants of Tropical Storm Lee deposited another 19 cm of rain onto saturated soils and caused additional flooding. Data from annual benthic macroinvertebrate surveys completed at 5 sites in the Upper Esopus Creek, a premier trout stream in the region, during August 2009–2011 (before the floods) were compared to data collected from the same sites in September 2011, November 2011, March 2012 and August 2012 (after the floods). The impact, rate of recovery and the factors which might affect the resilience of benthic macroinvertebrate communities were evaluated. The results of biological water quality assessment metrics immediately after the floods resembled those of highly polluted waters, yet severe floods were the only disturbance. Prior to the floods, standard biological assessment metrics showed that communities were not impacted and water quality was pristine. A large decrease in macroinvertebrate density was evident in the September 2011 surveys following the floods and bioassessment metrics reflected highly degraded water quality conditions. Most community metrics rebounded in 3–7 months (November 2011 and March 2012), and full recovery was evident in 12 months (August 2012) which suggests that macroinvertebrate assemblages are relatively resilient to the effects of extreme floods in these low-order streams. Therefore, macroinvertebrate samples collected from a flood-impacted stream before full recovery occurs might reflect loss of diversity and abundance from the flood disturbance and incorrectly attribute the impact to impaired water quality. The strong short-term impacts and the relatively rapid recovery of macroinvertebrate communities following catastrophic floods have important ramifications for routine bioassessment programs considering changing hydrologic regimes in streams across the Northeast and elsewhere.