Evidence of Streamflow and Sediment Effects on Juvenile Coho and Benthic Macroinvertebrates of Lagunitas Creek and San Geronimo Creek, Marin County, California

Lagunitas Creek and San Geronimo Creek in Marin County, California provide some of the best habitat for endangered coho salmon (Oncorhynchus kisutch) in the southern part of their range, making it a priority for local and federal agencies to collect habitat and biological data throughout the watershed. For this paper, we synthesized numerous years of existing data, including flow, sediment conditions, endangered coho salmon densities, and one year (2001) of macroinvertebrate biological assessment data to investigate biotic and abiotic interactions among physical habitat, juvenile coho, and macroinvertebrates. We found that summer juvenile coho densities in Lagunitas Creek were negatively correlated with annual peak mean daily flow, whereas in San Geronimo Creek, variation in peak mean daily flow did not significantly impact juvenile density. Although macroinvertebrate prey were not limiting factors for juvenile coho in 2001, increased coho density was correlated with significant declines in the percentage of vulnerable macroinvertebrates at sampling locations. In addition, San Geronimo Creek had relatively high densities of juvenile coho, despite local evidence of excess nutrients and fecal coliforms from 2001. Analyzing fourteen years of qualitative sediment accumulation observations from Lagunitas tributaries, we found that 1) regular sediment inputs from tributaries could be impacting local habitat quality and may require source investigation, and 2) the highest sediment delivery occurred during wet years, but localized sediment accumulation may also occur in low flow years due to lag time in sediment delivery. Our April 2009 physical habitat survey at the bioassessment site LAG220, above Irving Bridge on Lagunitas Creek, suggested that overall substrate conditions have improved at that location, compared to 2001 conditions. To further evaluate interactions among flow, sediment, coho, and macroinvertebrates in Lagunitas and San Geronimo Creeks, we recommend using a GIS model to evaluate data at multiple reach scales and over time