Explaining Patterns of Pelagic Fish Abundance in the Sacramento-San Joaquin Delta

Abstract

Investigating the effects of environmental, biological, and anthropogenic covariates on fish populations can aid interpretation of abundance and distribution patterns, contribute to understanding ecosystem functioning, and assist with management. Studies have documented declines in survey catch per unit effort (CPUE) of several fishes in the Sacramento-San Joaquin Delta, a highly altered estuary on the US west coast. This paper extends previous research by applying statistical models to 45 years (1967–2012) of trawl survey data to quantify the effects of covariates measured at different temporal scales on the CPUE of four species (delta smelt, Hypomesus transpacificus; longfin smelt, Spirinchus thaleichthys; age-0 striped bass, Morone saxatilis; and threadfin shad, Dorosoma petenense). Model comparisons showed that along with year, the covariates month, region, and Secchi depth measured synoptically with sampling were all statistically important, particularly in explaining patterns in zero observations. Secchi depth and predicted CPUE were inversely related for all species indicating that water clarity mediates CPUE. Model comparisons when the year covariate was replaced with annualized biotic and abiotic covariates indicated total suspended solids (TSS) best explained CPUE trends for all species, which extends the importance of water clarity on CPUE to an annual timescale. Comparatively, there was no empirical support for any other annualized covariates, which included metrics of prey abundance, other water quality parameters, and water flow. Top-down and bottom-up forcing remain important issues for understanding delta ecosystem functioning; however, the results of this study raise new questions about the effects of changing survey catchability in explaining patterns in pelagic fish CPUE.