Feeding habits and novel prey of larval fishes in the northern San Francisco Estuary

Abstract

Food limitation can dampen the survival and growth of fish species during early development. To investigate prey diversity important to the planktivorous larval longfin smelt (Spirinchus thaleichthys) and Pacific herring (Clupea pallasii) from the San Francisco Estuary, we used DNA metabarcoding analysis of the cytochrome oxidase I gene on the guts of these fishes and on environmental zooplankton samples, in comparison with results from abundance analyses of zooplankton samples. Overall, both fish consumed the prey that were commonly available and relatively abundant: Diets were dominated by arthropods but also included soft‐bodied and difficult to identify organisms. Prey taxa substantially overlapped between the two species (Schoener's index = 0.66), and alpha diversity analysis suggested high variability in the content of guts from different individuals. Abundant prey taxa in both fish species included the copepods Eurytemora carolleeae, Acanthocyclops americanus, and A. robustus; Acanthocyclops spp. are difficult to distinguish morphologically. Differential abundance analysis suggested that both species consumed the most abundant zooplankton in lower proportions than their availability in the environment. A few uncommon prey observed in the diets may hint at feeding strategies employed by the larvae, such as herring DNA in the longfin smelt diets, suggesting feeding on eggs near substrates or schooling behavior. Herring consumed the small (<0.5 mm) copepod Limnoithona tetraspina much more frequently (30%) than did smelt (2%), possibly indicating differences in foraging behavior or sensory abilities. Among the unexpected prey found in the diets was the cnidarian Hydra oligactis, the polychaete Dasybranchus sp., and the copepod Mesocyclops pehpeiensis, a species not previously described in the San Francisco Estuary. Our findings illustrate the power of molecular methods in revealing feeding patterns and novel plankton diversity that can be undetectable with complementary morphological analysis methods.

Long-term trends in seasonality and abundance of three key zooplankters in the upper San Francisco Estuary

“…Based on previous research (Lott 1998;Hobbs et al 2006;Hammock et al 2017;Jungbluth et al 2021), it is likely that the presence of our study species in fish diets is higher than our data indicates. This could be due to damage during digestion hindering identification of prey to the taxonomic level of our study taxa, or that most fish samples were limited to specific seasons and may not have been collected at a time that our zooplankton species were present in the environment.…”

“…Acanthocyclops robustus, Acanthocyclops americanus, and A. vernalis (Alekseev et al 2002;Dodson et al 2003;Miracle et al 2013;Alekseev 2021). Due to possible morphological misidentifications, it is not known whether all three species in the A. vernalis complex are native to the estuary and were present in the past; however, Jungbluth et al (2021) did find molecular evidence of all these species in larval Longfin Smelt diets recently collected in this region.…”

“…Acanthocyclops spp. is an important component of fish diets in this region, particularly Longfin Smelt (CDFW unpublished data; Hobbs et al 2006;Burris et al in review) and larval Pacific Herring (Clupea pallasii) (Jungbluth et al 2021), with Acanthocyclops spp. being the most common cyclopoid consumed by larval Longfin Smelt (Jungbluth et al 2021).…”

“…This is consistent with other studies that show the importance of Acanthocyclops spp. in Longfin Smelt (Hobbs et al 2006;Jungbluth et al 2021), Delta Smelt (Lott 1998;Hammock et al 2017), and larval Pacific Herring (Jungbluth et al 2021) diets. Acanthocyclops spp.…”

“…Bosmina spp. is consumed by fish in the estuary and elsewhere, including juvenile salmonids and Delta Smelt (Roegner et al 2015;Goertler et al 2018;Slater et al 2019), but it rarely occurs in diets of larval Longfin Smelt (Jungbluth et al 2021;Burris et al in review), likely due to a mismatch in the salinity preferences of Bosmina spp. and Longfin Smelt.…”