Early Life History of Fishes in the San Francisco Estuary and Watershed

<em>Abstract.</em>—We examined assemblage patterns of early life stages of fishes for two major tributaries of the upper San Francisco Estuary: (1) Sacramento River channel, and (2) Yolo Bypass, the river’s seasonal floodplain. Over four hydrologically diverse years (1999–2002), we collected 15 species in Yolo Bypass egg and larval samples, 18 species in Yolo Bypass rotary screw trap samples, and 10 species in Sacramento River egg and larval samples. Fishes captured included federally listed species (delta smelt <em>Hypomesus transpacificus </em>and splittail <em>Pogonichthys macrolepidotus</em>) and several game species (American shad <em>Alosa sapidissima</em>, striped bass <em>Morone saxatilis</em>, crappie <em>Pomoxis </em>spp., and Chinook salmon <em>Oncorhynchus tshawytscha</em>). As in other regions of the estuary, alien fish comprised a large portion of the individuals collected in Yolo Bypass (40–93% for egg and larval net samples; 84–98% for rotary screw trap samples) and Sacramento River (80–99% for egg and larval net samples). Overall ranks of species abundances were significantly correlated for Yolo Bypass and Sacramento River, suggesting that each assemblage was controlled by similar major environmental factors. However, species diversity and richness were higher in Yolo Bypass, likely because of a wider variety of habitat types and greater hydrologic variation in the floodplain. In both landscapes, we found evidence that timing of occurrence of native fishes was earlier than aliens, consistent with their life history and our data on adult migration patterns. We hypothesize that Yolo Bypass favors native fishes because the inundation of seasonal floodplain typically occurs early in the calendar year, providing access to vast areas of spawning and rearing habitat with an enhanced food web. Conclusions from this analysis have implications for the management of aquatic biodiversity of tributaries to the San Francisco Estuary and perhaps to other lowland rivers.

Early Life History of Fishes in the San Francisco Estuary and Watershed

<em>Abstract.</em>—We examined the spatial and temporal variability of native and alien ichthyoplankton in three habitat types (marsh edge, shallow open-water, and river channel) in one reference and three restored marshes in the Sacramento–San Joaquin Delta, California, during 1998 and 1999. More than 6,700 fish embryos and 25,000 larvae represented by 10 families were collected in 240 tows during the 2-year study. Overall, the assemblage was dominated by alien fishes, but natives were more abundant during winter and spring, whereas aliens were more abundant during summer. Overall abundance was highest in marsh edge habitats, suggesting that this habitat provides favorable larval rearing habitats for many fishes. The reference marsh was dominated by alien species making it difficult to assess whether it had attributes that promoted use by native fish. Ichthyoplankton abundance varied comparably at restored sites of similar configuration. The restored site, with minimal tidal exchange and greater lower trophic productivity, supported the highest densities of alien fish. We conclude that restoration projects in this region of the estuary must consider the potential impacts of alien fishes on natives and evaluate strategies designed to improve recruitment success of native fishes. Specifically, we suggest that restored wetlands that offer only winter and spring inundation periods may provide maximum benefits to natives while limiting access by many alien fishes regardless of specific habitat-use requirements.

Early Life History of Fishes in the San Francisco Estuary and Watershed

<em>Abstract.</em>—We used a coupled hydrodynamic-particle tracking model to assess the relative importance of factors affecting transport and entrainment risk for generalized fish larvae in Suisun Marsh in the San Francisco Estuary, California. Factors examined included location of particle release, time of particle release, vertical migration behavior, and combinations thereof. Model sensitivity was evaluated by observing particle entrainment into a water diversion in Suisun Marsh under various scenarios and compared to randomized input cases. Scenarios combining two or more factors indicated nonlinear interactions, including evidence of tidal pumping on neap tides near our test diversion. Our study suggests that accurate modeling of larval transport depends on accurate determination of fish locations, tidal and residual time scale circulation simulation, and fish behavior during early life stages. We also examined appropriate approaches to modeling these factors.

Early Life History of Fishes in the San Francisco Estuary and Watershed

<em>Abstract.</em>—We investigated factors affecting growth of larval striped bass <em>Morone saxatilis </em>in the San Francisco Estuary from 1984 to 1993. We estimated ages and growth rates of larval striped bass from daily otolith increments. Mean annual growth rates of 6–14 mm standard length striped bass varied from 0.13 to 0.27mm/d, the lowest rate occurring in 1989 and the highest in 1992. The 1989 growth rate was significantly lower than all other years, and growth rates for 1992 and 1993 were significantly higher than all other years, but did not differ from one another. Differences in annual growth rates apparently were due mainly to differences in mean annual prey densities because growth rate increased as prey density increased. Compared to both laboratory measured growth rates and growth rates of field-caught Chesapeake Bay larvae, growth rates from the San Francisco Estuary appeared to be high for the food available, indicating that larvae can grow at relatively high rates even at low prey densities. Correlation analyses did not support density-dependent control of growth rates. Growth rate was not significantly related to mean annual conductivity, water temperature, mortality rates, or the juvenile abundance index, but was significantly and positively correlated with densities of 1-mm length-groups of 9–14-mm striped bass.

Early Life History of Fishes in the San Francisco Estuary and Watershed

<em>Abstract.</em>—In July 2000 and 2001, we sampled adjacent screened and unscreened agricultural irrigation diversions in the Sacramento River, California to (1) evaluate the effectiveness of a custom fish screen for excluding four open-water fishes: native delta smelt <em>Hypomesus transpacificus </em>and alien threadfin shad <em>Dorosoma petenense</em>, inland silverside <em>Menidia beryllina</em>, and striped bass <em>Morone saxatilis</em>; and (2) examine factors affecting entrainment of each species. We also compiled trawl and beach seine data from contemporaneous monitoring programs to make inferences about microhabitat use by these fishes and its implications for entrainment vulnerability. The fish screen reduced entrainment of each species by 99% or more and excluded many fish less than 25 mm, the approximate minimum length it was designed to exclude. Tidal and diel influences on entrainment through the unscreened diversion were observed, but diel cycles appeared to be more important, as most entrainment occurred at night or during crepuscular periods. Except for delta smelt, our results suggested that open-water fishes may undergo ontogenetic changes in vulnerability to unscreened irrigation diversions. Fishes entrained during daylight (threadfin shad and striped bass) averaged only 15–16 mm in length. At night, average lengths of entrained threadfin shad and inland silverside were 22–25 mm, even though larvae continued to be entrained. Similarly, a diel influence on striped bass entrainment was observed only in 2000, when individuals larger than 20 mm were consistently collected. No striped bass were collected at sizes greater than 35 mm, even though larger individuals occupied the study area. We found no evidence of size-related changes in delta smelt vulnerability to entrainment, but the monitoring data indicated that delta smelt were abundant offshore, whereas the other three species were most abundant nearshore. We think that low and inconsistent entrainment of delta smelt reflected (1) predominantly offshore habitat use by delta smelt, and (2) the relatively small hydrodynamic influence of the diversion.

Early Life History of Fishes in the San Francisco Estuary and Watershed

<em>Abstract.</em>—We compared two approaches to back-calculation with otolith microstructure to develop a method for accurately estimating growth rates of juvenile fall-run Chinook salmon <em>Oncorhynchus tshawytscha </em>in California’s Central Valley. Total otolith width was a strong determinant of fork length (FL) in linear regressions used to determine the <em>y</em>-intercept in the fish size–otolith size relationship in two study groups of Chinook salmon. The Fraser-Lee back-calculation model estimated FL at first feeding in both study groups that did not differ significantly from lengths of first-feeding Chinook salmon in a reference group. In comparison, the biological-intercept method produced back-calculated lengths that were significantly greater in one study group than lengths of first-feeding Chinook salmon in the reference group. Chinook salmon emergence dates, estimated from counts of daily growth increments beyond the first-feeding check, corresponded with observed emergence periods in the river and hatchery populations from which the study groups were sampled. Size-at-age relationships were well described by a power function in both study groups, where mean FL over time approached an apparent asymptote at approximately 80 mm after 90 d postemergence. Growth rate estimates, using back-calculated size from the Fraser-Lee model, averaged 0.50 mm/d in one study group and 0.43 mm/d in the other study group. These estimates fell within the range of previous growth rate estimates for juvenile Chinook salmon in Central Valley riverine, floodplain, and delta environments and were about 2.5 times higher on average than an estimate for the San Francisco Estuary and about 2.3 times lower on average than estimates from the Strait of Georgia. We discuss the utility of otolith microstructure to not only estimate growth rates, but also to reconstruct emergence-date distributions in cohorts of emigrating juvenile Chinook salmon for stock identification purposes.

Early Life History of Fishes in the San Francisco Estuary and Watershed

<em>Abstract.</em>—San Francisco Bay provides spawning and rearing habitat for California’s largest population of Pacific herring <em>Clupea pallasi</em>. This population provides a food source for other species and supports a valuable fishery for Pacific herring roe. Since the inception of the roe fishery in 1973, the California Department of Fish and Game has conducted annual surveys of spawning in San Francisco Bay as part of an ongoing assessment of population status and management of the fishery. The purpose of this paper is to document (1) regions of San Francisco Bay used by Pacific herring as spawning grounds over time, and (2) time periods in which spawning took place. Spawn data were analyzed by geographic region in the bay and by month for the period 1973–2000. During this period, we documented 269 spawning events from Point San Pablo south to Redwood City. Estimates of spawning adult biomass (fish that were not harvested by the fishery) ranged from 80,813 metric tons in 1981–1982 to 3,199 metric tons in 1997– 1998 (mean = 34,688 ± 19,325 SD). January was the peak spawning month, followed by December and February; small variations in this pattern occurred during some years. Overall, the majority of spawning took place in the north-central bay region (Point Bonita to Richmond-San Rafael Bridge, Angel Island, Point San Pablo, Berkeley flats; 55%) and the San Francisco region (Golden Gate Bridge to Candlestick Point; 34%), although it alternated between these two regions over time. In some years, considerable spawning took place in the Oakland–Alameda region (San Francisco–Oakland Bay Bridge to Bay Farm Island). The largest spawns and peak periods of spawning may not contribute most toward the next generation of Pacific herring, due to differential mortality within the season. For this reason, all regions documented in this study are important spawning grounds for Pacific herring from November through March each year. A number of recent studies have furthered our understanding of Pacific herring early life history and the forces that drive year-class formation in San Francisco Bay. However, studies are especially needed that will improve our ability to adequately address the potential impacts of human activities on Pacific herring in this highly urbanized estuary.

Early Life History of Fishes in the San Francisco Estuary and Watershed

<em>Abstract.</em>—The goal of this study was to determine if the vertical distribution of larval delta smelt <em>Hypomesus transpacificus </em>and striped bass <em>Morone saxatilis </em>was affected by tidal stage and diel period. Delta smelt and striped bass have similar early life histories in that their larvae drift downstream from freshwater spawning habitats to brackish water rearing habitats. Little is known on whether the larvae undergo a vertical migration as they move downstream. Conical plankton nets were used to collect larval fishes from the surface, middle, and bottom sections of the main channels of the Sacramento and San Joaquin rivers. Samples were collected over six paired day–night cruises. Approximately 45,000 fish were collected; 82% were striped bass and 2% were delta smelt. More delta smelt were caught at night than during the day, and more at middle depth than surface or bottom. Very few striped bass were found at surface during day; most were caught at middle depth. Both species were fairly dispersed throughout the water column at night. Although striped bass larvae appear to undergo a small surface to middepth migration, this study indicates that larval delta smelt and striped bass do not undergo a mass diel or tidal vertical migration as they drift down the Sacramento and San Joaquin rivers.

Early Life History of Fishes in the San Francisco Estuary and Watershed

<em>Abstract.</em>—Runoff from dormant spray applications to orchards can contain the insecticides diazinon and esfenvalerate, which may affect the health of the threatened splittail <em>Pogonichthys macrolepidotus</em>. To examine the potential effects of these two insecticides, splittail larvae were exposed to diazinon, esfenvalerate, and a mixture of the two insecticides in static renewal 96-h acute toxicity tests. Surviving fish were transferred to clean water for an additional 14 d (18 d total). Mortality, morphological anomalies, histopathology, and growth determinations were made to assess lethal and sublethal effects. The combination of diazinon and esfenvalerate produced less than additive (independent) toxicity. Diazinon (singly and in combination with esfenvalerate) produced latent toxicity after the 96-h exposure, as demonstrated by reduced growth and increased spinal deformities. Metabolic dysfunction in the liver and inflammation of the pancreas were likely related to slower growth of diazinon-exposed fish. These symptoms were almost absent from esfenvalerate-exposed fish. The use of 14-d EC50 values and the measurement of biomarkers may more accurately describe the effects of 96-h shortterm exposure to these insecticides than traditional 96-h LC50 values.