278 
Fishery Bulletin 117(4) 
sturgeon (Acipenser oxyrinchus), the American shad (Alosa 
sapidissima ), river herring, a collective term for the blue- 
back herring (A. aestivalis) and the alewife (A. pseudoha- 
rengus), and the American eel ( Anguilla rostrata), have 
declined in Chesapeake Bay (Haro et al., 2000; Niklitshek 
and Secor, 2005; Limburg and Waldmen, 2009). Chesa¬ 
peake Bay is far from pristine, and anthropogenic activ¬ 
ities have resulted in major ecological changes, including 
the widespread loss of aquatic macrophytes, increased 
turbidity, and frequent hypoxic and anoxic events (Kemp 
et ah, 2005). Scientists and fisheries managers are now 
concerned that predation by invasive blue catfish may 
lead to further declines of depleted native fish species. 
Although blue crab (Callitiectes sapidus ) are not rare, 
there is concern about predation pressure on this species 
because it supports lucrative commercial fisheries in Vir¬ 
ginia, Maryland, and Delaware (Paolisso, 2002). Other 
studies have shown that blue catfish are consuming the 
aforementioned fish and crab species, with the exception 
of Atlantic sturgeon (Schmitt et ah, 2017, 2019). Previous 
research has revealed that blue catfish in Chesapeake 
Bay have remarkably broad diets that include vegeta¬ 
tion, numerous fish species, mollusks, crustaceans, birds, 
terrestrial mammals, reptiles, amphibians, and various 
invertebrates, yet the factors that drive dietary varia¬ 
tion have not yet been identified (Schmitt et ah, 2019). 
Furthermore, the predation dynamics of blue catfish on 
depleted species like the American shad and American eel 
have not been described. 
Although a substantial body of literature is dedicated to 
factors that influence the establishment of invasive species 
(Catford et al., 2009; Blackburn et ah, 2011), fewer works 
have focused on the impact phase of invasion (Fei et ah, 
2016), and most studies produce speculative results 
(Simberloff et al., 2013). This is especially the case for inva¬ 
sive fish species, for which more observational and experi¬ 
mental studies are urgently needed (Garcia-Berthou, 2007; 
Layman and Allgeier, 2012; Brandner et al., 2013). Diet stud¬ 
ies are not necessarily direct measures of impact, but they 
are useful for determining which organisms are most likely 
to be affected by an introduced predator (Caut et ah, 2008; 
Layman and Allgeier, 2012). Diet studies can also be used to 
determine when and where depleted species are most vul¬ 
nerable to predation by an invader, information that can be 
used to guide management efforts (Schmitt et ah, 2017). 
This study fulfilled 2 main objectives. First, we used 
multivariate modeling to determine the significance and 
relative influence of several factors that were suspected 
to influence the food habits of blue catfish. We hypothe¬ 
sized that diets would vary spatiotemporally, as these 
tidal systems support diverse assemblages that change 
in space and time due to salinity preference and seasonal 
migration patterns (Wagner and Austin, 1999; Jung and 
Houde, 2003). We also expected diets to vary with catfish 
size, as previous work has demonstrated that blue catfish 
undergo ontogenetic diet shifts from omnivory to piscivory 
at larger sizes (500-900 mm total length [TL], depending 
on the river; Schmitt et ah, 2019). Second, we incorporated 
any significant factors from the first objective into models 
of predation on American shad, river herring, American 
eel, and blue crab. These models were then used to eluci¬ 
date the circumstances that result in greater predation on 
these depleted or economically valuable native taxa. This 
information can be used to direct management efforts, 
which we discuss later. 
Materials and methods 
Study area 
Chesapeake Bay is the largest estuary in the continental 
United States and has a long history of commercial and 
recreational exploitation (McHugh and Bailey, 1957; Rich¬ 
ards and Rago, 1999). Although blue catfish are now found 
in all major tributaries of Chesapeake Bay (Schloesser 
et ah, 2011), many populations are still in the early stages 
of establishment and support low densities of fish (Agu¬ 
ilar et ah, 2017). We therefore focused our efforts on the 
James, Pamunkey, Mattaponi, and Rappahannock Rivers 
in eastern Virginia (Fig. 1). These rivers were stocked 
with hundreds of thousands of blue catfish between 1973 
and 1985 and now contain well-established populations 
that include mature individuals (Greenlee and Lim, 2011; 
Bunch et ah, 2018). 
Field methods 
Each river was divided into 3 strata according to average 
fall surface salinities during 1985-2016 by using data from 
the Chesapeake Bay Program’s Water Quality Database 
(available from website). The 3 strata included tidal fresh¬ 
water stretches (Practical Salinity [S P ]: 0.0-0.5), oligohaline 
stretches (S P : 0.6-5.0), and mesohaline stretches (S P : 5.0- 
18.0). We stratified each river by autumn salinities because 
density stratification is less problematic during fall (Shiah 
and Ducklow, 1994). Each stratum was divided into 2-km 
sections, which were numbered, and then a random num¬ 
ber generator was used to select each sampling location. 
During April-October, a minimum of 2 randomly selected 
sections were sampled monthly within each stratum for all 
4 rivers, with both nearshore and main-channel sampling 
occurring when possible. Most blue catfish were sampled 
by using low-frequency, pulsed direct current electrofishing 
(5-25 Hz, 100-400 V) because it captures blue catfish of all 
sizes (Bodine and Shoup, 2010) and is extremely effective 
in Virginia’s tidal rivers (Greenlee and Lim, 2011; Schmitt 
and Orth, 2015). In higher salinities, pulsed alternating cur¬ 
rent electrofishing was occasionally used, and anode design, 
voltage, and pulse frequency were adjusted on the basis of 
water conductivity and other environmental conditions. 
Upon capture, fish were immediately placed in a 568-L 
aerated livewell, and stomach contents were extracted 
within 30 min of capture to prevent regurgitation (Garvey 
and Chipps, 2012). Stomach contents were extracted 
either by excising the stomachs or with pulsed gastric 
lavage, which is highly effective for extracting stomach 
contents from blue catfish (Waters et al., 2004). Date, time, 
