9697

• UCREFRP Catalog Number: 9697
• Author: Bestgen, K.R., D.W. Beyers, J.A. Rice and G.B. Haines.
• Title: Factors affecting recruitment of young Colorado pikeminnow
• USFW Year: 2006.
• USFW - Doc Type: synthesis of predation experiments, field studies, and individual-based modeling.
• Copyright Material: YES

1738

Based on simulation results, this scenario suggested
that when large red shiners are present in early to
midsummer, wild Colorado pikeminnow must grow
quickly to survive to autumn. This hypothesis was
supported because autumn juveniles averaged 12-73%
faster growth than summer juveniles. A potential
explanation for this pattern is size-dependent predation
mortality from a gape-limited predator. Under this
hypothesis, slow-growing larvae should be susceptible
to predators for a longer period of time than fast-
growing fish, so only the fastest-growing fish in a
cohort survive (Houde 1987; Miller et al. 1988).
Colorado pikeminnow hatched later in summer sur-
vived to autumn, even though they were slower
growing, probably because large red shiners that
caused differential mortality were not present.
Even though relatively small and slow-growing
Colorado pikeminnow can survive to autumn, small-
bodied juveniles may have relatively low overwinter
survival. In a laboratory study, Thompson et al. (1991)
found that both fed and starved Colorado pikeminnow
that were relatively small (mean, 30 or 36 mm TL) had
lower survivallhan larger fish (44 mm) over simulated
winter conditions. In a Green River field study, Haines
et al. (1998) found that small Colorado pikeminnow
(mode, 28 mm TL) had only 6% overwinter survival
durillg a high-flow winter period, whereas in a different
year, when Colorado pikeminnow were larger (>38
mm), they had higher overwinter survival (56-65%),
albeit winter flows were lower. Poor overwinter
survival of small Colorado pikeminnow in the Green
River places added importance on the few larger fish
from early cohorts that survive, because those fish may
have higher overwinter survival.
Intra-armual variation in survival of cohorts of larvae
with different hatching dates has been observed for
other freshwater fishes (Crecco and Savoy 1985; Rice
et al. 1987; Limburg 1996; Cargnelli and Gross 1996;
Mion et al. 1998; Garvey et al. 2002). Most of those
studies invoked low temperatures or high flows as
mechanisms for reduced survival of portions of the
year-class. Poor survival of early spawned bloater
Coregonus hoyi was thought a function of high
predation rates on slow-growing larvae when water
was relatively cold (Rice et al. 1987; Luecke et al.
1990), which similar to our study, suggested Ihat
bloater recruitment may be structured by interacting
physical and biotic factors.

IBM Simulations

The ffiM simulations confirmed patterns observed in
field studies and provided support that predation
interacting with environmental variables was a main
mechanism structuring recruitment patterns. Simula-

BESTGEN ET AL.

tions were also useful to explore "what if' scenarios
such as the potential effect of reducing predator
abundance or exploring effects of different thermal
regimes on Colorado pikeminnow growth and recruit-
ment.
Additional data are needed to make ffiM simulations
more representative of the processes in the Green
River. For example, understanding seasonal size and
density shifts of red shiners and other potentially
predaceous fishes would better reflect Ihe predator
assemblage in backwaters. Information on the poten-
tially important effects of discharge level and flow
fluctuations on food availability and fish growth would
also be useful, particularly if ffiM simulations were
linked to bioenergetics modeling (Burke and Rice
2002). This is important because discharge fluctua-
tions, such as those produced by releases from
hydroelectric dams or natural flow increases, reduce
the abundance of nearshore chironomid assemblages
(Blinn et al. 1995), which are important in the diets of
most young fishes in the Green River (Muth and
Snyder 1995).

Alternative Recruitment Hypotheses

Starvation is another possible mechanism affecting
recruitment of early life stages of Colorado pike-
minnow in Green River backwaters, but we found no
support for that hypothesis. Starvation is primarily a
function of body size for fish larvae (Miller et al.
1988), and Colorado pikeminnow larvae are only of
moderate size at hatching (5.5 mm TL). However, 87%
of laboratory-reared larvae starved for 15 d after they
could eat (21 d posthatch) survived when offered food
(Bestgen 1996). Further, survival predictions from
Miller et al. (1988) for larvae that hatch at the same size
as Colorado pikeminnow suggested that pikeminnow
have higher than average resistance to starvation. Thus,
high starvation resistance makes it unlikely that direct
starvation mortality is an important mode of population
regulation.
We did find evidence that intra-annual flow
fluctuations affected the recruitment of cohorts of
Colorado pikeminnow larvae, perhaps through interac-
tions with predation. This was apparent in the lower
Green River in 1991 and 1992, when increased
discharge and very high turbidity from thunderstorm
runoff were associated with reduced growth and
recruitment of juvenile Colorado pikeminnow. At
discharge levels above 80 m3fs, which was in the
range observed in 1992, much of the available
backwater habitat was eliminated in that mostly
canyon-constrained reach (Haines and Tyus 1990;
K.R.B., personal observation) and habitat loss may
have reduced foraging efficiency and Colorado pike-