8.3.2.2 Continental shelves
Increasingly, trophic cascades are being detected
even in large open coastal and oceanic ecosystems.
Time-series data from offshore fisheries reveal, for
example, that the collapse of Atlantic cod (Gadus
morhua) stocks in recent decades as a result of fish-
ing have been accompanied by increases in both
their benthic crustacean prey (Worm and Myers
2003) and in overlying pelagic communities
(Franket al. 2005). Worm and Myers (2003) ana-
lysed time series from nine cod stocks throughout
the North Atlantic and searched for correlations of
boreal shrimp and cod abundance with one another
and with water temperature. They found that cod
and shrimp abundance showed opposite trends
through time at most sites, consistent with a trophic
cascade from humans through cod to benthic
shrimp. They were also able to reject an alternative
bottom-up hypothesis that these trends stemmed
from changing climate: although water tempera-
ture affected cod, it had no detectable influence in
shrimp abundance.
Clearly such marine trophic cascades are of more
than academic interest. Cod, shrimp, and lobster all
support major commercial fisheries, for example.
Another sobering case involves the decline of
sharks, which have been especially hard hit by fish-
ing (often as by-catch) throughout the world oceans
(Stevenset al. 2000; Baumet al. 2003). Along the
eastern seaboard of North America, time-series
data reveal that precipitous declines of large sharks
since 1970 were accompanied by increases of many
‘mesopredator’ rays and small sharks, most of
which are eaten almost exclusively by large sharks.
These patterns suggest that the mesopredators have
been released from predation by larger enemies
(Myerset al. 2007), as has been suggested for the
increase of small, mammalian predators on land as
well (Crooks and Soule 1999; Johnsonet al. 2007).
Notable among these marine mesopredators is the
cownose ray (Rhinoptera brasiliensis), which feeds
largely on bivalve molluscs, and which increased
by an order of magnitude over the last three dec-
ades. Both survey data and experiments show that
the growing population of cownose rays has in-
flicted heavy mortality on bay scallops in North
Carolina (Fig. 8.4c), resulting in a collapse of the
century-old fishery for this species. Reports from
Chesapeake Bay similarly suggest that cownose
rays are now causing severe damage to seagrass
beds as they forage for the infaunal bivalves living
there, some of which are commercially important.
Similar patterns of observed increases in mesopre-
datory rays and decreases in bivalve populations
have also been observed in the northeast Atlantic
and in Japanese waters (Myerset al. 2007). One
interesting question arising from such dramatic
changes in community structure is whether and
how predators are sustained after their prey are
driven to such low levels. Evidence consistent
with a trophic cascade has also been found in salt
marshes, where areas inaccessible to marine preda-
tors, notably blue crabs, have much higher densities
of rasping snails and marsh grasses achieve accord-
ingly lower biomasses; there is concern that declin-
ing populations of heavily fished blue crabs may
result in deterioration of salt marshes as a result of
this cascade (Silliman and Bertness 2002).8.3.2.3 Pelagic systems
Trophic cascades were first documented in fresh-
water pelagic (Carpenteret al. 1985) and benthic
(Power 1990) systems, and are common in ponds
and lakes (Carpenter and Kitchell 1993; Brett and
Goldman 1996). While they appear less common in
the more functionally diverse marine pelagic com-
munities (Micheliet al. 1999; Shurinet al. 2002),
there is growing evidence for trophic cascades in
both estuaries and the open ocean. In estuaries,
ctenophores (comb jellies) are voracious predators
of other zooplankton, and often reach very high
densities in summer, where in some systems they
can crop 20% of standing crustacean zooplankton
stock per day. A 6 year field study showed general-
ly synchronous but opposite fluctuations of preda-
tory ctenophores, their copepod prey and
phytoplankton, particularly the dominant diatom
species (Deason and Smayda 1982). Integrating
abundances of each group by month or season re-
vealed clear evidence of a trophic cascade as cteno-
phore ‘blooms’ were followed by decimation of
herbivorous zooplankton and subsequent phyto-
plankton blooms (Deason and Smayda 1982). Field
observations suggest that such trophic cascades
also occur in a variety of other marine pelagic sys-
tems, particularly those dominated by gelatinousSTRUCTURE AND FUNCTIONING OF EMERGING MARINE COMMUNITIES 103