abundance should be controlled alternately by bot-
tom-up and top-down processes as one descends
through successive levels of the food chain. The
chain of indirect effects emanating from top preda-
tors has since been called a trophic cascade (Paine
1980; Carpenteret al. 1985). Despite the obvious
greater complexity of real communities and the
several factors that would appear to work against
trophic cascades (Strong 1992; Polis 1999), many
controlled experiments in marine and other sys-
tems have supported the general HSS hypothesis,
showing that loss of predators indeed often releases
prey from top-down control, leading in turn to
strong reductions in the prey’s resources (Menge
1995; Shurinet al. 2002). Similar trophic cascades
have been documented experimentally in a wide
variety of ecosystems (Paceet al. 1999; Boreret al.
2005). Moreover, controlled experiments show that
changes in abundance of apex predators often have
cascading impacts throughout ecosystems (Pace
et al. 1999; Shurinet al. 2002), and that top-down
influences penetrate farther, on average, through
food chains than do bottom-up influences of nutri-
ent loading (Boreret al. 2006). Given the character-
istic shortening of marine food chains under human
influence, these generalizations imply that a consis-
tent consequence of human impacts will involve
cascading indirect effects of reduced predation
pressure.8.3.2 Evidence for trophic cascades in open marine systems
Marine ecosystems are open, with propagules and
apex predators moving over large distances, and
are subject to climate forcing and other influences
that could attenuate trophic cascades (Jennings
and Kaiser 1998). Perhaps surprisingly, the expec-
tations from simple theory of alternating bottom-
up and top-down control at adjacent trophic levels
are nevertheless supported by accumulating evi-
dence that fishing can drive trophic cascades. This
evidence includes time-series data from kelp beds
(Esteset al. 1998; Davenport and Anderson 2007),
coral reefs (Dulvyet al. 2004), open ocean plankton
(Shiomotoet al. 1997), the demersal communities
(Worm and Myers 2003) and pelagic communities
(Franket al. 2005, 2007) of continental shelves, and(^10080100)
ab c
60
40
20
0
(^6050)
(^4030)
(^2010)
0
10
(^86)
(^42)
(^01972198519891993)
Year
1997
Snail mortality(% loss/day
1 )
Snail density(adults/m
2 )
Spartina biomass
(g/m
3 )
Kelp density(No. 0.25 m
-2)
Grazing(%/24 hr
1 )
Urchin biomass
(g 0.25/m
2 )
Otters
(% max. count)
75
50
0.03 0.003
0.002
0.001
0.0
0.02
0.01
1970 1990
Great sharks
Elasmobranchmeasopredators
Scallops
1970 1990
1970 1990
1970 1990
1970 1990
0.0
0.6
0.4 0.02
0.01
0.0
0.2
0.0
200
150
100
50
25
0
800
600
400
200
0
4000
3000
2000
1000
(^0) Tall
zone
Short
zone
400
300
200
100
0
Figure 8.4Evidence for trophic cascades in disparate coastal marine ecosystems. (a) Killer whales to sea otters to sea
urchins to kelps in northeast Pacific kelp beds. Reproduced with permission from Esteset al. (1998). (b) Shell-crushing
crabs and terrapins to snails to cordgrass in west Atlantic salt marshes. Reproduced with permission from Silliman
and Bertness (2002). (c) Great sharks to cownose rays to bay scallops in west Atlantic estuaries. Reproduced with
permission from Myerset al. (2007).
STRUCTURE AND FUNCTIONING OF EMERGING MARINE COMMUNITIES 101