ICES Journal of Marine Science, 57: 628–640. 2000
lib
e
p
64
s
om
is
gen
over fishery effects in semi-enclosed seas and affect biodiversity and susceptibility to
fishing. Such impacts on marine fisheries beyond natural levels of outflow are referred
to here as marine catchment basin (MCB) effects. They require coordinated actions
within the catchment area to control them, since fisheries management actions alone
are unlikely to be effective in rectifying ecosystem impacts. Net nutrient inflows may be
positive or negative, depending on intensity and degree of retention by the receiving
marine system and may promote ecosystem change and lead to ecological dominance
by exotic species. Initially positive effects of enrichment may disguise the urgent need
for coordinated environmental and fishery management measures in semi-enclosed
seas. Fisheries impacts are serious, but may be secondary and synergistic, but
potentially catastrophic under hypoxic conditions since eutrophic processes make
demersal ecosystems particularly sensitive to disturbance of bottom habitats. Hence,
fishing with bottom gear may impact bottom fauna and demersal resources within or
above organic sediments affected by eutrophic processes and hypoxia. These effects
show up as changes in the ratio of pelagic to demersal landings, and support
broad-brush comparative studies of areas subject to different levels of environmental
impact.

2000 International Council for the Exploration of the Sea
Key words: anthropogenic effects, ecosystem impacts on fishing, (European) semi-
enclosed seas, marine eutrophication, nutrients.
J. F. Caddy: FAO, Via delle Terme di Caracalla, 00100 Rome, Italy. Present affiliation:
Research Fellow, T. M. Huxley School of Environment, Earth Sciences and Engineering,
Imperial College, London, and CINVESTAV, Merida, Mexico. Postal address: Via
Cervialto 3, Aprilia 04011, Latina, Italy; e-mail: jfcaddy@yahoo.co.uk.
Introduction
The role played by land-based run-off, either directly
into the sea or through river discharges, has been of
considerable interest to marine biologists. Cooper and
Brush (1993) noted that both anoxia and eutrophication
in Chesapeake Bay have been increasing since European
settlement. Odum (1980), quoted in Winter et al. (1996),
is associated with the ‘‘outwelling’’ hypothesis that states
that estuarine systems produce more organic material
and nutrients than are used internally and that excess
production is exported to nearshore seas. The effects
of nutrient run-off on semi-enclosed seas with limited
oceanic flushing may show up sooner than on open
ocean shelves. Even for open sea areas, temporary
(Rabalais et al., 1996), extensive hypoxia associated with
Mississippi River discharges exceeds in importance the
immediate effects of fishing, given that resources are
excluded from large areas. This paper stresses the impor-
tance of eutrophic phenomena, and contrasts their
effects on ecosystems with those due to fishing.
Increasing biological productivity of semi-enclosed
marine ecosystems (Caddy, 1993a) complicates a simple
description of direct or mechanical effects of fishing, and
hence needs to be summarized first. Semi-enclosed seas
such as the Mediterranean and the Baltic and Black Seas
were the cradles of human civilization, and have coex-
isted with moderate levels of anthropogenic influences
for a long time, with impacts of industrial-scale fisheries
only a recent addition. The combined downstreamdoi:10.1006/jmsc.2000.0739, available online at http://www.idea
Marine catchment basin effects v
semi-enclosed seas
J. F. Caddy
Caddy, J. F. 2000. Marine catchment basin effects versus im
semi-enclosed seas. – ICES Journal of Marine Science, 57: 628–
Synchronous anthropogenic effects on marine coastal system
World War II, make it difficult to separate effects of fishing fr
especially those caused by nutrient runoff. Natural enrichment
fisheries, but over the long term anthropogenic nutrient impactshypoxic conditions can cause mass mortalities (e.g.,
Haskin et al., 1983), and in the Gulf of Mexico
1054–3139/00/030628+13 $30.00/0rary.com on
rsus impacts of fisheries on
acts of fisheries on
0.
, particularly since
terrestrial inputs,
vital to productive
erally predominateeffects of growth in populations, industry, and agricul-
ture in catchment basins, although difficult to quantify,

2000 International Council for the Exploration of the Sea

(b) changes in fish production habitat;
(c) change in species/communities due to introductions
629Marine catchment basin effects versus impacts of fisheries on semi-enclosed seasor replacement of oligotrophic species by those
adapted to hypoxia;
(d) associated fishery effects of eutrophication and the
presence of contaminants (e.g., linkage of flatfish
papillomas with eutrophic conditions; Stich et al.,
1976).
Obviously (a), (b), and (c) can be easily confused with
incidental effects of fishing, as one example will illus-
trate: trawling on muddy sediments adjacent to sea-grass
beds suspends fines, hence reduces the euphotic zone,
and might make photosynthesis by deeper grass beds
unviable. A reduction of light penetration from algal
blooms is associated with ‘‘cultural eutrophication’’, i.e.,
a human-generated increase in nutrient runoff. How to
distinguish the two effects? Or should they be regarded
as synergistic stresses to the ecosystem, as suggested by
Rapport et al. (1985)?
Contemporaneous influences on coastal marine
ecosystems
Fisheries landings (FAO, 1995) and fleet size (FAO,
1994a) confirm that both global fishing effort and land-
ings have risen steadily since World War II. The increase
in landings to a recent plateau in 1996 of some 87 million
tonnes of fish and invertebrates is consistent with the
‘‘theory of fishing’’ which sees changes in fisheries pro-
duction as a function of changing fishing intensity,
modified by climatic factors.
The common assumption that the effects of fishing
dominate and that changes in landings simply reflect
changes in fishing effort may be misleading. As noted by
Stromberg (1997), distinguishing between anthropogenic
and natural effects requires ‘‘a significant understanding
of the dynamics of ocean’’ that is not attainable byappear to have been accentuated especially since World
War II.
Kerr and Ryder (1993) noted that ‘‘Few would dis-
agree that a primary effect of eutrophication . . . is
elevated levels of primary production. In judicious
measure, the ensuing effect on fishery production may
not necessarily be judged unwelcome in human terms’’.
Caddy (1993a) summarized evidence for this conclusion
for semi-enclosed seas, and suggested that fishery pro-
duction for formerly oligotrophic seas has increased in
recent decades, with moderate enrichment from land.
Kerr and Ryder (1992) recognized four categories of
eutrophication effects:
(a) modifications to the fish production environment
through reduction of suitable habitats for spawning
and larval survival, and increased vulnerability;considering the effect of only a few key variables.
Difference anthropogenic impacts on marine ecosys-tems, if contemporaneous, would be difficult to dis-
tinguish from the effects of fishing, given the highly
variable background ‘‘noise’’ behind the signal.
Environmental impacts on the freshwater components
of catchment basin ecosystems are easily documented
(e.g., Welcomme, 1995 ), but distinguishing these
changes from those due to fishing in coastal waters still
has to rely on anecdotal information or proxy variables.
Historical landings trends by FAO Statistical Areas
(Grainger and Garcia, 1996; Caddy et al., 1998) show
that, for most regions, peak multispecies landings
occurred in close sequence over the last few decades, as
industrial fisheries have spread out from ‘‘core’’ aeras,
largely in the northern hemisphere. The authors con-
cluded that system productivity to fisheries is in most
areas close to or beyond the top of the multispecies yield
curve. Thus, the effects of environmental changes,
whether anthropogenic or natural, are likely to predomi-
nate as long as the systems remain close to the fishing
intensity providing the theoretical peak multispecies
productivity.
Measurement of environmental driving functions usu-
ally depends on ‘‘proxy’’ variables that for instance
signal when changes in nutrient enrichment may have
occurred. For monitoring indirect effects of fishing,
direct observations are needed, and these have also
received little research attention. Fisheries landings,
being among the few time series available, seem obvious
variables for monitoring an ecosystem. However, they
reflect both types of human impacts; hence a compara-
tive approach between large marine ecosystems seems
essential, before one can separate out short-term ecosys-
tem effects of fishing on semi-enclosed seas (de Leiva
Moreno et al., in press).
During the 20th century, almost all anthropogenic
signals have trended in a similar direction, namely
towards increased stress on natural freshwater and
inshore systems as well as on semi-enclosed marine
ecosystems (Rapport et al., 1985; Caddy, 1993a). Symp-
toms of this stress include simplifications of ecosystem
complexity and dominance by r-selected species, but the
original causes are difficult to distinguish. Figure 1
illustrates the contemporaneous nature of trends in time
series available for those few variables, which actually or
potentially affect aquatic systems on the global scale.
Both world populations and industrial production of
phosphate and nitrate fertilizers (spread largely within
catchment basins) have risen over the same period that
witnessed a dramatic increase in marine landings and
fleet size. Regression analysis alone does not allow us to
distinguish between the relative rates of change in these
trends.
A simple index based on normalized linear regres-
sions, the Linking Ratio (Stamatopoulos and Caddy,
1991), allows one to rank linear regressions of positive-
value dependent variables in terms of their relative trend