Environmental Engineering Reference
In-Depth Information
20% of i sh meal went to pigs, the rest to poultry). Although aquatic ectotherms
convert this feed with higher efi ciencies than land animals—typical feed-to-product
ratios are 1.9-2.4 for salmon and 1.2-1.5 for shrimp (Jackson 2009)—aquafarming
of carnivorous species fed i sh-derived protein and oils will always result in a net
loss of digestible protein, arguably justii ed by the fact that people prefer to consume
salmon and shrimp rather than anchoveta.
The only unaccounted-for claim of aquaculture on aquatic primary production
is phytoplankton, zooplankton, and detritus captured by farmed herbivorous i sh
and invertebrates. The upper limit of this requirement can be estimated by assuming
(contrary to actual practices) that no cyprinids (trophic level 2) receive any supple-
mentary feeding: producing some 15 Mt (1.8 Mt C) of these i shes in 2008 would
have required about 18 Mt C in aquatic phytomass, an insignii cant total compared
to inherent inaccuracies in calculating the feeding requirements of wild i sh. The
previously calculated share of about 6% of the worldwide aquatic NPP can thus
represent the total claim by all harvested marine and freshwater species, caught or
cultured.
This seems to be a comfortably low rate, but it is yet another perfect illustration
of the limits of using the human-appropriated NPP shares as supposedly revealing
measures of human impact. This particular rate ignores the changing primary
productivity base, and it entirely misses three critical factors affecting modern
marine catches that cannot be captured by a simple tracing of total landings and
their conversion to primary production requirements. The i rst one is a long-term
decline in phytoplankton production, the second one is the extent of overi shing,
and the third one (less assuredly) is the shift in average trophic level of harvested
species.
Satellite-derived observations of phytoplankton concentrations have been avail-
able only since 1979, but numerous standardized ocean transparency measurements
(made using a Secchi disk) have been available since 1899, and they can be used to
calculate surface chlorophyll concentration and augmented by direct in situ optical
measurements. Boyce, Lewis, and Worm (2010) used this information to estimate a
century of phytoplankton changes on regional and global scales and found signii -
cant interannual and decadal phytoplankton l uctuations superimposed on long-
term declines in eight out of ten ocean regions, a worrisome trend whose continuation
would have signii cant effects on future catches.
The second concern, the increasing extent of overi shing, has been masked for a
long time by higher landings. Worm et al. (2006) concluded that the rate of i sheries
collapses (when the catches drop below 10% of the recorded maxima) has been
