Practicable and Useful Partitions in Coastal Seas
113
A cutoff level of 0.4 isolated the Coastal biome from the remainder. Separation of
additional groups was achieved at progressed cutoff levels; the principal entities recognized
may be equated (at least informally) to the GFST, NADR, NAST and NECS provinces.
However that may be, it is not very useful to attempt to verify a partition that is
usually presented as a series of rectangular spaces, fixed in location, by reference to the
ungridded observed distribution of biological properties; difficulty of verification by this
method becomes the more acute the longer the period over which the observations are
accumulated: in this case over more than 50 years, during which time we know that
species distribution patterns have changed.
I believe we can go little further at the moment in assuring ourselves that the boundaries
proposed in this book for the open ocean have some objective reality. The data fields
needed for a purely objective ecological partitioning of the ocean neither exist, nor are
likely to exist in the foreseeable future. The proof of the pudding must be in the eating.
If the provinces proposed here, or something like them, prove to be useful for practical
purposes, such as the partitioning of fishery statistics to investigate the biological basis of
fisheries production, then they probably resemble those that might have been objectively
drafted had the needed data been available. If not, they will rapidly join the heap of
discarded biogeographic systems.
Practicable and Useful Partitions in Coastal
Seas
For a partition of the coastal zone, one cannot proceed so methodically as for the open
ocean, and it is here that the greatest difficulty exists in proposing a logical and objective
system. It is here that the fractal nature of spatial diversity is most pronounced. It is
here that one is faced with the greatest number of choices in how to make a partition
and no single solution is evident. I have, therefore, adopted a very flexible approach
to the selection of unit areas within which it seems useful to consider the regional
ecology in an integrated manner. A series of provinces within the Coastal biome has
been established that match the characteristic dimension of the oceanic provinces: there
are 22 of these, bringing the number of provinces close to the total of 50 that was the
original target. Their boundaries are, seaward, the shelf-break or upwelling front present
along many continental slopes. Longshore boundaries of coastal provinces are set to
match the natural features of the coastline: at prominent capes, especially at those where
a major coastal current passes offshore into the open ocean. The suggested boundaries
also match the major partitions between the Trades, Westerlies, and Polar biomes so that
coastal provinces may be associated with each for some purposes. Several of the provinces
proposed in this system correspond at least approximately with one of the well-known
“Large Marine Ecosystems” proposed by Ken Sherman as suitable units within which to
integrate fishery management research. The boundaries of these “ecosystems” are not,
however, entirely based on natural features, and LMEs are not considered further here. For
those who may be interested in making a comparison between these and biogeochemical
provinces, Pauly et al. (2000) provide an analysis of the compatibility of the two systems.
Of course, almost all of the 22 coastal provinces used here are too large to be entirely
satisfactory for some purposes, but it is simple for users of the system to establish their
own smaller partition. For instance, the Northeast Atlantic Continental Shelf province
runs from Cape Finisterre to southern Norway, a region that may be subdivided quite
naturally and easily. The following would represent a sensible second-level partition:
(i) North Sea from the Straits of Dover to the Shetlands, (ii) English Channel west to
Ushant, (iii) Bay of Biscay from Ushant to Cape Finisterre, (iv) northern outer shelf, from
113
A cutoff level of 0.4 isolated the Coastal biome from the remainder. Separation of
additional groups was achieved at progressed cutoff levels; the principal entities recognized
may be equated (at least informally) to the GFST, NADR, NAST and NECS provinces.
However that may be, it is not very useful to attempt to verify a partition that is
usually presented as a series of rectangular spaces, fixed in location, by reference to the
ungridded observed distribution of biological properties; difficulty of verification by this
method becomes the more acute the longer the period over which the observations are
accumulated: in this case over more than 50 years, during which time we know that
species distribution patterns have changed.
I believe we can go little further at the moment in assuring ourselves that the boundaries
proposed in this book for the open ocean have some objective reality. The data fields
needed for a purely objective ecological partitioning of the ocean neither exist, nor are
likely to exist in the foreseeable future. The proof of the pudding must be in the eating.
If the provinces proposed here, or something like them, prove to be useful for practical
purposes, such as the partitioning of fishery statistics to investigate the biological basis of
fisheries production, then they probably resemble those that might have been objectively
drafted had the needed data been available. If not, they will rapidly join the heap of
discarded biogeographic systems.
Practicable and Useful Partitions in Coastal
Seas
For a partition of the coastal zone, one cannot proceed so methodically as for the open
ocean, and it is here that the greatest difficulty exists in proposing a logical and objective
system. It is here that the fractal nature of spatial diversity is most pronounced. It is
here that one is faced with the greatest number of choices in how to make a partition
and no single solution is evident. I have, therefore, adopted a very flexible approach
to the selection of unit areas within which it seems useful to consider the regional
ecology in an integrated manner. A series of provinces within the Coastal biome has
been established that match the characteristic dimension of the oceanic provinces: there
are 22 of these, bringing the number of provinces close to the total of 50 that was the
original target. Their boundaries are, seaward, the shelf-break or upwelling front present
along many continental slopes. Longshore boundaries of coastal provinces are set to
match the natural features of the coastline: at prominent capes, especially at those where
a major coastal current passes offshore into the open ocean. The suggested boundaries
also match the major partitions between the Trades, Westerlies, and Polar biomes so that
coastal provinces may be associated with each for some purposes. Several of the provinces
proposed in this system correspond at least approximately with one of the well-known
“Large Marine Ecosystems” proposed by Ken Sherman as suitable units within which to
integrate fishery management research. The boundaries of these “ecosystems” are not,
however, entirely based on natural features, and LMEs are not considered further here. For
those who may be interested in making a comparison between these and biogeochemical
provinces, Pauly et al. (2000) provide an analysis of the compatibility of the two systems.
Of course, almost all of the 22 coastal provinces used here are too large to be entirely
satisfactory for some purposes, but it is simple for users of the system to establish their
own smaller partition. For instance, the Northeast Atlantic Continental Shelf province
runs from Cape Finisterre to southern Norway, a region that may be subdivided quite
naturally and easily. The following would represent a sensible second-level partition:
(i) North Sea from the Straits of Dover to the Shetlands, (ii) English Channel west to
Ushant, (iii) Bay of Biscay from Ushant to Cape Finisterre, (iv) northern outer shelf, from