The distribution of diversity: challenges and applications
83
4.6 NATURAL UNITS IN
THE MARINE REALM
As on land, both compositionalist and functionalist
approaches are used to map marine environments.
However, there are important factors of the marine
realm that differ from the terrestrial, affecting the distribution of organisms and habitats, limiting our
understanding (the Linnean and Wallacean shortfalls
are even more pronounced in the sea than on land) and
determining the tools that can be used to create biogeographical classifi cations. Some of these physical, biological and socio - political factors are highlighted in
Table 4.6 , as they have important implications for
marine biogeography and the development of conservation initiatives (see Chapter 5 , Section 5.4 ).
The marine realm is more heterogeneous, fl uid and
obviously interconnected than the terrestrial, and the
high density of water enables a fully pelagic lifestyle
that is absent on land. Indeed, over 90 per cent of the
marine realm ’ s living space is pelagic and pelagic
systems support the majority of marine biomass.
Pelagic ecosystems represent a three - dimensional
continuum from the surface, epipelagic layer through
to the deepest trenches or bathypelagic zones. Water
chemistry, salinity, depth/pressure, currents and
primary productivity gradients create different regions
within these systems. Demersal or benthic ecosystems
form a two - dimensional layer from the shoreline to
deep sea, where bathymetry (depth) corresponds to
elevation within the terrestrial realm and in which different physiographical features such as seamounts,
submarine canyons, trenches, hydrothermal vents,
volcanoes and hypersaline lakes are embedded.
Overlain on the physical underwater landscape are different biogenic habitats, e.g. sea grass, coral reefs,
oyster beds, and maerl (maerl beds are formed by
certain species of calcifi ed red seaweed).
Efforts to undertake biogeographical classifi cation of
marine systems have a lengthy history, dating back to
work on coral reefs by James D. Dana published in
1848 and on molluscs by S.P. Woodward published in
1856 (both cited in Hedgepeth, 1957 ). Perhaps the
most widely accepted functional classifi cation divides
the sea into neritic (continental shelf) and oceanic
(beyond the 200 m isobath) zones (Hedgepeth, 1957 ).
Within the neritic zone are four primary biomes: estuarine, coastal marine, demersal shelf and pelagic shelf.
Within the oceanic zone are four primary biomes:
continental shelf, abyssal, epipelagic and meso/
bathypelagic.
Regions of open ocean were fi rst demarcated by
direction, velocity and persistence of currents (Dietrich,
1963 ), which provided the basis for the development
of a system of domains and divisions of the sea (Bailey,
1998 ). Temperature was also highlighted early on as
an important controlling factor of species ’ distributions and, hence, as a basis for spatial divisions at the
highest level of biogeographical classifi cation (Ekman,
1953 ). Faunal records and per cent endemism have
been used to defi ne provinces (Ekman, 1953 ; Briggs,
1974 ) and a combination of geomorphology and biotic
associations have been used to defi ne coastal zones
(Ray, 1975 ) (Table 4.7 ).
These older qualitative efforts can be distinguished
from more recent applications of multivariate clustering algorithms as applied to faunal composition (Ray
& Hayden, 1993 ), interpretations of remotely sensed
oceanographical data (Longhurst, 1998 ), biophysical
data (Harris & Whiteway, 2009 ), hierarchical geophysical approaches (Roff et al ., 2003 ) and systematically collected community data. Such analyses can be
repeated and tested for sensitivity to assumptions
(Shears et al ., 2008 ).
The urgent need for marine biogeographical classifi -
cations in order to proceed with immediate conservation goals has stimulated the development of numerous
regional marine systems, for example in Australia,
using both a compositionalist and controlling factors
approach (Environment Australia, 1998 ), and in
Canada, primarily using controlling factors (Roff et al .,
2003 ). A functional approach has also been taken in
developing the non - hierarchical set of Large Marine
Ecosystems (Sherman, 1993 ).
Recently, Spalding et al . (2007) reviewed the global
and regional marine classifi cations in order to develop
a consensus map (the Marine Ecoregions of the World)
of nested realms, provinces and ecoregions, at a scale
appropriate for global/regional conservation planning
in continental shelf regions (Spalding et al ., 2007 ).
Since different classifi cations resulted in different
boundaries being drawn, comparisons were made and
expert regional advice sought in order to determine the
best ecoregional defi nitions (Figure 4.15 ).
A classifi cation of the high seas (areas beyond
national jurisdiction), for both pelagic and benthic
biomes – Global Open Oceans and Deep Seabed classifi cation – has also recently been published, using both
83
4.6 NATURAL UNITS IN
THE MARINE REALM
As on land, both compositionalist and functionalist
approaches are used to map marine environments.
However, there are important factors of the marine
realm that differ from the terrestrial, affecting the distribution of organisms and habitats, limiting our
understanding (the Linnean and Wallacean shortfalls
are even more pronounced in the sea than on land) and
determining the tools that can be used to create biogeographical classifi cations. Some of these physical, biological and socio - political factors are highlighted in
Table 4.6 , as they have important implications for
marine biogeography and the development of conservation initiatives (see Chapter 5 , Section 5.4 ).
The marine realm is more heterogeneous, fl uid and
obviously interconnected than the terrestrial, and the
high density of water enables a fully pelagic lifestyle
that is absent on land. Indeed, over 90 per cent of the
marine realm ’ s living space is pelagic and pelagic
systems support the majority of marine biomass.
Pelagic ecosystems represent a three - dimensional
continuum from the surface, epipelagic layer through
to the deepest trenches or bathypelagic zones. Water
chemistry, salinity, depth/pressure, currents and
primary productivity gradients create different regions
within these systems. Demersal or benthic ecosystems
form a two - dimensional layer from the shoreline to
deep sea, where bathymetry (depth) corresponds to
elevation within the terrestrial realm and in which different physiographical features such as seamounts,
submarine canyons, trenches, hydrothermal vents,
volcanoes and hypersaline lakes are embedded.
Overlain on the physical underwater landscape are different biogenic habitats, e.g. sea grass, coral reefs,
oyster beds, and maerl (maerl beds are formed by
certain species of calcifi ed red seaweed).
Efforts to undertake biogeographical classifi cation of
marine systems have a lengthy history, dating back to
work on coral reefs by James D. Dana published in
1848 and on molluscs by S.P. Woodward published in
1856 (both cited in Hedgepeth, 1957 ). Perhaps the
most widely accepted functional classifi cation divides
the sea into neritic (continental shelf) and oceanic
(beyond the 200 m isobath) zones (Hedgepeth, 1957 ).
Within the neritic zone are four primary biomes: estuarine, coastal marine, demersal shelf and pelagic shelf.
Within the oceanic zone are four primary biomes:
continental shelf, abyssal, epipelagic and meso/
bathypelagic.
Regions of open ocean were fi rst demarcated by
direction, velocity and persistence of currents (Dietrich,
1963 ), which provided the basis for the development
of a system of domains and divisions of the sea (Bailey,
1998 ). Temperature was also highlighted early on as
an important controlling factor of species ’ distributions and, hence, as a basis for spatial divisions at the
highest level of biogeographical classifi cation (Ekman,
1953 ). Faunal records and per cent endemism have
been used to defi ne provinces (Ekman, 1953 ; Briggs,
1974 ) and a combination of geomorphology and biotic
associations have been used to defi ne coastal zones
(Ray, 1975 ) (Table 4.7 ).
These older qualitative efforts can be distinguished
from more recent applications of multivariate clustering algorithms as applied to faunal composition (Ray
& Hayden, 1993 ), interpretations of remotely sensed
oceanographical data (Longhurst, 1998 ), biophysical
data (Harris & Whiteway, 2009 ), hierarchical geophysical approaches (Roff et al ., 2003 ) and systematically collected community data. Such analyses can be
repeated and tested for sensitivity to assumptions
(Shears et al ., 2008 ).
The urgent need for marine biogeographical classifi -
cations in order to proceed with immediate conservation goals has stimulated the development of numerous
regional marine systems, for example in Australia,
using both a compositionalist and controlling factors
approach (Environment Australia, 1998 ), and in
Canada, primarily using controlling factors (Roff et al .,
2003 ). A functional approach has also been taken in
developing the non - hierarchical set of Large Marine
Ecosystems (Sherman, 1993 ).
Recently, Spalding et al . (2007) reviewed the global
and regional marine classifi cations in order to develop
a consensus map (the Marine Ecoregions of the World)
of nested realms, provinces and ecoregions, at a scale
appropriate for global/regional conservation planning
in continental shelf regions (Spalding et al ., 2007 ).
Since different classifi cations resulted in different
boundaries being drawn, comparisons were made and
expert regional advice sought in order to determine the
best ecoregional defi nitions (Figure 4.15 ).
A classifi cation of the high seas (areas beyond
national jurisdiction), for both pelagic and benthic
biomes – Global Open Oceans and Deep Seabed classifi cation – has also recently been published, using both