Fróðskaparrit - 31.12.2000, Qupperneq 115

TITTLEIKIN AV VANLIGASTU FØROYSKU DýRA- OG TARASLØGUNUM
A KLETTASTROND I MUN TIL YMISK UMHVØRVISVIðURSKIFTI
119
cation of marine biotopes. Species noted as
frequent, common, abundant, or super
abundant in biotopes on eulittoral or
supralittoral rock (not including rockpools)
are marked in the table for the exposure in-
terval in which the biotopes occur. Since
quantitative descriptions of individual
species responses to wave exposure was not
the purpose either of Borgesen (1902;
1905) or Connor et al. (1996), our interpre-
tation is, thus, somewhat subjective and
should be considered with caution.
Assuming that wave exposure was the
main structuring factor at the sites with
tidal amplitude larger than 0.4 m, the ef-
fects of the substrate and fjord variables
seen in the Canonical Correspondence
Analysis (CCA) may have been partly due
to the correlation of these variables with the
exposure variable. There was, however,
some added effect from these variables,
particularly from the substrate variable,
that could not be attributed to the wave ex-
posure variable. Some of this added effect
might still have been connected to wave ex-
posure. A high substrate index signified
boulders, or, less commonly, stones (only
one station). Provided these were stable,
they might provide some shelter from wave
action. The wave-modifying effect of the
substrate appears plausible, particularly at
low wave exposure levels that occur at most
of the bouldery or stony sites.
The unique effect of the fjord index may
also be explained partly by wave exposure.
On the open coast, there may have been
more effects generated by reflected or de-
flected waves, or small islets may have pro-
vided less shelter than they would have in
fjords. These factors would lead to under-
estimation of the exposure at some open
coast stations by the wave exposure index,
which might be corrected by the fjord in-
dex.
The substrate and fjord factors may have
had other effects, but these were not detect-
ed unequivocally in the present analysis. A
high substrate index may signify reduced
stability of the substrate as well as hetero-
geneity with respect to light conđitions,
risk of desiccation, etc. A high fjord index
may indicate greater temperature variation
or, perhaps less likely, reduced salinity. The
stability of the substrate did not seem to be
reflected in the species pattern in any obvi-
ous way. In particular, the high score of
Ascophyllum nodosum along the substrate
vector could be difficult to explain. It is a
species that may need several years to reach
maturity and is considered to require stable
substrate (Baardseth, 1970). The fjord ef-
fect is difficult to untangle from that of
wave exposure due to the high correlation
between the variables.
It seemed most likely that wave exposure
was the main structuring factor behind the
first axis in the DCA and the CCA, as well
as in the partial CA, for the stations with
tidal amplitude larger than 0.4 m. This was
supported by the results of the forward se-
lection of variables in the CCA and the cor-
relation between the variables and the axes
in CCA and DCA, which suggested that
wave exposure was the most important
variable. This conclusion is further sup-
ported by the conformance with descrip-
tions made by other authors, and by the
possibility of explaining the effects of the