CHAPTER 22 . (hemometric Applications to Seawater Analysis
417
and 5 m depth waters is similar, the PC score values are different, to indicate that, although an equilibrium ratio between PAHs in surface water and in 5 m depth water
may exist, the PAH composition is also governed by specific processes occurring in
each layer, as on the other hand evidenced by the vertical distribution. As concerns nalkanes in surface water, PCA analysis shows no connection between the geography
of sampling and the composition in n-alkanes in the samples. Also, for water from 5 m
depth the distribution of the PC scores is not connected with the geography of sampling, but the PC score plots for surface water and water from 5 m depth differ significantly. Likely, the vast majority of n-alkanes due to their extremely low solubility in
water are present in emulsion form or adsorbed on organic matter. For sediment
samples, the main factor governing the composition and concentration of n-alkanes
seems to be the depth from the bottom. In conclusion, in water and sediment samples
no connection was found between variable patterns and the geography of sampling:
these results suggest that n-alkanes cannot be used as a measure oflong term anthropogenic contamination.
The concentration of 28 PAH was measured in superlayer and underlayer deep sea
sediments collected at 15 sampling sites in the Gulf of Lion, in the Mediterranean Sea
(Domine et al. 1994). For comparison, samples of sediments of the Rhone River were
also collected. The Mediterranean Sea receives large amounts of toxic contaminants
from in flowing rivers and the atmosphere as well as industrial and tourist activities
along its coasts. In particular, PAHs are generated on land by the combustion of fossil
fuels and wood. PAH-Ioaden particles can be transported by the wind to distant locations and are then removed from the atmosphere by rain and dry fall-out. Some particles are deposited on water surfaces and then incorporated in the sediments, others
are redeposited on land and then delivered to natural waters by erosion and runoff,
followed by the resuspension and transport of contaminated sediments. The results
of GC and HPLC methods emphasise that contamination is relatively important in deep
sea sediments, and it is mainly due to material coming from the Rhone River. The data
were treated by means of the non linear mapping (NLM) method. The NLM method,
particularly suitable for the analysis of large sets of data, is a non linear display method
to represent the points of an n-dimensional space in a 2-D space, preserving as possible the interpoint distances. The NLM was performed both on the 15 sites with the
PAH concentrations as variables and on the 28 compounds with the site concentrations as variables and the results interpreted by means of multiple graphical displays.
It was possible to obtain a better knowledge of the PAH contamination of underwater
canyons and deep sea fans in Gulf of Lion. It also allowed some correlation between
the PAH structures and their origins and concentrations in the different samples. The
use of NLM coupled with graphical tools has shown that it is possible to easily visualise
the level of PAH contamination: lower concentrations are generally observed in the
underlayer sediments, with the only exception of retene and perylene which only in
fact have a biogenic origin.
22.2.1.7
Bioindicators
Information about marine ecosystem health can be obtained also from studies on the
state of flora and fauna present. Thus species of algae, fish, turtles, etc. are used as
417
and 5 m depth waters is similar, the PC score values are different, to indicate that, although an equilibrium ratio between PAHs in surface water and in 5 m depth water
may exist, the PAH composition is also governed by specific processes occurring in
each layer, as on the other hand evidenced by the vertical distribution. As concerns nalkanes in surface water, PCA analysis shows no connection between the geography
of sampling and the composition in n-alkanes in the samples. Also, for water from 5 m
depth the distribution of the PC scores is not connected with the geography of sampling, but the PC score plots for surface water and water from 5 m depth differ significantly. Likely, the vast majority of n-alkanes due to their extremely low solubility in
water are present in emulsion form or adsorbed on organic matter. For sediment
samples, the main factor governing the composition and concentration of n-alkanes
seems to be the depth from the bottom. In conclusion, in water and sediment samples
no connection was found between variable patterns and the geography of sampling:
these results suggest that n-alkanes cannot be used as a measure oflong term anthropogenic contamination.
The concentration of 28 PAH was measured in superlayer and underlayer deep sea
sediments collected at 15 sampling sites in the Gulf of Lion, in the Mediterranean Sea
(Domine et al. 1994). For comparison, samples of sediments of the Rhone River were
also collected. The Mediterranean Sea receives large amounts of toxic contaminants
from in flowing rivers and the atmosphere as well as industrial and tourist activities
along its coasts. In particular, PAHs are generated on land by the combustion of fossil
fuels and wood. PAH-Ioaden particles can be transported by the wind to distant locations and are then removed from the atmosphere by rain and dry fall-out. Some particles are deposited on water surfaces and then incorporated in the sediments, others
are redeposited on land and then delivered to natural waters by erosion and runoff,
followed by the resuspension and transport of contaminated sediments. The results
of GC and HPLC methods emphasise that contamination is relatively important in deep
sea sediments, and it is mainly due to material coming from the Rhone River. The data
were treated by means of the non linear mapping (NLM) method. The NLM method,
particularly suitable for the analysis of large sets of data, is a non linear display method
to represent the points of an n-dimensional space in a 2-D space, preserving as possible the interpoint distances. The NLM was performed both on the 15 sites with the
PAH concentrations as variables and on the 28 compounds with the site concentrations as variables and the results interpreted by means of multiple graphical displays.
It was possible to obtain a better knowledge of the PAH contamination of underwater
canyons and deep sea fans in Gulf of Lion. It also allowed some correlation between
the PAH structures and their origins and concentrations in the different samples. The
use of NLM coupled with graphical tools has shown that it is possible to easily visualise
the level of PAH contamination: lower concentrations are generally observed in the
underlayer sediments, with the only exception of retene and perylene which only in
fact have a biogenic origin.
22.2.1.7
Bioindicators
Information about marine ecosystem health can be obtained also from studies on the
state of flora and fauna present. Thus species of algae, fish, turtles, etc. are used as