This paper not to he eiteu without prior rdercnce to the authors.
INTERNATIONAL COUNCIL FOR
THE EXPLORATION OF THE SEA
C. M. 1'.1X-I/D: 10
Statistics Committce
Rd.: Marine EnvironmcIlui
Quality Committce
TIME·RELATED CHANGES OF ORGANIC AND INORGANIC NITROGEN AN)) OTHER
INTER·RELATIONSHIPS INTHE SOUTHERN NORTH SEA OFFTHE DUTCH COAST;
ASTATISTICAL ANALYSIS.
hy
H. Pictcrs anu H.B. Ikekcr.
Ncthcrlanus Institute for Fishcry Il1\cstigations
1'.0. Box (iX , )970 AB IJmuiucn'
Thc Ncthcrlanus.
This paper not to be cited without prior reference to the authors
INTERNATIONAL COUNCIL FOR THE
EXPLORATION OF THE SEA
C.M. 1984/D :10
Statistics Committee
Ref: Marine Environmental
Quality Committee
Time-related changes of organic and inorganic nitrogen and other
inter-relationships in the southern North Sea off the
Dutch coast; a statistical analysis.
by
H. Pieters and H.B. Becker
Netherlands Institute for Fishery
Investigations
P.O. Box 68, 1970 AB IJmuiden
The Netherlands
ABSTRACT
Monthly measurements of nutrients and salinity have been performed at
sixteen selected locations in the Dutch coastal area over the period
1980 - 1983. From these nutrient data organic nitrogen contents were
calculated as the difference between total nitrogen (both dissolved as
particulate) and the
sum
of
nitrate,
nitrite
and
ammonia
concentrations.
By means of a multiple linear regression model applied to combined
sub-datasets of two neighbouring locations the authors were able to
elucidate the influence of five variables (nitrate, nitrite, ammonia,
salinity and river Rhine discharge) on the occurrence of organic
nitrogen (ORGN). T - ratios and fractions for each variable to total
variance of ORGN were calculated, resulting in a spatial distribution
of significant and non-significant contributions to ORGN.
Partial regression plots were produced in order to identify outlying
data points, whereas most of them could be omitted on account of very
high values of ORGN in the winterperiod. The relative influences of
these five variables on the spatial occurrence of ORGN, indicated by
their contributions to total variance of ORGN reflects closely the
average distribution of river Rhine water in the Dutch coastal area.
INTRODUCTION
This paper gives a preliminary description of the nutrient variations
in space and time in the Dutch marine waters and' concerns mostly with
the changes of and relations between
organic and inorganic nitrogen
in that part of the coastal area that is directly influenced by the
river Rhine outflow (see fig. 1).
PAGE 2
Large quantities of nutrients (nitrate, ammonia, phosphate) from
domestic and industrial sources are carried to the coastal waters by
the freshwater run-off from the Rhine and some minor sourees.
This
nutrient enrichment of the Dutch marine environment has effected an
increasing eutrophication in this area during the last decades (1,2),
concomitantly with higher production of organic material.
Due to the south-north directed residual-current along the Dutch coast
the out- flowing Rhine water is transported in northerly direction,
producing a flowing pattern with low salinities parallel to the coast.
This plume of river water contains a relatively high amount of
suspended matter and can therefore be observed from space by remote
sensing
techniques
distinctly separated from surrounding water
masses(3,4).
Simple dilution by mixing of river water and sea water results in a
conservative behaviour of nutrients in the coastal zone, e.g. a
linear decrease with increasing salinity. However several
other
processes
can
also
have
a
strong
impact on the nutrient
concentrations,
such
as
primary
production,
mineralisation,
adsorbatory and sedimentary processes.
Due to these influences
deflections from conservative mixing can be found with higher or lower
nutrient values than expected from mixing only.
As a result two important gradients occur in the Dutch coastal zone.
Going offsshore from the Dutch coast salinity increases gradually
and nutrient concen- trations decreases and going to the north from
Hoek van Holland to Texel com- parable changes in these parameters are
detectable. As a matter of fact a lot of fluctuations and anomalies
in the overall distribution of nutrients will often occur due to
stream variations, windintensity, phytoplankton blooms and changes in
the river outflow.
•
In this complex picture of nutrient movements we have tried to
elucidate the relation of the organic nitrogen content to some
crucial variables as salinity, inorganic nitrogen species (nitrate,
nitrite and ammonia) and the monthly averaged discharge of the river
Rhine. By applying an interactively used multiple regression model
it was possible to identify extraordinary data points and to correct
for it.
The aim of this paper was to reveal the possible usefulness of such a
regression model to interprete nutrient data sets in estuarine waters.
PAGE 3
METRODS AND ANALYSES
Analytical methods:
Water sampies (surface) were collected every month from ships of
Rijkswaterstaat at selected locations along the Dutch coast. The
position of the sampling sites are shown in figure 1, arranged in four
lines off Callantsoog, Egmond, Noordwijk and Terheyden respectively.
After transportation to the laboratory the samples were filtrated
through 0.45 u Oxoid filters and kept deep frozen until analysis.
All analyses, except for ammonia, were carried out with Technicon
Auto-analyzerr 11 equipment. Total-nitrogen analyses were performed
in the unfiltered water according to the method of Koroleff (5).
Nitrogen compounds were oxidized to nitrate by boiling with an
alkaline persulfate solution. Nitrate ions were reduced then in a Cd
Cu reductor column and subsequently analyzed as nitrite ions
according to the method of Bendschneider and Robinson (6), whereby nitrite reacts to a diazo compound with sulfanilamide and then couples
with a diamine to form a reddish-purple Azo dye.
Ammonia analysis was performed according to the manual method as
described by Koroleff (7), with citrate buffer as a complexing agent
to avoid precipitaties of calcium- and magnesiumhydroxides at the
alkaline pR of the reaction mixture as described by Solorzano (8).
Description of variables:
•
TOTN = total nitrogen.
NTRA = nitrate.
NTRI
nitrite.
AMMO = ammonia.
SALI = salinity.
RAF = river Rhine discharge / monthly averages(9).
ORGN = TOTN - NTRA - NTRI - AMMO = organic nitrogen.
Statistical analysis:
A multiple linear
between ORGN and
NTRI and RAF at
combined datasets
regression model is used to examine the relation
the following nominated carriers: NTRA, SALI, ~10,
different locations (see fig.
1).
The following
were compiled:
C3IO, CI020, C2030
E3IO, EI020, E2030
N310, NI020, N2030
T310, TI020, T2030
CE3, CEIO, CE20, CE30
EN3, ENIO, EN20, EN30
TN3, TNIO, TN20, TN30
C3IO = C3 + CIO; C3 = 3 km off the coast
at Callantsoog.
CE3 = C3 + E3
P~E4
On each of these combined datasets multiple regression of ORGN on the
other carriers was carried out. Plots of residuals against fitted
values did not show any significant trend and the residual Gaussian
quantile - quantile plots were reasonable straight.
In order to identify extraordinary data points partial regression
plots for the carriers NTRA, SALI and AMMa were produced (10). As an
example a partial reg- gression plot for NTRA and dataset EN20 is
given in fig. 2. On the Y-axis the residuals from the least-squares
regression of aRGN on the other carriers except NTRA is plotted(r1)
and on X-axis the residuals from the regression of NTRA on the other
carriers(r2).
This plot has a least squares slope equal to the
regression
coefficient of NTRA in the full model and least squares residuals
equal to the final residuals of the full multiple regression model.
lf the outlying data points in the partial regression plots seem to be
unrealistic after inspec- tion of the observed values; they were
omitted from the datasets.
An other multiple regression was then
carried out with the reduced dataset~ In our example in fig. 2 five
points have been omitted as indicated.
One of these points (with the lowest r1 value) has negative value for
ORGN and the other four points have a value for TaTN which is too high
for the winter period (January, February and March), and is probably
caused by false measurements.
RESULTS
Gradients:
•
In figure 3 some examplcs of the occurrence of nutrient gradients in
the Dutch coastal area are drawn. Mean values of TaTN, NTRA and AMMO
for the summer period (June - October) of the years 1980 - 1983 have
been used. The decreasing con- centrations of the nutrients at longer
distance from the coast are in agreement
with
the
salinity
distribution of figure 4. The lowest salinity values off Terheyden
indicate percentages of Rhine water greater than 30 %. The continious
mixing with the inflowing freshwater diminishes the concentration of
the nutrients also in northerly direction (fig 3B), hence much lower
values are found off Callantsoog.
The strong influence of the discharge of the river Rhine is shown in
fig 4A a and 4B.
The high discharge of freshwater in August 1980
produced lower salinitities at longer distances from the coast and
would also have a large impact on the nutrient values in the northern
part of the coastal area.
The relationship between nutrient concentration and salinity is shown
for nitrate and ammonia in fig 5, estimated with da ta of the
winterperiod of 1981/1982.
Nitrate behaves conservatively duc to
normal mixing, while ammonia shows a non-conservative behaviour with
lower values than according to mere mixing processes.
PAGE 5
Seasonal chariges:
The seasonal changes in organie nitrogen and nitrate. given as the
monthly mean values at 20 km off the Dutch eoast. are depicted in fig.
6. Low values for organic nitrogen could be calculated in the
wintermonths and elevated valuess in the summer period. corresponding
with the growing season. For nitrate an reversed picture has been
obtained.
It is thereforeclear that an inversely linear relationship
for nitrate and organic nitrogen has been visualized: consumption
of nitrate by the phytoplankton to produce organic nitrogen.
In figure 7A one can observe a large deviance for this relationship
due to the influence of many other variables. which all have a more or
less strong impact on the formation of organic nitrogen.
~
Hence. a multiple regression model applied to.the total data set of
the examined time period might reveal more informatioQ about local
impacts of each variable on the production of organic nitrogen.
Trend analysis:
•
To apply multiple regression analysis on the total dataset obtained
from the monitoring program of 1980 till 1982. it was necessary to
deduce if a significant change in the nutrient load of the coastal
area did occur during this time interval Only without theoccurrence
of such a change it was allowed to integrate all sub-datasets of the
different years of investigation to one cODbined dataset for each two
localities. To perform such a trend analysis the relation of total
nitrogen and nitrate to salinity was estimated. The concentrations
of these nutrients at a fixed salinity for each year establishes then
the yearly fluctuations of the nutrient influx and could reveal a
possible trend. Only data collected in the winterperiod (Dee. Jan and
Feb) have been used to avoid impacts of primary production and
mineralisation processes(11) •
The results of this exercise are shown in fig. 7B. The variation of
nitrate lies within the range of the standard error. hence no
significant trend has been demonstrated.
Multiple linear regression:
The configuration of the regression model is as foliows:
ORGN
=A+
Bx(NTRA) + Cx(SALI) +Dx(AMMO) + Ex(NTRI) + Fx(RAF)
The coefficients B to F represent the regression coefficients in the
full regression model.
T
ratios ( = estimate,of the regression
coeffieient in the full model divided by its standard error) have been
calculated to reveal
significant terms in this model. For t ratios smaller than 2. the variable should not be accepted as
significant term. Tables 1 - 2 give the t - ratios and the fractions
of the total variance of ORGN accounted for by NTRA t SALI and AMMO.
These fractions are reproduced in figures 8a t 9a and 10a t whereas the
spatial configurations of the t - values for the five variables are
given in figures 8 - 12.
PAGE 6
From the configurations of the t - values and the fractions of total
variance of ORGN for NTRA, SALI and AMMO the following general remarks
can be made:
1. Nitrate(NTRA) has in the whole coastal area (except for C310
and CE3) significant contributions to the total variance of
ORGN (organic nitrogen) with the highest fractions at 3 and
20 km off the coast andminimum values at about 10 km (fig. 8, 8a,
values within rectangles).
2. At longer distance from the coast the significant fraction to
total variance for the salinity (SALI) increases, while at 10 km
hardly any significant contribution is observable (fig. 9, 9a,
values within rectangles).
3. At scattered locations ammonia (AMMO) appears to have a minor
fraction to total variance of ORGN with the highest values near
Callantsoog (fig. 10, 10a). However, at 10 km off the coast no
significant contribution could be detected(values within rectangles).
4. Nitrite (NTRI) has only significant contributions to the total
variance of ORGN near the river Rhine outflow(fig. 11,values
within rectangles).
S. Only at about 10 km off the coast the discharge of the Rhine
has a significant contribution to ORGN (fig. 12,values within
rectangles). With all the other datasets of the combined locations
no significant contribution has been found.
All the above mentioned features can be related unequivocally to the
plume of river Rhine water outflowing into the Dutch coastal zone:
Rhine water contains high amounts of suspended matter and therefore
turbidity is highest at those locations in the coastal area where the
percentage of river Rhine water is maximal, consequently at about 10
km off the coast.
Light intensisity might be here occasionally the limiting factor for
primary production (12) and will therefore be more important for the
production
of organic nitrogen
than
the
elevated
nutrient
concentrations present in excess.
It is also evident that the
discharge of the Rhine will have some impact just at these 10- cations
because of its positive correlation to the turbidity of the coastal
zone. At longer distance from the coast light intensity increases
highly and so nitrate and salinity (closely related to nutrient
concentrations, see above) become more important for the formation of
organic nitrogen.
Simularly, nitrate and ammonia exhibit significant contributions to
ORGN with a strong impact on its formation at the shallow waters
inshore, where light intensity is not limiting.
Nitrite is an intermediate compound in mineralisation processes and
will have more influence on the formation of organic nitrogen at
locations where mineralisation is favoured above primary production,
thus near the outflow of river water into the coastal area, because of
the high concentrations of suspended matter present and stimulated
bacterial activity (12).
PAGE 7
Conclusions:
- At about 10 km off the Dutch coast minor contributions to the
total variance of ORGN have been estimated for nitrate, ammonia
and salinity, while river Rhine discharge exhibited the only
significant contribution just at this very distance from the coast.
- The spatial configuration of the relative importance of the five
variables in terms of accounting for the total variance of
=organic nitrogen, measured over a time period of three years,
reflects closely the average distribution of the plume of river
Rhine water outflowing in the North Sea along the Dutch coast.
From our results it can be inferred that the application of an
interactively used multiple regression model may be very valuable
to reveal the influence of nutrient enrichment of coastal and
estuarine zones on the production of organic material:
REFERENCES
•
1) P.Hagel and J.W.A. van Rijn van Alkemade, Eutrophication of the
North Sea,
lCES C.M.1973/L:22
2) A.J.van Bennekom et al, Eutrophication of Dutch coastal waters,
Proc.R.Soc.Lond.B.189 ,359-374(1975)
3) P.L.Larsen and P.C.Jorgensen, Coastal zone color scanner data
applied for mapping of the phytoplankton pigment
concentration in the North Sea.,
report 1981
4) F.Vegter, Nutrient intrusion by the river Rhine in the Dutch
Coastal waters,
lCES C.M.1971/C:38
5) F.Koroleff, Determination of total nitrogen in natural waters by
means of persulfate oxidation,
lCES C.M.1969/C:8jrevised 1970
6) K.Bendschneider and R.J.Robinson, A New Spectrophotometric Method
for the Determination of Nitrite in Sen Water,
J.Marine Res. 11(1952)
7) F.Koroleff, Direct determination of ammonia in natural waters as
indophenolblue,
lCES C.M.1969/C:9
8) L.Solorzano, Determination of ammonia in natural water by the
phenolhypochlorite method,
Limnol.Oceanogr.4(1969):799-801
9) Kwaliteitsonderzoek in de rijkswateren,
kwartaalverslagen RIZA,Lelystad
10) H.V.Henderson and P.F.Velleman, Building Multiple Regression
Models lnteractively,
Biometrics 37(1981) 391
11) A.R.Folkard and P.G.W.Jones. The distribution of nutrient salts in
the southern North Sea during early 1974,
lCES C.M.1974/C:17
'
12) F.Colijn, Primary production in the Ems-Dollard estuary,
dissertation,1983
rA BL E 1.
Ca 1 C IJ 1 a ted t v 2. llJ e 'E an d fra c t. i
01"1 S
t1
f
t
('I
t ;d v i1
rj
an ce elf 0 RGN f r' CI III t hp
nil I
1 t i. F' 1 Po
r ., S res<.: i
0
r,
111
n d ... J.•
.
======================================================
==========~========================~================~::='
I
I1
C310
I
Cl020
I
C2030
1
E310
I
E1020
1
E2030
1-------1 1----------------1----------------1----------------1----------------1----------------1--------------I
I1
: f r
ti rl1
: f r c t ion I
: fra r' t. ). r iI
: fra c t i
I
: f r'
t. j. n I
:f r
t on
1v r i a t, 1I t r t 1. n f t (") t 1. I ;', rat i. f t, tal I t r t i. n : f t, t, 1 1t a t 1. f t, t.;; 1. I t r t i f t.. t nIl t r t i f tot al
Cl C
Cl
~
p.
~
0
0 :
<'l
I") :
0
0
<'l
I")
r)
0
ill:
01"1
I'
"3
0 : ,")
0
;:j
I") : I")
0
Cl C
~
0
I") :
1
0
I
11
:variancel
:variah~el
:variancel
:vari~ncel
:varjancpl
:variance
1-------11----------------1----------------1----------------1----------------1----------------1--------------·
!
1
I I
INTRA
1 SALI
I AMMO
1 NTRI
I RAF
1I
I
I
I 4.37
O.~O
11 2.50
I I ~.1A
1 I 0.28
I I 0.79
0.08
0.16
I
I
1
1
0.19
0.12
3.63
0.38
0.31
2.A3
I
~.13
0.06
1 4.53
O.?~
1
I
I
I
3.70
1.49
0.22
1.ht
0.1f~
I 2.57
I 0.94
1 0.65
I O.A~
0.07
0.01
0.03
.I
I
I
I
I
I
I
I
3.76
3.40
1.22
0.78
0.15
0.13
0.02
0.~7
I
I
1
I
1
O. , :'
4.62
6.79
2.07
0.43
0.39
O.;\f,
0.03
~====================~================================ ===============================:=======================::
1
II
N310
I
Nl020
I
N2030
I
T310
1
Tl020
I
12030
1-----------------_··_---------------------------------------------------------------------------------------_.
I
I I
1
I
I
1
I I 4.15
114.03
1\ 1.87
I I O.RO
NTRA
SAI.l
AMMO
NTRI
I F: AF
I I
I
1. 15
0.21
0.?0
0.04
1
I 5.37
I 4.09
1 3.47
I 4.~R
I
O. 7 17'
0.22
O.l~
0.09
I 4.94
14.45
I ~.?8
:
1 5.h2:
1 O. 3 ,L.
I
0.17
0.l4
0.04
1
1
I
I
I
2.73
S.06
2.13
2.]9
I O. 7 f.
0.08
O.;~9
O.O~
1
I
I
I
1
I
~.92
0.10
4.03
?OO
0.77
O.~O
1. 4 9
0.05
I 4.01
16.41.
I 2.4;'
I 0.02
O. 15
O.4()
("I.
I O. 4 0
I
Cl:>
I
\J,i.,
::=======================::=
I
II
=================~=====~=========================~=
CE3
CE10
1-------1 1·------------·-·--·I
I I
:fraction
1v a r i a tel I t, rat i (> : .J f I', 0 t, a 1 t
I
11
:variance
CF.20
:fraction
ra~io:of
tot~l
t
:variance
CE30
:fraction
:fraction
ratio:of tot~l t ratin:of total
:variance
:varj~nr~
1-------11---------------I
INTRA
I SALI
tl
O.OJ
0.09
4.::.! 1.
0.17
4.41
O.?O
3.03
1 .01.
o .1.~
4.72
0.::t:5
1..85
0.03
0.22
3.51
5.9<'>
O.;"lo
2. t'j7
I AHMO
I I 3.9:-;
0.07
0.30
0.53
I NTRI
I I 0.02
0.15
0.73
0.51
I RAF
I I 0.48
1 .98
1.07
0.04
:::===================:::====================================== ================I I
Etl] ..M.
EN10
I::.N20
EN30
:
..
...
.
.
1I
I I
._~
NH,A
SAL I
AMMO
NTRI
RAF
,
.
:
I
I
I
I
I
I
I
I
I
~
.~
I
.1.08
3.07
2.11
0.89
1. 44
0.20
0.16
0.05
0.10
0.02
3.05
1 .41
0.03
1 .97
1.88
7. :1 ?
7.04
5.61
7.01
0.92
0.21.
3.06
o • 11
O.?O
5.2~
0.31
0.1 ::t:
0.4<'>
O.~O
0.47.
::~============::======:==============================
===========================
I
TN3
1I
TN10
I
TN20
I
TN30
I
I
1----------------------------------------------------------------------------I
I
I
I
I
I NTRA
I
I
I
I
SALI
AHMD
NTRI
RAF
I I 3.55
5.40
I I 1.40
I1
11
I I
0.13
0.31
0.07.
I 3.tt
3.78
I 1.64
I
0.10
0.15
0.03
I 5.77
6.09
I 2.89
I
1.98
I
2.3::'
I
1.1.2
I
3.0~
I 0.30
5.4l
0.20
0.23
0.05
I 3.45
5.00
I 1.59
I
0.1.4
0.29
0.03
1
I
I
0.21:
I
I 0.31:
I
I
I
\0
I
- 10 -
•
I·
"t
FIG. 1: THE DUTCH COASTAL AREA WITH THE
SAMPLING SITES.
Eg mond
52
•
70.
3(}
..
'-..
. -.....~
40
~m
Ol
- 11 Rl
;·:100
FIG. 2.
8.00
7.00
o
6.00
o
5.00
4.00
o
3.00
2.00
•
1.00
0.00
• • 00
•
0
*
*. * *
* *** i*
*
*
**
*
* * * t *. * *.• *
*
.* * *.
i
* *•
*
* ••*
*
~
••
*
*
•
*
* *
-2.00
0
-3.00
-4.00
+-------t-------t-------+-------t-------t-------t-------t-------+-------t-------+-------+-------t
-4.00
-2.40
-0.80
0.80
2.40
4.00
5.60
r-:2 :.:
PARTIAl. REGRESSION PLOT FOR NTRA
******:
NOT ONITTED
naaaoo: aMITTED
: : : : : :: DATA (lVERLAP
~
00
-
12 -
Y xtOO
FIG. 3A: DECREASING VALUES FOR NUTRIENT CONCENTRATIONS WITH INCRESING DISTANCE FROH
THE DUTCH COAST OFF NOORDWIJK.
y= TOTAL NITROGEN
y= NITRATE
Y= AHHONIA
in mg/m}.
z.
z.
I.
I.
o.
O'~~--=:::===:;;;===~~=------;;i
0.00
15.00
30.00
<5.00
60.00
15.00
90.00
iJISTANCE (KM)
NUTRIENTS IN COASTAL
~ATERS
Y xtOO
6.00
•
FIG. 38: DECREASING NUTRIENT VALUES IN SOUTH
- NORTH DIRECTION AT 20 KM OFF THE
COAST. ALL DATA ARE HEAN SUHHER
VALUES.
oooeo
Y= TOTAL NUROGEN
+++++
Y= AMHONIA
y. NITRATE
in mg/m 3 •
O·~0":.OO-:----':':IS"'.OO"""----:::JO"".OO:::------:';<5"'.0::::0----60~.O;::0----;:;'s~.o;;;O-----;;;;90 oe
iJISTANCE (KM)
NUTRIENTS IN COASTAL ~ATERS
SOUTH-NORTH OIRECTION
fIG. 4A: SALINITY DISTRIBUTION IN
fIG. 4B: SALINITY DISTRIBUTION
DUTCH COASTAL WATERS
34.0
IN DUTCH COASTAL WATERS
LOW RHINE WATER DISCHARGE.
AUGUST 1980 ,HIGH RHINE
•
WATER DISCHARGE.
9
•
•
•
70.
•
w
"
40 KM
..
40 KM
I
!
FIG. 5: CONSERVATIVE AND NON-CONSERVATIVE BEHAVIOUR OF NITRATE (A) AND AMMONIA (B)
RESPECTIVELY, DURING THE WINTERPERIOD OF 1982. DATA USED WERE MEASURED AT
ALL SAMPLING SITES.
AHHONIA xl00
NITRA TE x1000
1.20
A
SALINITY
SALINITY
RELATIDNSnIP NUTRIENTS AND SALINITY
CDNSERVAT!VE 8EHAVIDUR OF NITRATE
B
6.00
RELATIONSHIP NUTRIENTS AND SALINITY
NON-CONSERVATIvE 8EHAVIDUR OF AMMONIA
Y x100
FIG. 6
y= NITRATE
++++ +
y= ORGANIC NITROGEN
Ir
)/
\
\
/
I
f
I
~~
1
__-'-
32.00
.1-
40.00
. ..o-._----l
48.00
MONTH NUMBER
SEASONAL VARIATIONS OF NUTRIENTS
ORGANIC NITROGEN AND NITRATE AT 30 KM OFF THE COAST
_ _,
•
~J
..-J_
FI~
• 7A
Plot of DRGN versus NTRA for data set EN1D.
ORGN
bOOt
*
0*
570.
540.
510.
480.
450.
420.
390.
360.
330.
300.
270.
240.
210.
180.
150.
120.
90.0
60.0
30.0
).000
•••••
0
0
0
*
*
*2
tOO
0
0
0
*0
:j
0
000
7-
•••
0.000
00
0
0
0000
0
0
0
00 0
0
0
0
.' 0
"l
"-
0
0
*
t
0
022
~2
*0 00
*l:
*
*
*
*
**
o
0
0
0
0
0
t
.*
250.
*
500.
*
750.
0
0
*..
t
••••••
O.100Ef04
*
*
O.12~E+04
NTRA
O.150Ef04
0'\
I
I
Y x1000
1.41
1.3
I
h. h. h. h. h.
I
I
Y= NITRATE
Y= TOTAL NITROGEN
1.2(}_
I
,
I
. I
U(}_
FIG. 7B: TRENDS OF NITRATE AND TOTAL NITROGEN.
i
FOR EXPLANATION SEE THE TEXT.
I
!
1.0dr-
I
0.90
r
II
0.8C!;.,
i
i
0.7ri.
,
1
!
0.60_
i
i
0.50...
I
0.40_
!
0.30_
o
!
0.2L_--,.__ ._L_ •. _.L_.. _.---1
74.00
76.00
78.00
..- ...--.__.L__.--l.....__ ..l...
80.00
82.00
-L._._l.. __ --l---l
84.00
86.00
YEAR
TRENDS OF NUTRIENTS
NUTRIENTS-YEAR VARIATION
I
- - - - - - - - - . . , - - - -18-
53
30 KM
20 KM
10 KM 3 KM
4.37
~
Ca Ifantsoog
NORTH SEA
C")
o
C")
E g mond
'0
.
o
C")
-
C'\t
IJMUIDEN
•
N oordwij k
. 'FIGURE 8.
t
Terheyden
va lu es
.far NTRA
o
30 KM
20 KM
0006[
10
KM
3 KM
o . 19
Co
f:2J
NORTH SEA
110 ntsoog
0..
0
0
~
0
"-
0
o
I
0·22
Eg mond
--
o
-
~
o
IJMUIDEN
N oordwij k
FIGURE eA.
NT RÄ· fractions of
, .
total var~ance
0 f ORGN
Terheyden
o
52
- 20 -
30 KM
f
~
NORTH SEA
V)
20 KM
3.70
f
Ca I (a n t 500 9
~
"'t
6.79
3.40
I
,.
Egmond
/
i
/
i
/
!
I
I
I
~
er;
~
".
•
I
I
/
0
/
I
/J
/
I
I
I
i
i
/
/
R
C'J•
IJMUIDEN
N oordwijk
·FIGURE 9.
t
Terheyden
values rar SALI
- 21 -
53
30 KM
/ 0.,.
~
NORTH SEA
C")
o
0.36
20 KM
!
Ca 110 ntsoog
'"
'"
o'
o. r3
Egmond
/
/
I
/
I
i
I
I
-
/
J
f'?
o
•
/
/
lf7
lsJ
i
I
,i
~
JMU I DEN
o'
N oordwij k
. FIGURE 9A.
SALI; fractions of
Terheyden
total variance of ORGN
o
=-------I~-:22l-
530
30 KM
20 KM
(
',49
10
0·38
KM
3 KM
3'38 t-
NORTH SEA
Ca (fa ntsoog
!
I
!;{
/
o
2.87
/
i
Eg mond
/
I
I
!
I
,-
/
,/
'0
~
o
"'
'0
IJMUIDEN
lt)
•
/
~/
l).,'
~·~S/
/
/
/
/
/
N oordwij k
FIGURE 10.
t values far AMMO
RHINE
OUTFLOw'
Terheyden
,5-2°
'._
- 23 -
53
o
30 KM
20 KM
10 KM
3 KM
NOHTH SEA
6
O"
r
Ca Ila ntsoog
,
.'
I
"
I
o
o
!
o
C"I
o
0.03
0.02
Egmond
/
-.
C'?
o
ß
IJMUIDEN
0Q
.<t
,
./
./
/
N oordwij k
FIGURE 10A.
RHINE
OUTFLOW
AMMO; fractions of
Terheyden
total variance of ORGN
5 2(
- 24 -
53
30 KM
20KM
10 KM
0.22
0.3,
NORTH SEA
0
...,
.....
0
.
/0
-.
lt'l
3 KM
28
Ca 110 ntsoog
I
-.
0
lt'l
0
N
0
0
.
N oordwij k
FIGURE 11.
t
Terheyden
values. for
NTRI
o
- 25 -
30 KM
20 KM
10
1 • 61
KM
f!:EJ (O.79/3/rKM
NORTH SEA
.
o"
~
{Jj
•
e.·
Ca ffa ntsoog
~
.
'<;/"
o
0.39
I
Egmond
/
.
t\I
'<;/"
o
/
IJMUIDEN
/
()
.
~()
/
/
N oordwij k
FIGURE 12.
t
RHINE
OUTFLOW
Terheyden
values.for
RAF