UNDP/GEF Danube Regional Project
Strengthening the Implementation Capacities for Nutrient
Reduction and Transboundary Cooperation
in the Danube River Basin
Orientation on environmental quality
standards for nutrients and other Danube
specific priority substances
Project Component 2.2: Development of operational tools for
monitoring, laboratory and information management with
particular attention to nutrients and toxic substances
October 22, 2003
Prepared by: Rodeco Consulting GmbH
Author:
Paul H.L. Buijs
Preface
In its original settings, the activities under the project tasks 2.1 and 2.2 envisaged to develop (2.1) a
proposal on water quality objectives for nutrients in line with requirements of EU WFD, and (2.2) a
proposal on water quality standards for toxic substances from the Danube River Protection Convention
list of priority substances.
In total two person-weeks of input were allocated for these tasks. With such limited time, it is not
feasible to cover the subjects in-depth and to be exhaustive for all different regions within the Danube
basin. Therefore, while aiming at formulating preliminary recommendations for Environmental
Quality Standards, it was considered equally important to provide background material and arguments
to the MLIM and EMIS expert groups for supporting their further activities. Besides the amount of
information included in this report, the references contain many links to WebPages in Internet from
where more (background) information easily can be retrieved.
The Draft Final report of August 2003 was discussed during the Second Joint MLIM/EMIS meeting in
Bratislava (September 17th, 2003) and the 31st MLIM meeting on September 19th, 2003 (in Bratislava).
A number of questions and comments were raised during the meetings and the author received e-mails
afterwards summarising the comments. Some of the remarks critically commented parts of the report,
like not having followed the typespecific approach in line with the Water Framework Directive.
The author decided to add the several questions and comments, together with responses by the author,
in the separate `Epilogue' chapter 7. The remaining contents are similar to the draft final report. By
doing so, it remains more clear on which specific grounds the various comments were based, and why
certain items were raised as discussion points. In this way, it also remains more transparent how to use
(and not to use) the report during further discussions and studies.
Paul Buijs
22 October 2003
Amersfoort, the Netherlands
Orientation on environmental quality standards for nutrients and other Danube specific priority substances
I 1
Table of contents
Preface
Abbreviations, Acronyms, and Definitions
Executive summary
1 Introduction.............................................................................................................................I - 7
1.1 Scope.................................................................................................................................I - 7
1.2 General departure points......................................................................................................I - 7
1.3 Comments to draft report.....................................................................................................I - 7
2 Nutrients (Ntot, Ptot)...................................................................................................................I - 8
2.1 Nutrients and the WFD........................................................................................................I - 8
2.2 Considerations for formulating EQO and EQS for nutrients in the Danube .............................I - 9
2.3 Natural background concentrations of nutrients...................................................................I - 10
2.3.1
Nitrogen...................................................................................................................I - 10
2.3.2
Phosphorous .............................................................................................................I - 12
2.3.3
Historic data on nutrients in the Danube .....................................................................I - 13
2.4 Nutrients levels avoiding risks for eutrophication (`good status') .........................................I - 14
2.4.1
Nitrogen...................................................................................................................I - 14
2.4.2
Phosphorous .............................................................................................................I - 17
2.5 Comparison of proposed EQSs with actual concentrations ...................................................I - 19
2.5.1
Seasonality...............................................................................................................I - 19
2.5.2
Joint Danube Survey.................................................................................................I - 19
2.6 Synthesis and discussion of previous findings .....................................................................I - 23
3 Ammonium (NH4)..................................................................................................................I - 25
3.1 Introduction......................................................................................................................I - 25
3.2 Inventory of EQSs for NH4................................................................................................I - 25
3.3 Synthesis of findings for `good status' of NH4 ....................................................................I - 26
3.4 Comparison of proposed EQS with actual concentrations ....................................................I - 27
3.4.1
Joint Danube Survey.................................................................................................I - 27
3.4.2
TNMN .....................................................................................................................I - 27
3.4.3
Seasonality...............................................................................................................I - 28
4 Chemical oxygen demand (COD)............................................................................................I - 29
4.1 Introduction......................................................................................................................I - 29
4.2 Inventory of EQSs for COD...............................................................................................I - 29
4.2.1
United Nations..........................................................................................................I - 30
4.2.2
Danube basin ............................................................................................................I - 30
4.2.3
Japan (lakes).............................................................................................................I - 31
4.3 Synthesis of findings for `good status' of CODMn................................................................I - 31
4.4 Comparison of proposed EQS with actual concentrations ....................................................I - 31
5 Metals (As, Cr, Cu, Zn) ..........................................................................................................I - 32
5.1 Introduction......................................................................................................................I - 32
5.2 Total, dissolved, adsorbed? ................................................................................................I - 32
5.2.1
Short primer on some key features of total, dissolved, and adsorbed metals ..................I - 32
I 2
UNDP/GEF Danube Regional Project
5.3 Natural background levels .................................................................................................I - 34
5.3.1
Examples of limitations for using the formulas in textbox 5.1 ......................................I - 36
5.4 Inventory of water quality criteria for metals resembling `good status' .................................I - 37
5.4.1
United Nations..........................................................................................................I - 37
5.4.2
European Union........................................................................................................I - 38
5.4.3
United States ............................................................................................................I - 39
5.4.4
The Netherlands; Rhine .............................................................................................I - 39
5.4.5
Canada.....................................................................................................................I - 41
5.4.6
Joint Danube Survey.................................................................................................I - 41
5.4.7
Synthesis of findings for `good status' of metals .........................................................I - 41
5.5 Comparison with actual metal concentrations .....................................................................I - 42
6 Conclusions and recommendations ..........................................................................................I - 43
6.1 Nutrients: Ntot and Ptot........................................................................................................I - 43
6.2 NH4 .................................................................................................................................I - 43
6.3 CODMn.............................................................................................................................I - 43
6.4 Metals: As, Cr, Cu, Zn ......................................................................................................I - 43
7 Epilogue: comments to the draft final report.............................................................................I - 44
7.1 Remarks to draft report submitted by e-mail .......................................................................I - 44
7.1.1
Germany ..................................................................................................................I - 44
7.1.2
Austria .....................................................................................................................I - 44
7.2 Heavy metals ....................................................................................................................I - 45
7.3 Typespecific approach.......................................................................................................I - 45
7.4 Background of the values mentioned in the report ...............................................................I - 47
7.5 Compliance testing ...........................................................................................................I - 48
7.6 Concentration and loads ....................................................................................................I - 48
7.7 Closing remarks................................................................................................................I - 49
References
RODECO Consulting GmbH
Orientation on environmental quality standards for nutrients and other Danube specific priority substances
I 3
Abbreviations, Acronyms, and Definitions
CCC
The Criterion Continuous Concentration is an estimate of the highest concentration of
a material in surface water to which an aquatic community can be exposed indefinitely
without resulting in an unacceptable effect
CMC
The Criteria Maximum Concentration is an estimate of the highest concentration of a
material in surface water to which an aquatic community can be exposed briefly
without resulting in an unacceptable effect.
COD
Chemical Oxygen Demand
DIN
dissolved inorganic nitrogen (NH4, NO2, NO3)
DRPC
Danube River Protection Convention
Eutrophication The enrichment of water by nitrogen compounds, causing an accelerated growth of
algae and higher forms of plant life to produce an undesirable disturbance to the water
balance of organisms present in the water and to the quality of the water concerned
(91/676/EEC, Article 2).
EQO
Environmental Quality Objective. Policy or management goal to achieve within a
certain period of time. This may be a specific use/function of the water system or any
goal, e.g. 50% reduction in nutrient load within 10 years. An EQO also can be
expressed as a set of numerical standards for each designated use of the water, which
specify the maximum permissible level of pollutants, which must not be exceeded in
the shorter and longer term. The timeframe for achieving an EQO is directly
dependent on analysis of the technical, financial and other implications associated
with the desired improvement in water or sediment quality (Reynolds, 2001).
EQS
Environmental Quality Standard. The concentration of a parameter that should not be
exceeded in the receiving water in order to protect the use of the water. The EQS for
the protection of aquatic life is derived to protect all aquatic species (Reynolds, 2001).
IRC
International Rhine Commission
ISQG
Interim sediment quality guideline
Kd
partition coefficient
N
nitrogen
NH
+
4
NH4 , ammonium
NL
The Netherlands
NO
-
2
NO2 , nitrite
NO
-
3
NO3 , nitrate
NOEL
No Observed Effect Level
Ntot
total nitrogen
P
phosphorous
PEL
Probable effect level
PO
3-
4
PO4 , ortho-phosphate
Ptot
total phosphorous
RAP
Rhine Action Programme
SS
suspended solids
TNMN
Transnational Monitoring Network
US-EPA
United States Environmental Protection Agency
Orientation on environmental quality standards for nutrients and other Danube specific priority substances
I 5
Executive summary
The underlying study aimed at formulating Environmental Quality Standards (EQSs) for those Danube
specific substances that are not included in the list of priority substances of the EU Water Framework
Directive. The Danube specific priority substances comprise: total nitrogen (Ntot), total phosphorous
(P
+
tot), ammonium (NH4 ), chemical oxygen demand (COD), and the metals As, Cr, Cu and Zn.
Given the resources and the time available for the study, it was merely feasible to collect and assess
already existing systems of EQSs. Further, it was feasible to prepare preliminary recommendations for
EQSs at a rather generic and broad scope. The typespecific approach as required by the WFD could
not be implemented.
For the nutrients (Ntot, Ptot) the study used the following working definition in order to make the
WFD's `good status' description for physico-chemical parameters more operational: "nutrient
concentrations such that chances on the occurrence of eutrophication are minimised, or (preferably)
avoided". The query into existing systems for water quality assessment and standards resulted in the
following preliminary recommendations for EQSs to be used as representing `good status' thresholds
for nutrients: Ntot: 1.0 - 1.5 mg N/l; Ptot: 0.02 0.08 mg P/l. Compared to estimated natural background
levels for the Danube river (Ntot ~ 0.8 mg N/l; Ptot ~ 0.03 mg P/l) the proposed EQSs seem rather steep.
The major comment during the presentation of the previous values was that they do not meet with the
requirements of the typespecific approach (which was acknowledged by the study). Following the
typespecific approach, the conditions (including natural background) and requirements of the specific
water body should be assessed and taken into account when setting its corresponding EQS. Therefore,
the figures for both the EQS as well as for natural background mentioned in the report are considered
merely indicative.
For ammonium (NH +
4 ) a separate EQS has been proposed, since ammonium can have toxic effects
under certain conditions and concentration levels. The proposed threshold value representing the
physico-chemical `good status' of NH +
4 is =0.2 mg N/l.
The proposed threshold value representing the physico-chemical `good status' of chemical oxygen
demand is CODMn =10 mg O2/l.
For the metals As, Cr, Cu and Zn it was not possible to extract common denominators from the
existing systems of water quality standards. Firstly, existing systems can differ for the matrices
included in the defined standards (total, dissolved, suspended solids and/or sediment). Secondly,
differences in an order of magnitude of 10 can be observed between comparable water quality
standards, like the `No Observed Effect Level' for zinc applying in the Netherlands (total= 12 µg/l)
versus the one used by the US-EPA (dissolved= 120 µg/l). Since comparing existing systems is not
expected to provide a common ground for reaching consensus, for possible follow-up it has been
proposed to a) `pragmatically' adopt of one existing system of EQSs, or b) to infer EQSs for the
Danube specific metals applying the methodology used by the Fraunhofer Institute for setting the
EQSs for the WFD priority pollutants. As it turned out, Austria already has implemented option b) for
dissolved concentrations of As, Cr, Cu and Zn. The final report is expected to be made public around
the end of the year 2003.
Orientation on environmental quality standards for nutrients and other Danube specific priority substances
I 7
1 Introduction
1.1 Scope
In its original settings, the activities under the tasks 2.1 and 2.2 envisaged to develop (2.1) a proposal
on water quality objectives for nutrients in line with requirements of EU WFD, and (2.2) a proposal on
water quality standards for toxic substances from the DRPC (Danube River Protection Convention)
list of priority substances.
During the 30th MLIM-EG Meeting, it was decided that the present project should focus on
parameters, specific for the DRPC (general parameters: COD, NH4, N, P, and Danube Specific Priority
Substances: As, Cr, Cu, Zn). The remaining Danube priority substances are similar to the WFD
priority substances. The recommendations on water quality standards to be formulated by the Expert
Advisory Forum on Priority Substances are expected to be implemented for the Danube basin as well
and therefore does not require further elaboration here.
During the first Joint MLIM/EMIS Working Group meeting in Vienna (3 February 2003) it was
agreed that the focus of the work would be on the main course of the Danube River. If possible,
recommendations for major trans-boundary tributaries, such as Morava, Tisza, Sava, and Drava,
should be drawn.
1.2 General departure points
With the time and resources available under this proje ct, only existing systems of EQO/EQS could be
taken into consideration as references for formulating recommendations for water quality objectives
and standards for the Danube. Preference was given to systems that were formulated to apply at
regional (international) scales. At least the following EQO/EQS systems are taken into consideration
for the assessments:
· EU-guidelines, notably: 76/464/EEC: on pollution caused by certain dangerous substances
discharged into the aquatic environment of the Community (incl. daughter directives);
78/659/EC: on the quality of fresh waters needing protection or improvement in order to
support fish life; 75/440/EC: concerning the quality required of surface water intended for the
abstraction of drinking water in the Member States;)
· Rhine Action Programme;
· UN "ECE Standards Statistical Classification of Surface Water Quality for the Maintenance of
Aquatic Life;
· other systems where appropriate.
It is presumed, that the approach for deriving the water quality standards in the systems mentioned
above have followed the core features of the procedures as outlined in the WFD (Annex V, 1.2.6
Procedure for the setting of chemical water quality standards by Member States).
1.3 Comments to draft report
The main comments made to the draft final report of August 2003 report are included in chapter 7,
together with brief responses by the author. For proper understanding and perception of the previous
chapters, readers should take the remarks from chapter 7 into account.
I 8
UNDP/GEF Danube Regional Project
2 Nutrients (Ntot, Ptot)
2.1 Nutrients and the WFD
Nutrients are not included in the final list of WFD priority substances as such (decision No
2455/2001/EC). They are mentioned in Annex VIII: Indicative list of the main pollutants, "11.
Substances which contribute to eutrophication (in particular, nitrates and phosphates)". Further,
nutrients are addressed in the definitions of ecological status (Annex V, table 1.2, physico-chemical
quality status):
· High status : nutrient concentrations remain within range normally associated with
undisturbed conditions.
· Good status : nutrient concentrations do not exceed the levels established as to ensure the
functioning of the ecosystem and the achievement of the values specified above (author: this
is a reference to table 1.1 in Annex V) for the biological quality elements.
The WFD puts nutrients dominantly in the perspective of eutrophication. For nitrite (N02), ammonia
(NH3), and ammonium (NH4) also toxic effects on aquatic life forms can apply (without the
occurrence of eutrophication).
Although WFD Annex VIII specifically mentions nitrates, it is common to consider total nitrogen
when dealing with nitrogen-based eutrophication parameters (refer also to textbox 2.1). The priority
substances specific for the Danube include total-nitrogen and ammonium. Therefore, this document
will mainly focus on these two parameters.
Textbox 2.1
Appearances of N and P in riverine systems
Nitrogen compounds occur both dissolved in the water phase, as well as in particulate matter. Generally,
following subdivisions are made:
·
Dissolved inorganic nitrogen: nitrite (NO2), nitrate (NO3), ammonia (NH3), and ammonium
(NH4). NO2 and NH3 are quite instable, and only occur in more extreme conditions (like: low
oxygen levels). Dominating are NO3 and NH4.
·
Dissolved organic nitrogen, like amino-acids, peptides, proteins, etc...
·
Particulate inorganic nitrogen: mainly adsorbed NH4.
·
Particulate organic nitrogen: all kinds of occurrences of N, like in decaying organic plant and
animal material etc...
·
Dissolved gas: N2 or N2O, which can be neglected in the present context.
Total nitrogen (Ntot) in principle encompasses the sum of all the above mentioned occurrences.
(Especially in the former Soviet region it was a practice to label the sum of NO2, NO3, and NH4 as total
nitrogen, which is not correct. Analysing organic/particulate nitrogen was not a tradition at all).
What is available as data partially dep ends on the methods of analysis applies. When for instance
determining nitrogen with the Kjeldahl method, the results comprises both the organic particulate
nitrogen + ammoniacal nitrogen (NH4/NH3). The sum of Kjeldahl-nitrogen+NO2+NO3 then is normally
considered as being total nitrogen.
Phosphorous also occurs in both dissolved and particulate forms. In water quality monitoring, normally
samples are determined for ortho-phosphate and total phosphorous.
·
orthophosphate (PO4) is the major dissolved inorganic form of phosphorous
· total phosphorous (Ptot) includes all occurrences of phosphorous, also PO4, of course assuming
that samples were not filtered prior to analysis.
RODECO Consulting GmbH
Orientation on environmental quality standards for nutrients and other Danube specific priority substances
I 9
2.2 Considerations for formulating EQO and EQS for nutrients in the
Danube
The final aim of the WFD can be considered as one overall Environmental Quality Objective: reaching
(at least) good ecological status before the end of the year 2015.
With regard to the physico-chemical quality elements (including nutrients), conditions are implied
such as they can enable at least good ecological status. The nutrients can be combined with biological
quality elements in terms of phytoplankton, being one of the phenomena that can be related to
eutrophication (like algae blooms). High respectively good status for Phytoplankton are defined as
(Annex V, table 1.2.1):
· High status : The taxonomic composition of phytoplankton corresponds totally or nearly
totally to undisturbed conditions.
The average phytoplankton abundance is wholly consistent with the type-
specific physico-chemical conditions and is not such as to significantly alter
the type-specific transparency conditions.
Planktonic blooms occur at a frequency and intensity, which is consistent with
the type-specific physicochemical conditions.
· Good status : There are slight changes in the composition and abundance of planktonic taxa
compared to the type-specific communities. Such changes do not indicate any
accelerated growth of algae resulting in undesirable disturbances to the
balance of organisms present in the water body or to the physico-chemical
quality of the water or sediment.
A slight increase in the frequency and intensity of the type-specific planktonic
blooms may occur.
As for macrophytes and phytobenthos (also using nutrients as `fertilisers') the differences between
high and good status contain similar wordings (good status: "there are slight changes in the
composition and abundance of macrophyte and phytobentic taxa compared to type-specific
communities. Such changes do not indicate any accelerated growth ...").
A more precise definition and quantification of `slight changes, slight increase, and accelerated
growth' are not considered part of this specific project component (expected to be addressed in the
activities dealing with the issues typology, reference conditions, and ecological classification).
Nevertheless, the above can be translated to formulating an EQO for nutrients as the situation where
nutrient concentrations are such that chances for the occurrence of eutrophication are minimised, or
(preferably) avoided. Two (complementary, but slightly different) angles were followed in this study
to formulate nutrient EQSs.
1. The interpretation of `high status' is a situation with no or only minor anthropogenic
impacts. An inventory of natural background concentrations for nutrients has been
made in order to provide with such perspective when defining the water quality
standards for nutrients.
2. Another part of the inventory aimed at collecting data about nutrient concentrations
that are considered safe in relation to the occurrence of eutrophication.
The value-added of this combined approach is that the discrepancy/similarity between both sets of
results (high and good status approximations) provide additional arguments when defining the EQSs
for nutrients.
As mentioned in the introduction, the main scope of present study is the major course of the Danube
River itself. This can be amended as following.
· Standing waters are more prone to eutrophication compared to running waters. In many
tributaries to the Danube (more-or-less) `standing waters' occur in the form of reservoirs (for
irrigation, drinking water supply or hydropower purposes). Further, the Danube delta
comprises many sections with standing waters (incl. lakes) that are fed by the Danube. In the
queries for data about nutrient levels considered safe in relation to eutrophication, information
on standing freshwaters was included. Applying (basically: more stringent) standing waters'
criteria also to the running parts of the Danube will better safeguard the good status situation.
I 10
UNDP/GEF Danube Regional Project
· The water quality of the northwestern part of the Black Sea is heavily influenced by the inflow
of the Danube. Since the Black Sea is not part of this assignment, it has not been taken into
account yet as additional criterion when formulating the recommended EQSs for nutrients.
Unless mentioned otherwise, figures for water quality standards apply to freshwater. A
narrative in section 2.6 briefly puts the findings of underlying study in the perspective of the
Black Sea.
· Eutrophication in freshwaters is often related to phosphorous being the decisive (limiting)
factor, while for marine waters nitrogen would be more determining. Nevertheless, sufficient
situations exist where the situation is opposite (N being the limiting factor in freshwaters and
P in marine/tidal waters). The queries for EQSs included both nitrogen and phosphorous, to
anticipate different environments, and to keep the perspective on the Black Sea. The search
has been limited to straightforward concentrations. In reality, not only the absolute
concentrations, but also N/P-ratios can be determinative in the actual occurrence of
eutrophication.
2.3 Natural background concentrations of nutrients
As applying to virtually all parameters that originate from natural sources, also for nutrients not one
single natural background concentration in freshwaters exists. Natural background nutrient
concentrations for the Danube are available from calculations with the MONERIS model [Schreiber
et. al., 1993; Behrendt, 2003]. Additional data for other basins were added for the sake of
completeness.
2.3.1 Nitrogen
Table 2.1
Estimates for natural background concentrations of total nitrogen (in [mg N/l])
Average
Min
Max
water(s)
reference
0.8
-
-
Danube
Schreiber, 2003 see
text below
0.75
0.55
1.2
main Danube Behrendt, 2003
tributaries
1.5
-
-
Danube
Adamkovį et. al.,
(TNMN, class I)
2003
0.64
0.27
1.00
river Rhine at Lobith Veldstra, 1989
(Dutch-German
border)
0.6
-
-
Dutch rivers
Breukel, 1993
0.60+(0.0024*SS)X
-
-
rivers in temperate Riet, 1998
zones
1.0
-
-
national
US Dept., 2002
(undeveloped areas)
USA
<1.5
-
-
background rivers in EEIC, 2000
Estonia
X contents of particulate N estimated as 2400 µg/g; multiplying the suspended solids (SS) contents (in mg/l) with a factor
0.0024 provides the particulate N fraction as mg N/l
The average concentration of 0.8 mg N/l for the Danube was calculated as follows. The total natural
background emissions of total nitrogen into the Danube basin estimated by MONERIS is 163 kt/a
[Schreiber, 2003]. Dividing this load by the long-term average downstream flow of the Danube
-6500 m3/s- results in 0.8 mg N/l. Since the load of 163 kt/a represents emissions over the whole
basin, the actual load in the Danube downstream near the delta could be lower, e.g. due to retention.
RODECO Consulting GmbH
Orientation on environmental quality standards for nutrients and other Danube specific priority substances
I 11
Most of the other estimates, ranging between 0.6 1.0 mg N/l, are in line with the value calculated for
the Danube. The interim water quality class I of the Danube TNMN is with 1.5 mg N/l the highest
estimate. TNMN class I is to represent reference conditions or background concentrations
There is seasonality in nutrients concentrations in water systems. Concentrations tend to be lower in
summer period, when for instance assimilation of nutrients by phytoplankton is relatively high.
Therefore, an (annual) average concentration might not properly represent the specific features of
nutrients like nitrogen. This issue will be further elaborated in section 2.5.
Results for other occurrences of nitrogen are summarised in the tables below.
Table 2.2
Estimates for natural background concentrations of dissolved nitrate
(in [mg NO3_N/l])
Average
Min
Max
water(s)
reference
1
-
-
Danube
Adamkovį et. al.,
(TNMN, class I)
2003
0.13
0.05
0.20
river Rhine at Lobith Veldstra, 1989
(Dutch-German
border)
-
0.1
1.0
rivers and lakes
Meybeck, 1989
0.3
-
-
most European rivers EEA, 2001
0.6
-
-
national
US Dept., 2002
(undeveloped areas)
USA
Table 2.3
Estimates for natural background concentrations of dissolved ammonium
(in [mg NH4_N/l])
Average
Min
Max
water(s)
reference
0.2
-
-
Danube
Adamkovį et. al.,
(TNMN, class I)
2003
0.07
0.03
0.10
river Rhine at Lobith Veldstra, 1989
(Dutch-German
border)
0.015
-
-
most European rivers EEA, 2001
0.1
-
-
national
US Dept., 2002
(undeveloped areas)
USA
Table 2.4
Estimates for natural background concentrations of dissolved organic nitrogen
(in [mg N/l])
Average
Min
Max
water(s)
reference
0.01
-
-
Danube
Adamkovį et. al.,
(TNMN, class I)
2003
0.30
0.12
0.50
river Rhine at Lobith Veldstra, 1989
(Dutch-German
border)
I 12
UNDP/GEF Danube Regional Project
Table 2.5
Estimates for natural background concentrations of particulate organic nitrogen
(in [mg N/l])
Average
Min
Max
water(s)
reference
0.14*
0.07
0.20
river Rhine at Lobith Veldstra, 1989
(Dutch-German
border)
* based on suspended solids concentration of 30 mg/l, with 10% organic matter
2.3.2 Phosphorous
Table 2.6
Estimates for natural background concentrations of total phosphorous (in [mg
P/l])
Average
Min
Max
water(s)
reference
0.028
-
-
Danube
Schreiber, 2003
see text below
0.027
0.021
0.037
main Danube Behrendt, 2003
tributaries
0.1
-
-
Danube
Adamkovį, 2003
(TNMN class I)
0.11
0.023
0.19
river Rhine at Lobith Veldstra, 1989
(Dutch-German
border)
0.06
-
-
Dutch rivers
Breukel, 1993
0.011+(0.00115*SS)X
-
-
rivers in temperate Riet, 1998
zones
-
0.0
0.05
various catchments
EEA, 2001
0.10
-
-
national mean of Pope, 2002
streams in USA
0.10
-
-
national
US Dept., 2002
(undeveloped areas)
USA
0.05
-
-
natural rivers in EEIC, 2000
Estonia
X contents of particulate P estimated as 1150 µg/g; multiplying the suspended solids (SS) contents (in mg/l) with a factor
0.00115 provides the particulate P fraction as mg P/l
The average Danube concentration of 0.028 mg P/l was derived from a natural background of 5.8 kt/a
as calculated by MONERIS, divided by a flow of 6500 m3/s.
The natural background concentrations of Ptot calculated for the Danube based on MONERIS are the
lowest compared to the estimates for other rivers: 0.03 mg P/l. The remaining data in table 2.6 range
between 0.05 - 0.10 mg P/l, except for the river Rhine with an estimated maximum of 0.19 mg P/l.
Data about other appearances of phosphorous are included in the following two tables.
RODECO Consulting GmbH
Orientation on environmental quality standards for nutrients and other Danube specific priority substances
I 13
Table 2.7
Estimates for natural background concentrations of ortho-phosphate
(in [mg PO4_P/l])
Average
Min
Max
water(s)
reference
0.05
-
-
Danube
Adamkovį, 2003
(TNMN, Class I)
0.05
0.003
0.10
river Rhine at Lobith Veldstra, 1989
(Dutch-German border)
Table 2.8
Estimates for natural background concentrations particulate bound
phosphorous
(in [mg P/l])
Average
Min
Max
water(s)
reference
0.06*
0.02
0.09
river Rhine at Lobith Veldstra, 1989
(Dutch-German border)
* based on suspended solids concentration of 30 mg/l, with 10% organic matter
2.3.3 Historic data on nutrients in the Danube
According to Kroiss "... in 1961 the nutrient discharge was close to `natural' conditions as there were
no adverse effects reported from Black Sea" [Kroiss, 2002]. This implies that, measurement data
around the year 1960 could serve as an adequate reference for (more or less) natural background
concentrations.
Loads discharged into the Black Sea in 1961 are estimated as 140 kiloton DIN (dissolved inorganic
nitrogen) and 12 kt PO4-P [Kroiss, 2002]. Following the method used above (dividing the load by a
flow of 6500 m3/s), these loads would imply the following average concentrations:
· DIN: 0.68 mg N/l
· PO4_P: 0.063 mg PO4_P/l
In 1960, the DIN load at Sulina was estimated to 140 kton [NIMRD]. The Ptot load at Reni was in 1960
13 kt [VITUKI, 1997]. These loads would result in the following average concentrations:
· DIN: 0.68 mg N/l
· Ptot: 0.063 mg P/l
The loads mentioned above are higher than the natural background emissions calculated by
MONERIS.
According to MONERIS the natural background loading for Ntot is 163 kt/a. Since the above DIN
loads (140 kton) do not include organic nitrogen, the accompanying total nitrogen loads can be
assumed to be higher than the MONERIS estimate.
The 1961 phosphorous load (12 kton) comprises PO4 only, hence the Ptot load will be higher. The 1960
Ptot load at Reni (13 kt) is two times higher than the total natural P-emissions calculated by
MONERIS.
The notice of Kroiss ("close to `natural' conditions as there were no adverse effects reported from
Black Sea") can be considered a working definition of a WFD `good status'. As within the WFD itself,
it is still arbitrary how large `close' would be.
Collecting and analysing individual measurement data around the 1960-ies could provide additional
support in defining the nutrient EQSs for the Danube. One advantage of collecting measurement data
is that seasonality phenomena better can be estimated, and data themselves are actual concentrations
I 14
UNDP/GEF Danube Regional Project
(and not approximates from loads). Main disadvantages of this approach are that it could be an
elaborate process to retrieve the data from paper archives, there will be missing data (notably organic
nitrogen), and the comparability of analyses/analytical results needs to be crosschecked.
2.4 Nutrients levels avoiding risks for eutrophication (`good status')
In section 2.3 it is proposed to use as (limits for) natural background concentrations of nutrients in the
Danube basin: Ntot: =1.0 mg N/l, and Ptot: =0.10 mg P/l. Concentrations at background levels are
regarded to represent a `high status'. The WFD does not impose high status, but allows for at least
`good status'. Under good status, nutrient concentrations may deviate from an undisturbed situation, as
long as the functioning of the ecosystem is ensured and the values specified for the biological quality
elements can be achieved. As argued in section 2.2, eutrophication is considered the major
phenomenon occurring when nutrient levels are too high. From this, it can be postulated that nutrient
concentrations not exceeding levels associated with causing eutrophication can be regarded as good
status levels. The tables below contain the results of literature and Internet queries on this topic.
2.4.1 Nitrogen
2.4.1.1 United Nations
The UN "ECE Standards Statistical Classification of Surface Water Quality for the Maintenance of
Aquatic Life" defines five classes in relation to eutrophication (for the parameters Ntot, Ptot and
chlorophyll) [UN/ECE, 1992]. As major criteria apply "Trophic state and best available expert
judgement regarding the impact of trophic state on aquatic life, maintaining consistency between the
three variables".
The concentrations for Ntot are shown below.
UN-ECE
class I1
class II
class III
class IV
class V
Ntot
Oligotrophic
Mesotrophic
moderately
strongly
extensively
Eutrophic
Eutrophic
polluted
[mg N/l]
<0.3
0.30 - 0.75
0.75 - 1.50
1.50 2.50
>2.50
2.4.1.2 United States
Quoting part of the opening EPA page at http://www.epa.gov/ost/standards/nutrient.html "The
United States Environmental Protection Agency (EPA) is publishing recommended water quality
criteria to reduce problems associated with excess nutrients in waterbodies in specific areas of the
country. EPA will work with states and tribes to adopt regional-specific and locally appropriate water
quality criteria for nutrients in lakes, reservoirs, rivers, streams, and wetlands in seventeen
ecoregions." Interesting in their approach is the use of "Ecoregional Nutrient Criteria". In total
seventeen ecoregions are recognised within the United States. Different water quality criteria can
1 UN/ECE eutrophication classes:
I
Clear, oligotrophic water with, at most, a very slight, occasional anthropogenic pollution with organic matter. Low nutrient
content, providing spawning ground for salmonoids
II
Slightly polluted, mesotrophic water receiving small discharges of organic matter. The loadings may lead to slightly increased
primary productivity.
III
Moderately euthrophic water receiving considerable amounts of discharges of organic matter and nutrients. The level of primary
production is considerable, and some changes in the community structure, including fish species, can be observed
IV
Strongly eutrophic, polluted water, receiving discharges of organic matter, nutrients and harmful substances. Algal blooms are
common. Increased decomposition of organic matter together with stratification of water bodies may entail anaerobic conditions
and fish kills. Mass occurrences of more tolerant species; population of fish and benthic organisms are affected.
V
Extensively polluted, hyperthropic water. Decomposers dominate over producers. Fish or benthic species do not occur
permanently.
RODECO Consulting GmbH
Orientation on environmental quality standards for nutrients and other Danube specific priority substances
I 15
apply to different regions. This approach/philosophy resembles the approach of the WFD that also
acknowledges ecoregions. (No attempt has been made to identify similarities in the ecoregions defined
by the EPA, and the Danube basin, going beyond the scope of this project, and the present state of
information concerning the typology setting for the Danube.) The overall strategy "to reduce
overenrichment in surface waters" and the accompanying water quality criteria can be considered
equivalent to "avoiding eutrophication".
The following boundaries for trophic cla ssification are suggested:
(http://www.epa.gov/waterscience/criteria/nutrient/guidance/rivers/chapter_2.pdf)
US EPA
Oligotrophic-
Mesotrophic-
mesotrophic
eutrophic
N
boundary
boundary
tot
[mg N/l]
0.7
1.5
The recommended nutrient criteria for the different ecoregions range between the following values
(http://www.epa.gov/waterscience/criteria/nutrient/ecoregions/sumtable.pdf)
US EPA
nutrient criteria
nutrient criteria
Lakes and reservoirs Rivers and Streams
Ntot [mg N/l] (12 ecoregions)
(13 ecoregions)
minimum
0.10
0.12
maximum
1.27
2.18
average
0.51
0.67
median
0.45
0.56
2.4.1.3 Sweden
The Swedish EPA uses the following criteria for assessment of the trophic state in lakes [SEPA,
2002]:
Sweden
Oligotrophic Mesotrophic
Eutrophic
Eutrophic
Hypertrophic
(level Low)
(level Moderately (level High)
(level Very High) (level Extremely high)
N
high)
tot [mg N/l]
average May-Oct <0.300
0.300 - 0.625
0.625 1.25
1.25 5
>5
2.4.1.4 The Netherlands
The Dutch system of water quality standards generally distinguishes two concentration levels:
· A "landelijke streefwaarde" (national target va lue), which can be considered equivalent to a
No Observed Effect Level (NOEL). Water quality meeting these NOELs are objectives of the
medium to long-term policy strategy.
· A "Maximaal Toelaatbaar Risico" (maximum allowable risk), equivalent to MAC values.
These standards are mandatory for current water policy and -management; exceeding requires
immediate remedial actions.
I 16
UNDP/GEF Danube Regional Project
The Netherlands
NOEL
MAC
Ntot
[mg N/l] [mg N/l]
standing waters: average of concentrations during summer (Apr-Sep) 1.0
2.2
running waters: annual mean concentration
1.0
2.2
2.4.1.5 TNMN
An interim water quality classification scheme had been approved by the Monitoring, Laboratory and
Information Management Sub-Group in 2001. The interim class II represents the target values.
Danube, TNMN
[mg N/l]
Ntot
Class II (interim target value) 4.0
The Class II value of 4 mg N/l is considerably higher than the values mentioned in the previous
sections.
2.4.1.6 Synthesis of eutrophication thresholds for Ntot
Refining the descriptions in the section 2.2 and the introduction of 2.4, it is suggested that the
approximation of a `good status' situation in the context of the eutrophication issue could be set equal
to "Mesotrophic level or better". The concentrations associated with the boundary Mesotrophic
Eutrophic in the data above range from 0.6 (0.625 Sweden) through 0.8 (UN/ECE) to 1.5 (US EPA)
mg N/l. The recommended US EPA nutrient criteria for Ntot (average 0.6 mg N/l for lakes and
reservoirs, and 0.7 mg N/l for rivers and streams) imply that the (US EPA) Oligotrophic -Mesotrophic
boundary prevails. (It is not immediately clear whether the US-EPA nutrient criteria could be
considered more equivalent to a `high' status situation.) Nevertheless, setting 1.5 mg N/l as the
ultimate limit for the `good status' total nitrogen concentration seems reasonable with the above data.
The UN-ECE and Swedish trophic level boundaries indicate that 1.0 mg N/l can be considered a safer
threshold.
Combining these various findings, for total nitrogen in the (freshwater part of the) Danube basin an
EQS in the range 1.0 1.5 mg N/l is recommended. Input from other tasks, Danubian experts, and
additional activities will be needed to fine-tune the recommended value (beyond the lifetime of this
project) towards just one concentration. Part of the fine-tuning is also to comprise defining additional
criteria, like whether the EQS should be an annual, or a summer average concentration (see also
section 2.5 below).
Taking the Dutch 2.2 mg N/l as MAC into consideration, then it seems prudent to qualify Ntot
concentrations =2 mg N/l as `moderate' (or worse) physico-chemical status.
The TNMN Class II target value (4 mg N/l) is an outlier in the series. Systems like those from the
UN/ECE, US EPA, or Sweden would associate such a concentration with eutrophic waters. This does
not imply that the Class II value not would be appropriate. It merely shows that during the follow-up
activities (beyond the lifetime of the underlying project) the arguments and findings of earlier
discussions are to be considered as well.
RODECO Consulting GmbH
Orientation on environmental quality standards for nutrients and other Danube specific priority substances
I 17
2.4.2 Phosphorous
2.4.2.1 Overview of different water quality/eutrophication criteria for phosphorous
For background information about the various water quality standards/criteria mentioned in the tables
below, refer the previous subsection, except for the following.
· The OECD has defined boundary value criteria for trophic categories. They apply to temperate
region lakes and reservoirs. The figures shown further below were quoted from [Meybeck,
1989].
· The UK Environment Agency has defined interim targets for phosphorous in fresh waters
[UK/EA, 2001].
· EU. The Council Directive 78/659/EEC "on the quality of fresh waters needing protection or
improvement in order to support fish life" mentions no concentration figures for Ptot. But, the
table in Annex I of this directive mentions under the header Observations "In other cases limit
values of 0.2 mg PO4/l (author: corresponding to 0.065 mg PO4_P/l) in salmonid waters and
of 0.4 mg PO4/l (author: corresponding to 0.13 mg PO4_P/l) in cyprinid waters may be
regarded as indicative in order to reduce eutrophication."
· The Rhine Action Programme has defined 0.15 mg P/l as the target-value for Ptot [IKSR,
1992]. This target value aims at reduction of algae growths.
UN/ECE, 1992
class I
class II
class III
class IV
class V
Oligotrophic
Mesotrophic
moderately
strongly
extensively
P
eutrophic
eutrophic
polluted
tot [mg P/l]
standing water
<0.01
0.010 0.025
0.025 0.050
0.050 0.125
>0.125
running water
<0.015
0.015 0.040
0.040 0.075
0.075 0.190
>0.190
US EPA
Oligotrophic-
Mesotrophic-
mesotrophic
eutrophic
P
boundary
boundary
tot
[mg P/l]
0.025
0.075
US EPA
nutrient criteria
nutrient criteria
Lakes and Reservoirs Rivers and Streams
Ptot [mg P/l] (12 ecoregions)
(13 ecoregions)
minimum
0.008
0.010
maximum
0.038
0.128
average
0.017
0.041
median
0.016
0.033
OECD
Ultra-
Oligotrophic
Mesotrophic
Eutrophic
Hypertrophic
P
logographic
tot
[mg P/l]
0.004
0.01
0.01-0.035
0.035-0.1
0.1
I 18
UNDP/GEF Danube Regional Project
Rhine
IRC target-value
Ptot
(`Zielvorgabe')
[mg P/l] - average 0.15
Sweden
Oligotrophic Mesotrophic
Eutrophic
Eutrophic
Hypertrophic
(level Low)
(level Moderately (level High)
(level Very High) (level Extremely high)
P
high)
tot [mg P/l]
average May-Oct <0.0125
0.0125 0.0250
0.025-0.050
0.050-0.100
>0.100
The Netherlands
NOEL
MAC
Ptot
[mg P/l] [mg P/l]
standing waters: average of concentrations during summer (Apr-Sep) 0.05
0.15
running waters: annual mean concentration
0.05
0.15
Danube, TNMN
Ptot
[mg P/l]
Class II (interim target value) 0.1
United Kingdom
Oligotrophic Mesotrophic
Meso-
Eutrophic
P
Eutrophic
tot [mg P/l]
standing waters (annual geometric mean) 0.008
0.025
-
0.085
running waters (annual mean)
0.020
0.060
0.100
0.200
EU 78/659/EEC salmonid waters cyprinid waters
PO4
[mg P/l]
0.065
0.13
2.4.2.2 Synthesis of eutrophication thresholds for Ptot
Following the approach for Ntot, the boundary Mesotrophic-Eutrophic for standing waters- for Ptot
ranges from 0.025 (UN-ECE, Sweden) through 0.035 (OECD) to 0.075 (US-EPA) mg P/l. As was the
case with Ntot, also the US-EPA nutrient criteria for Ptot are lower than this boundary (average 0.02 mg
P/l for lakes and reservoirs, and 0.04 mg P/l for rivers and streams).
The findings suggest recommending the EQS for Ptot in the Danube to 0.02-0.08 mg P/l. Excluding the
US-EPA, the data gear towards a range of 0.02 0.05 mg P/l. Both ranges completely fit within the
range of natural background concentrations as suggested in subsection 2.3.2!
A shared feature in most of the above systems is that Ptot concentrations =0.1 mg P/l are associated
with strongly eutrophic and worse states. Implying that a freshwater water quality with concentrations
of 0.1 mg P/l or more anyway cannot qualify as `good status' waters. The major exception is the
target-value defined for the Rhine Action Programme (0.15 mg P/l). The TNMN Class II target value
(0.1 mg P/l) is also relatively high.
RODECO Consulting GmbH
Orientation on environmental quality standards for nutrients and other Danube specific priority substances
I 19
2.5 Comparison of proposed EQSs with actual concentrations
The above-recommended EQSs are put into perspective by comparing them with actual measurements
in the Danube. The results of the Joint Danube Survey (JDS) are elaborated in this report. Another
output of the underlying project consists of an assessment of 5 years of TNMN data in the report
entitled "Five-years Report on Water Quality in Danube River Basin Based on TransNational
Monitoring Network - 1996-2000" [Adamkovį, 2003]. Readers are referred to this report for more
details.
2.5.1 Seasonality
Before going into further details it is considered appropriate first to outline some features that are
rather typical for nutrients. Generally, during summer period nutrient concentrations, notably nitrogen
compounds, tend to be lower because of the assimilation and fixation by phytoplankton. An example
is shown in the graph below.
Figure 2.1
Nitrate concentrations at Reni (left bank) 1996-2000 [mg NO3_N/l]
.
.
.
.
.
Jan-
Jul -
Jan-
Jul-
Jan-
Jul-
Dec-
Jun-
Dec-
Jun-
Dec-
The implications of such seasonal variations are following:
· Eutrophication normally occurs in warmer periods, tentatively in the period May September.
If the EQSs are formulated in relation to (avoiding) eutrophication, then basically it would be
important to set the standards for the period most prone to eutrophication. In the case of
nitrogen this actually implies that one could allow for lower concentrations for the EQS.
Eutrophication is less likely to occur during the colder season (November March) during
which N-concentrations tend to be higher!
· Status assessments comprising annual mean concentrations basically are not sufficient. Season
(averaged) concentrations would prevail.
2.5.2 Joint Danube Survey
The results of the JDS are exhaustively elaborated and discussed in the technical report [ICPDR,
2002]. Readers are referred to this report for more details. This section mainly contains some
highlights, together with a few additional assessments.
2.5.2.1 Total nitrogen
Many of the JDS data are unprecedented, including the systematic sampling and analysis for organic
nitrogen along the whole stretch of the Danube river. Data for organic and total nitrogen in the Danube
I 20
UNDP/GEF Danube Regional Project
basin are still scant, also in the 1996-2000 TNMN database. The JDS took place in the warmer
summer-autumn period. Hence, the nutrient/nitrogen concentrations presumably are reflecting the
lower ranges over the year 2002.
The JDS Technical report does not contain an assessment of total nitrogen. These figures (based upon
data retrieved from the ICPDR Danubis website) have been generated for this reported in the
following way. The water samples were analysed for organic nitrogen with application of Kjeldahl
method ammonium analysis by spectrophotometric method [ICPDR, 2002, subsection 5.1.3]. The
Kjeldahl method results in the sum of organic plus (the inorganic) ammonium, NH4, nitrogen. Hence,
the total nitrogen concentrations were calculated as the sum of organic nitrogen + NO2 + NO3. The
`pure' organic nitrogen concentrations were calculated by subtracting the (separately reported) NH4-
concentrations from the organic -N concentrations for each sampling site.
Since the JDS took place in an `eutrophication-sensitive' period, the concentrations can be considered
to be `low-year values'.
The total nitrogen figures for the main part of the river are shown in the graph below.
Figure 2.2
Total nitrogen concentrations in the Danube's main course, JDS survey [mg N/l]
River km (distance to Black Sea)
The average concentration along the main course of the Danube was 1.9 mg N/l. The maximum of the
recommended EQS range (1.0 - =1.5 mg N/l) was exceeded in 75% of the occasions (43 out of 57
sampling locations marked as `Danube'). The total nitrogen concentrations of the samples taken at
(near the mouth of) the tributaries and/or Danube arms (code `Tributary/arm') exceed the proposed
maximum of the EQSs range in 19 out of 25 (76%) of the cases. Not surprisingly, the concentrations
in the tributaries tend to be (slightly) higher than in the main course of the Danube River.
RODECO Consulting GmbH
Orientation on environmental quality standards for nutrients and other Danube specific priority substances
I 21
Table 2.9
Summary statistics of Ntot in Joint Danube Survey [mg N/l]
Main Danube course Tributaries/arms
average
1.9
2.4
median
1.9
2.1
5-percentile
1.0
1.1
95-percentile
2.9
4.5
minimum
1.0
0.9
maximum
6.6
7.0
2.5.2.2 Organic nitrogen
When subtracting the NH4 concentrations from the (Kjeldahl determined) organic nitrogen contents,
the results are as follows.
Figure 2.3
`Plain' organic nitrogen concentrations in the Danube's main course, JDS survey
[mg N/l]
River km (distance to Black Sea)
Table 2.10
Summary statistics of `plain' organic N in Joint Danube Survey [mg N/l]
Main Danube course Tributaries/arms
average
0.85
0.98
median
1.28
0.90
5-percentile
0.30
0.49
95-percentile
1.36
1.66
minimum
0.10
0.44
maximum
5.10
2.21
On average, the organic_N concentrations comprise about 40% of the total_N concentrations.
I 22
UNDP/GEF Danube Regional Project
2.5.2.3 Total phosphorous
Since Ptot was directly analysed in the water samples, no intermediate calcula tions were required. The
graph below shows the Ptot concentrations along the main course of the river. The table includes the
summary statistics for the main course of the Danube River and for the tributaries.
Figure 2.4
Ptot concentrations in the Danube's main course, JDS survey [mg P/l]
.
.
.
.
.
.
.
River km (distance to Black Sea)
Table 2.11
Summary statistics of Ptot in Joint Danube Survey [mg P/l]
Main Danube course Tributaries/arms
average
0.12
0.23
median
0.10
0.16
5-percentile
0.07
0.08
95-percentile
0.23
0.67
minimum
0.06
0.08
maximum
0.59
0.92
Compared to the proposed EQS-range (0.02 0.08 mg P/l), most of the JDS samples would exceed the
maximum of 0.08 mg P/l.
The graph below contains the annual mean Ptot concentrations of the TNMN stations along the main
course of the Danube for the year 2000.
RODECO Consulting GmbH
Orientation on environmental quality standards for nutrients and other Danube specific priority substances
I 23
Figure 2.5
Annual mean Ptot concentrations in the Danube, TNMN data 2000 [mg P/l]
.
.
.
.
.
.
River km (distance to Black Sea)
The annual mean concentrations tend to be slightly higher than the JDS results. More details on the
results of the TNMN can be found in [Adamkovį, 2003].
2.5.2.4 `Eutrophication'
One should keep in mind that the suggested Ntot and Ptot thresholds are merely physico-chemical
approximates for the occurrence of eutrophication. Finally, the occurrence of the more distinctive
features of eutrophication, like algae blooms, chlorophyll-a concentrations, very low/either
oversaturated oxygen levels, etcetera, are better indicators.
Section 4.5: Phytoplankton of the JDS technical report indeed mentions the occurrences of
eutrophication and/or eutrophic states in various instances [ICPDR, 2002]. Linking these observations
with the measured nutrient concentrations in the main course of the Danube (5/95-percentile
concentrations were: Ntot: 0.3 1.9 mg N/l; Ptot: 0.07 - 0.23 mg P/l), one can infer that the proposed
`good status' EQS for Ntot (1.0 1.5 mg N/l) is supported, while the proposed EQS for Ptot (0.02 0.08
mg P/l) may be considered too stringent.
2.6 Synthesis and discussion of previous findings
While trying to formulate recommendations for EQSs for nutrients in the Danube River basin, the
underlying study at the same time aimed at providing with (background) material to support the
MLIM and other expert groups in the completion of their tasks.
The recommended EQSs are:
· 1.0 1.5 mg N/l for Ntot
· 0.02 0.08 mg P/l for Ptot
I 24
UNDP/GEF Danube Regional Project
The proposed values first of all are suggested to become the focus of future discussions. Among
others, further agreements finally should be reached on:
· the proposed ranges as such;
· the upper/lower/medium of the ranges (if agreed as such);
· setting the EQSs as `season variable' values (e.g. just summer-average means);
· checking and testing the physico-chemical values against ecological/biological criteria and
empirical findings.
The proposed EQSs seem rather stringent. Nevertheless, they were formulated from a common
denominator that can be recognised in the `best-available-knowledge' contained in a representative
series of references.
When applying them for instance to the results of the JDS, then the conclusion would be that the
physico-chemical situation would not comply with "good status" for nutrients. The EQSs could
partially be supported by yet other sets of data more specific for the Danube, notably the Joint Danube
Survey findings. The biological results of the JDS indicate occurrences of eutrophication, with nutrient
concentrations not that much exceeding the recommended EQSs.
The Black Sea has not could be taken into account in the assessments of the underlying study.
Acknowledging the fact that the discharge of the Danube has an significant impact on the status of the
Black Sea (at least in the north-western part), implies that, while formulating water quality criteria for
the freshwater part of the Danube basin, the final resulting water quality of the Black Sea should be
taken into consideration as well. Additional considerations, when taking the Black Sea into account as
well, include:
· Discharged loads as a criterion; this not necessarily conflicts with the approach of formulating
EQSs, but may result in yet other viewpoints. The concept of "critical loads" is still under
development.
· Seasonality criteria: for loading of the Black Sea with nutrients it finally may not matter
whether they are discharged in summer of wintertime, since they will be retained in the sea
anyway. This could conflict with the option to `optimise' EQSs in the freshwater Danube
basin for the summer period.
RODECO Consulting GmbH
Orientation on environmental quality standards for nutrients and other Danube specific priority substances
I 25
3 Ammonium (NH4)
3.1 Introduction
Ammonium2 is one of the nutrients that can contribute to eutrophication. However, NH4 is also related
with toxic impacts, notably the toxicity of (ammonia) NH3. Ammonia is instable, and in freshwaters
only will occur in larger concentrations under more extreme conditions (relevant parameters are water
temperature and pH; low oxygen levels also can be relevant). Because of its potential toxicity, it has
been decided to deal separately with NH4 in this chapter.
In this context, ammonium can be regarded as a `specific non-synthetic pollutant'. The WFD defines
the status of these specific non-synthetic pollutants as follows (annex V, table 1.2)
· High status : Concentrations remain within the range normally associated with undisturbed
conditions (background levels = bgl).
· Good status : Concentrations not in excess of the standards set in accordance with the
procedure detailed in section 1.2.6 (2) without prejudice to Directive
91/414/EC and Directive 98/8/EC. (<EQS).3
3.2 Inventory of EQSs for NH4
Compared to the nutrients (chapter 2) and the metals (chapter 5), the available information for EQS for
ammonium will be presented rather straightforward. The major information is included in table 3.1.
Table 3.1
Overview of EQSs for ammonium
Water quality Concentration
Concentration
Remarks
Reference
criteria system
(lower range)
(higher range)
[mg NH4_N/l]
[mg NH4_N/l]
Rhine
0.2
Target value (Zielvorgabe)
IKSR, 1992
United States
3.8
20 (salmonids present)
see note (1) below
EPA, 1999
30 (salmonids absent)
EU
=0.04 salmonid
=1 salmonid waters
low= G(uide) value
Directive
=0.2 cyprinid
=1 cyprinid waters
higher= I(mperative)
78/559/EEC
UN-ECE
n.a.
The document just mentions UN-ECE,
NH3 without any values
1992
TNMN
0.2 Class I
0.6 - >1.5 (Class III V)
Adamkovį,
0.3 Class II
2003
Czech Republic 0.2 Class I
0.4 Class II
I: very clean
Haskoning,
5 Class IV
II: clean
1994
>5 Class V
IV: intensely polluted
V: very polluted
Hungary
0.8 Class I
1.9 Class II
I: high quality, clean waters
Haskoning,
2 This report will use the name ammonium for NH4. Sometimes, NH4 is called "total ammonia", where NH3 then is called "un-ionised"
ammonia.
3 Section 1.2.6 of WFD Annex V outlines the procedure for the setting of chemical quality standards by Member States. Directive
91/414/EC is the COUNCIL DIRECTIVE of 15 July 1991 concerning the placing of plant protection products on the market. 98/8/EC refers
to the Directive 98/8/EC of the European Parliament and of the Council of 16 February 1998 concerning the placing of biocidal products on
the market.
I 26
UNDP/GEF Danube Regional Project
Water quality Concentration
Concentration
Remarks
Reference
criteria system
(lower range)
(higher range)
[mg NH4_N/l]
[mg NH4_N/l]
>1.9 Class III
II: polluted; no detrimental 1994
impact on aquatic
ecosystem
III: polluted, detrimental to
aquatic ecosystem
Romania
0.1 Class I
0.2 class II
I: drinking water
Haskoning,
0.4 class III
II: recreation and fishing
1994
III: irrigation and industry
The
-
0.02 mg NH3_N/l (MAC)
no values for ammonium
V&W, 2000
Netherlands
(1)
The US-EPA actually mentions formulas as criteria for total ammonia, where the result depends on the water temperature and pH.
In the above table as `low range' concentration the CCC (chronic criterion, see also chapter 5) for pH=7.2 and T= 20 oC is
mentioned [EPA, 1999, page 87]. As high range, the CMC (acute criterion) for pH= 7.2 is mentioned [EPA, 1999, page 86].
3.3 Synthesis of findings for `good status' of NH4
A natural background concentration of ammonium will be less than 0.1 mg NH4_N/l (compare Table
2.3). Hence, as EQS for `high' status of ammonium, a concentration of 0.1 mg NH4_N/l is suggested.
The US-EPA concentrations are exceptionally high compared to the other criteria. The lowest CMC
(Criteria Maximum Concentration is an estimate of the highest concentration of a material in surface
water to which an aquatic community can be exposed briefly without resulting in an unacceptable
effect) concentration in [EPA, 1999, table page 86] is 0.9 mg N/l, applying to water with pH=9 and
salmonids present.
The remaining concentrations mentioned in the `lower range' column tend towards 0.2 mg NH4_N/l or
less (except Hungary Class I). Based upon this overview it is proposed to use =0.2 mg NH4_N/l as the
`good status' EQS for ammonium.
RODECO Consulting GmbH
Orientation on environmental quality standards for nutrients and other Danube specific priority substances
I 27
3.4 Comparison of proposed EQS with actual concentrations
3.4.1 Joint Danube Survey
Figure 3.1
NH4 concentrations in the Danube's main course, JDS survey [mg N/l]
.
.
.
.
.
.
.
.
.
.
River km (distance to Black Sea)
Table 3.2
Summary statistics of NH4 in Joint Danube Survey [mg N/l]
Main Danube course Tributaries/arms
average
0.05
0.23
median
0.02
0.03
5-percentile
0.01
0.01
95-percentile
0.18
0.78
minimum
0.01
0.01
maximum
0.19
3.24
The JDS findings would comply with 0.2 mg NH4_N/l as "good status" EQS for ammonium. (Because
of one extreme outlier, the average tributaries/arms concentration is higher than the median
concentration; also the 95%-concentration seems relatively high).
3.4.2 TNMN
Since the JDS was conducted during the warm August-October period, one may expect relatively low
concentrations. The graph below contains the annual mean concentrations of the TNMN stations along
the main course of the river for the year 20004.
4 Please refer for details to the separately reported TNMN 1995-2000 results in [Adamkovį, 2003].
I 28
UNDP/GEF Danube Regional Project
Figure 3.2
NH4 concentrations in the Danube, TNMN stations, annual mean 2000 [mg N/l]
.
.
.
.
.
.
.
.
River km (distance to Black Sea)
Indeed this shows quite a different picture compared to the JDS results. Similar patterns
(concentrations levels and higher concentrations in the downstream half of the river) can be recognised
for the years 1996-1999. There are many locations where the annual mean concentration exceeds the
proposed EQS of 0.2 mg N/l (and higher concentrations can be expected during the winter period).
3.4.3 Seasonality
The argument that could be applied to eutrophication (an EQS could refer to a summer, low
concentration, period only) is not valid from a toxicity point of view. It is the actual concentration that
might be harmful to organisms. In fact, one should apply the potential winter maximum conditions
when formulating the EQS for NH4.
RODECO Consulting GmbH
Orientation on environmental quality standards for nutrients and other Danube specific priority substances
I 29
4 Chemical oxygen demand (COD)
4.1 Introduction
COD is a bit of an exception in the parameters dealt with in this report. The other parameters are
distinct compounds, whereas COD is merely an indicative `sum' parameter. COD can be defined as:
· (1) A measure of the chemically oxidizable material in the water, which provides an approximation of
the amount of organic and reducing material present. The determined value may correlate with
Biochemical Oxygen Demand (BOD) or with carbonaceous organic pollution from sewage or industrial
wastes; (2) A chemical measure of the amount of organic substances in water or wastewater. A strong
oxidizing agent together with acid and heat are used to oxidize all carbon compounds in a water sample.
Nonbiodegradable and recalcitrant (slowly degrading) compounds, which are not detected by the test
for BOD, are included in the analysis. The actual measurement involves a determination of the amount
of oxidizing agent (typically, potassium dichromate) that is reduced during the reaction [NDWR, 1999].
· The mass concentration of oxygen consumed by the chemical breakdown of organic and inorganic
matter [UN/ECE, 1992].
The following quote shows how COD can be used and interpreted in an environmental context (note:
the situation described in the quotation applies to Japan).
"Chemical Oxygen Demand: CODMn is used as an organic pollution index including phytoplankton
growth. A COD of less than 1 mg/l is assumed not to be caused by anthropogenic influence. Waters
under this condition are suitable for conservation of the natural environment. According to the drinking
water law, the standard value for KmnO4 consumption is 10 mg/l, which is equivalent to 2.5 mg/l of
COD. A survey, conducted by the Ministry of Health and Welfare, found that most lakes being used for
drinking water supply have a COD of less than 3 mg/ l. Water quality for fisheries were classified as
either oligotrophic or eutrophic. In oligotrophic lakes, having very clear water, COD should be less than
1 mg/l that is required for oligosaprobic species such as rainbow trout. In general, the COD of
oligotrophic and eutrophic lakes containing oligosaprobic fish such as smelt, should be less than 3 mg/l.
In eutrophic lakes containing carp, the COD should be less than 5 mg/l (Water Quality Standards for
Fisheries, 1965). Less than 8 mg/l COD is desirable for waters used for swimming. High COD
interferes with oxygen transfer to the soil, resulting death of rice plants. Experimental results show that
a COD of less than 6 mg/l are desirable for agriculture use. In general 8 mg/l of COD is acceptable for
most industrial uses and for conservation of environment" [EMECS, 2003].
There are two methods for analysing COD, using the dichromate method (results then should be
indicated as CODCr), or the permanganate method (CODMn). Analysis of the same water sample with
CODCr results in higher (factor 2 to 3) concentrations than with CODMn.
4.2 Inventory of EQSs for COD
Compared to the other parameters, relatively few information on EQS systems could be identified.
From the major used reference systems (EU, UN, Rhine, US-EPA) only the UN-ECE presents EQSs
for COD. The only EU Directive mentioning a distinct value for COD is the Council Directive
75/440/EEC (30 mg O2/l as Guide value for the A3 drinking water treatment category).
I 30
UNDP/GEF Danube Regional Project
4.2.1 United Nations
The UN-ECE groups COD under the Oxygen regime category. Major criteria for this category are
oxygen content, together with presence of oxygen-demanding substances, and the impact of oxygen
content levels on aquatic life.
UN-ECE
class I5 class II class III class IV class V
CODMn (mg O2/l] <3
3 10
10 20
20 -30
>30
4.2.2 Danube basin
Table 4.1
Overview of CODMn criteria in various Danube countries
Water quality Concentration
Concentration
Remarks
Information
criteria system
(lower range)
(higher range)
source
[mg O2/l]
[mg O2/l]
TNMN
5 Class I
20 - >50 (Class III V)
Adamkovį,
10 Class II
2003
Czech Republic 5 Class I
10 Class II
I: very clean
Haskoning,
25 Class IV
II: clean
1994
>25 Class V
IV: intensely polluted
V: very polluted
Hungary
8 Class I
15 Class II
I: high quality, clean waters
Haskoning,
>15 Class III
II: polluted; no detrimental 1994
impact on aquatic
ecosystem
III: polluted, detrimental to
aquatic ecosystem
Romania
10 Class I
15 class II
I: drinking water
Haskoning,
25 class III
II: recreation and fishing
1994
III: irrigation and industry
5 UN/ECE oxygen regime classes:
I
Constant near-saturation of oxygen content. Insignificant presence of oxygen demanding substances from the point of view of
aquatic life.
II
The oxygen saturation of water is good. Oxygen demanding substances do not normally disturb oxygen saturation
III
Oxygen deficiencies may occur in the hypolimnion. The presence of oxygen demanding substances risks having sometimes
considerable negative impact s on aquatic life through the reduction of oxygen contents.
IV
Oversaturation of oxygen or oxygen deficiency occurs in the epilimnion and oxygen deficiencies are frequent in the hypolimnion,
possibly owing to chronic problems with the presence of oxygen demanding substances.
V
Acute problems occur in oxygen regime, i.e. oversaturation or oxygen deficiency in the epilimnion, and oxygen deficiency
leading to anaerobic conditions in the hypolimnion. The high level of presence of oxygen demanding substances may equally
cause acute oxygen deficiencies.
RODECO Consulting GmbH
Orientation on environmental quality standards for nutrients and other Danube specific priority substances
I 31
4.2.3 Japan (lakes)
The information mentioned in the introduction of this chapter is summarised in the table below
[EMECS, 2003].
Table 4.2
Overview of CODMn criteria in Japan (lakes)
CODMn
Description
[mg O2/l]
=1
not to be caused by anthropogenic influence
=3
oligotrophic and eutrophic lakes containing oligosaprobic fish such as smelt
=5
eutrophic lakes containing carp (Water Quality Standards for Fisheries, 1965)
=6
for agriculture use
=8
desirable for waters used for swimming;
acceptable for most industrial uses and for conservation of environment
4.3 Synthesis of findings for `good status' of CODMn
The EQSs in the previous section are rather well comparable. The UN-ECE class II range (3 10 mg
O2/l) encompasses the lower ranges of the other EQS criteria. The class II description ("The oxygen
saturation of water is good. Oxygen demanding substances do not normally disturb oxygen
saturation") is an appropriate approximation of the `good' status of the oxygen regime. Hence, the
recommended EQS for `good status' for CODMn is set to =10 mg O2/l.
4.4 Comparison of proposed EQS with actual concentrations
COD was not analysed during the JDS. Using the annual average data (as can be retrieved from
Danubis), the average CODMn concentration of the TNMN data over the period 1996-2000 was 4.7 mg
O2/l. This easily fits within the recommended EQS of =10 mg O2/l. Maximum concentrations though
could go as high as 49 mg O2/l. More details on the results of the TNMN can be found in [Adamkovį,
2003].
I 32
UNDP/GEF Danube Regional Project
5 Metals (As, Cr, Cu, Zn)
5.1 Introduction
Metals can be present in rivers originating from natural (geogenic) sources. The WFD defines the
status of these specific non-synthetic pollutants as follows (annex V, table 1.2)
· High status : Concentrations remain within the range normally associated with undisturbed
conditions (background levels = bgl).
· Good status : Concentrations not in excess of the standards set in accordance with the
procedure detailed in section 1.2.6 (2) without prejudice to Directive
91/414/EC and Directive 98/8/EC. (<EQS).6
At low concentration levels many (heavy) metals as natural trace elements can be essential for most
living organisms. However, at higher concentrations, metals can become toxic.
The general purpose one can derive from the text of WFD Annex V, 1.2.6 is the need for defining a
No Observed Effect Concentration. The differences between the description of `high' and `good'
status of the specific non-synthetic pollutants like metals in the WFD in principle allow for
environmental conditions influenced by anthropogenic activities (read: polluted beyond the natural
background loading).
5.2 Total, dissolved, adsorbed?
A discussion about monitoring and setting EQSs for heavy metals often concerns which occurrence(s)
should be taken into account: only the dissolved part (more readily bioavailable), or the total
concentration (adsorbed + dissolved)?
In this context, it is interesting to notice that the target-values (in German: "Zielvorgaben") for heavy
metals in the Rhine Action Programme are formulated for suspended solids only. These
`Zielvorgaben' for suspended solids/sediment take into account: disposal of dredged sediment on land
and sea, plus protection of organisms living in the sediment.
As far as the queries could identify, actually few existing water quality criteria systems for heavy
metals in aquatic environments comprise EQSs for suspended solids/sediment. The Netherlands has
elaborated quite an extensive set of EQSs for heavy metals. This includes water quality standards for
total, dissolved, and sediment concentrations.
With the majority of water quality systems having defined standards for total (and/or dissolved)
concentrations, the focus of this project will be on total concentrations. Nevertheless, relevant data
concerning suspended solids and sediments will be incorporated as well.
5.2.1 Short primer on some key features of total, dissolved, and adsorbed metals
Heavy metals in aquatic environments normally are present in both the water phase (dissolved) and
adsorbed to particles (suspended solids, sediment). The ratio dissolved/adsorbed varies among the
metals. The table below shows the percentage dissolved in the total concentration for some metals for
average conditions that apply to the river Rhine (refer also to textbox 5.1 for details).
6 Section 1.2.6 of WFD Annex V outlines the procedure for the setting of chemical quality standards by Member States. Directive
91/414/EC is the COUNCIL DIRECTIVE of 15 July 1991 concerning the placing of plant protection products on the market. 98/8/EC refers
to the Directive 98/8/EC of the European Parliament and of the Council of 16 February 1998 concerning the placing of biocidal products on
the market.
RODECO Consulting GmbH
Orientation on environmental quality standards for nutrients and other Danube specific priority substances
I 33
Textbox 5.1
Partitioncoefficient Kd and formulas for (re-)calculation of metal concentrations
The partitioncoefficient Kd represents for metals in aquatic systems the `equilibrium' between the amount of the
metal dissolved in the water phase versus the amount adsorbed to the suspended solids (adsorbed; solid phase).
Ctot - Cdis
Kd =
(1)
SS *Cdis
The Kd values mentioned in table 5.1 were calculated from three years of field data collected in various Dutch
waterways [RIZA, 1989].
In case only dissolved concentrations are known (measured), then the total metal concentration can be calculated
using formula (2a), either (2b)
Ctot =
SS
*
Kd
* Cdis+ Cdis
(2a)
Ctot Cdis
=
*(1+
SS)
*
Kd
(2b)
The metal concentration adsorbed to the suspended solids could be calculated directly from the field data (in this
case, the total and dissolved metal concentration plus the suspended solids concentration are measured) using
formula (3),
Ctot - Cdis
Cads =
(3)
SS
or approximated with formula (4) (following from (1) and (3)), when only the dissolved concentration is known
Cads
= Kd *Cdis
(4)
either through formula (5), when the total metal concentration plus the suspended solids concentration are
known
Ctot
Cads = SS +
(5)
1
( / Kd)
with:
Kd:
partitioncoefficient water/solid phase [l/g]
Ctot:
total heavy metal concentration [µg/l]
Cdis:
dissolved heavy metal concentration [µg/l]
Css:
concentration heavy metals in suspended solids fraction [mg/kg]
SS:
suspended solids concentration in [g/l]
Formulas quoted from [RIZA, 1989].
Please refer to subsection 5.3 for examples of the limitations and pitfalls of the formulas above.
I 34
UNDP/GEF Danube Regional Project
Table 5.1
Percentages dissolved metals of total concentration
Kd
Dissolved fraction
[l/g] (with SS= 30 mg/l)
Ni 8
81%
As 10
77%
Cu 50
40%
Zn 110
23%
Cd 130
20%
Hg 170
16%
Cr 290
10%
Pb 640
5%
SS= suspended solids
The table illustrates that in an unfiltered (freshwater) water sample for instance arsenic is dissolved in
the water phase for 77%, while in the case of chromium 90% is adsorbed to the suspended solids.
For particulate (adsorbed) concentrations of micropollutants further the composition of the suspended
solids and sediment are relevant. Micropollutants tend to adsorb to the smaller suspended solids
fraction (e.g. clay) and to the organic matter. The Dutch EQSs therefore apply to `standard suspended
solids', consisting of 20% organic matter and 40% `lutum' (clay <2 µm fraction). `Standard sediment'
consists of 10% organic matter and 25% lutum [V&W, 2000].
For a proper comparison of sample concentrations with the Dutch EQSs, the results are first to be
(re-)calculated to the standard conditions 7. This, in order to remove the bias that would result from
differences in the amount of organic matter and/or small solids fractions. (A more sandy sediment
sample might seem to be less polluted than a sample taken at the same location containing finer
particles. If no sieving prior to analysis, or standardisation like above on the analysis results has been
carried out, such would be the interpretation.)
An alternative approach for standardising sediment (and suspended solids , given a sufficient amount
of material) prior to analysis is sieving; common are 63 or 20 µm pores.
5.3 Natural background levels
Data on natural background levels of heavy metals for the purposes of this study could be identified
for: the Danube, the Rhine basin, and the Netherlands. The International Rhine Commission (IRC) has
defined natural background for sediment/suspended solids for the Rhine River basin. The Netherlands
has defined natural background concentrations of metals in sediment and in water (total
concentration). Data related to the Danube basin are included in section 5.2.3 of the JDS Technical
report [ICPDR, 2002]. The various values are included in the tables below.
7 The actual (re-)calculations are a bit more complicated than mentioned here. Additional coefficients are used in the calculations, which
differ among the various metals.
RODECO Consulting GmbH
Orientation on environmental quality standards for nutrients and other Danube specific priority substances
I 35
Table 5.2
Natural background concentrations of metals in sediment [mg/kg]
Rhine [IKSR, Rhine [IKSR, The Netherlands [V&W, 2000]
Danube
1993a,b]
1993a]
(sediment with 10% organic matter and 25% baseline
(< 20 µm fraction)
(< 20 µm fraction)
<2 µm)
[ICPDR, 2002]
AVERAGE
RANGE
(< 63 µm)
As
20&
not defined
29
10
Cr
80
40 160
100
10 - 50 (?)
Cu
20
10 40
36
35
Zn
100
50 200
140
130
& The Rhine target value for As has been defined as 2 x natural background, hence 20 mg/kg can be deferred.
Table 5.3
Natural background concentrations of metals in water [µg/l]
Rhine
The Netherlands The Netherlands `Danube'
[IKSR]
[V&W, 2000]
[V&W, 2000]
[ICPDR, 2002]
total
dissolved
total
As not defined
1.0
0.8
-
Cr
not defined
1.6
0.2
1.3 5.0
Cu not defined
1.1
0.4
0.5 2.0
Zn not defined
12
2.8
1.8 7
Table 5.2 implies that the IRC has defined lower background concentrations for Cr, Cu, and Zn in
sediment than the Netherlands8. These are not necessarily `real' differences (both the IRC and the
Dutch values are estimated `best-expert' approximations). The IRC for instance applies the same
natural background levels for the metals in both sediment and in suspended solids [IKSR, 1993a,
1993b]. The explanation of the IRC is that sieving suspended solids either sediment over 20 µm more
or less will equalise possible differences in the composition of the original material (like unsieved
sediment samples containing more sand). No data were immediately ava ilable showing whether or not
sieving sediment over 20 µm would result in `standard sediment' complying with the Dutch
definitions.
`Indirectly' one can infer that the IKSR and the Dutch systems actually share similarities that could
explain the apparent differences in the background concentrations. Being a bit ahead of the EQSs
presented later on, this can be illustrated as follows. The `Zielvorgaben' for metals of the Rhine Action
Programme (RAP) are defined for suspended solids [IKSR, 1992]. The table with the EQSs for the
Netherlands contains numeric values for sediment [V&W, 2000]. As mentioned in this document, the
EQSs for metals in suspended solids are a factor 1.5 higher than the sediment concentrations
(assuming both `standard sediment' and `standard suspended solids'. Refer also to [RIZA, 1989]). The
various values are included in the table below. (As mentioned in subsection 2.4.1.4, in the Netherlands
actually two sets of EQSs are defined, a No Effect Level and a MAC value. The Rhine target-values
are considered to be equivalent to NOELs).
8 Arsenic is not a priority substance in the Rhine Action Programme, hence no `Zielwert' or background concentration have been
formulated.
I 36
UNDP/GEF Danube Regional Project
Table 5.4
Target values for particulate metal contents in the RAP and in NL [mg/kg]
The Netherlands The Netherlands
Rhine [IKSR]
sediment, NOEL SS, NOEL (=sediment * 1.5) SS, Zielvorgabe
As
29
44
not defined
Cr
100
150
100
Cu
36
54
50
Zn
140
210
200
Cd
0.8
1.2
1.0
Hg
0.3
0.45
0.5
Ni
35
52
50
Pb
85
128
100
The Dutch `calculated' suspended solids target values are comparable to the RAP target values9,
except for chromium. From this, one can infer that the associated sediment quality (also of the natural
background) actually can supposed to be similar as well.
5.3.1 Examples of limitations for using the formulas in textbox 5.1
The Rhine Action Programme has formulated target-values for suspended solids only. Of course this
triggers the question how these will relate to dissolved or total metal concentrations. This, in order to
be able to compare with other sets of EQS that for instance only contain total or dissolved
concentrations.
The formulas presented in textbox 5.1 imply this should be feasible, and only requires rather basic
mathematics. Unfortunately, it turns out that the Dutch EQSs for total, dissolved, and sediment
mutually do not comply with the results as would be expected when applying these formulas. The
Dutch NOEL water quality standards for the metals of present study are summarised in the table below
[V&W, 2000].
dissolved total
sediment suspended solids
(= sediment * 1.5)
[µg/l]
[µg/l] [mg/kg]
[mg/kg]
As
1.0
1.3
29
44
Cr
0.3
2.4
100
150
Cu
0.5
1.1
36
52
Zn
2.9
12.0
140
210
Applying the formulas in textbox 5.1, together with the Kd values mentioned in Table 5.1, and 30 mg/l
as the `standard suspended solids concentration', following results are obtained.
9 Of course, this is not just a coincidence. The major part of the Netherlands is part of the Rhine basin. There are many cross-references and
overlaps between the national Dutch and international Rhine policy settings and strategies.
RODECO Consulting GmbH
Orientation on environmental quality standards for nutrients and other Danube specific priority substances
I 37
Total ó Dissolved
The calculated results match quite well with the defined values, maybe except for the total-Cr
concentration calculated from the dissolved chromium concentration (see table below)
dissolved dissolved
total
total
defined
calculated defined calculated
from total
from dissolved
[µg/l]
[µg/l]
[µg/l]
[µg/l]
As
1.0
1.00
1.3
1.3
Cr
0.3
0.25
2.4
2.9
Cu
0.5
0.44
1.1
1.25
Zn
2.9
2.79
12.0
12.5
Adsorbed ó Total and Dissolved
The results of using the total and dissolved concentrations for calculating the suspended solids
significantly differ from the actual ones!
Suspended solids Suspended solids Suspended solids
defined
calculated
calculated
from total
from dissolved
[mg/kg]
[mg/kg]
[mg/kg]
As
44
10
10
Cr
150
72
87
Cu
52
22
25
Zn
210
307
319
The above findings indicate that one should be quite careful when applying formulas like the ones
presented in textbox 5.1. More considerations then merely equilibrium coefficients can apply when
establishing EQSs, for instance depending the specific feature of the media (like organisms living in
the water phase, versus those mainly living in bottom sediments).
5.4 Inventory of water quality criteria for metals resembling `good status'
5.4.1 United Nations
The UN "ECE Standards Statistical Classification of Surface Water Quality for the Maintenance of
Aquatic Life" defines five classes of metal concentrations [UN/ECE, 1992]. Major criterion
underlying the metal section is "Toxicological impact on aquatic life as established in US-EPA
practices." Although not explicitly mentioned in the document, metal concentrations can be inferred to
apply to total concentration in water.
I 38
UNDP/GEF Danube Regional Project
UN-ECE
class I10
class II
class III
class IV
class V
[µg/l]
Asc
<10
10 - 100
100 -190
190 360
>360
Crc
<1
1 6
6 11
11 16
>16
Cud
<2
2 7
7 12
12 18
>18
Znd
<45
45 77
77 110
110 -120
>120
c Applicable for hardness from about 0.5 meq/l to 8 meq/l. Arsenic V (chromium III) to be converted to arsenic III (chromium VI).
d Applicable for hardness from about 0.5 meq/l to 8 meq/l.
5.4.2 European Union
EU Directives, explicitly mentioning EQSs for As, Cr, Cu, and/or Zn are:
· 75/440/EEC, concerning the quality required of surface water intended for the abstraction of
drinking water in the Member States;
· 78/659/EC: on the quality of fresh waters needing protection or improvement in order to
support fish life.
The standards of the 75/440/EEC Directive are merely mentioned for the sake of completeness. The
Guide respectively Imperative values for total concentrations of the most stringent A1 category are:
75/440/EEC
A1
A1
[µg/l]
Guide
Imperative
As
10
50
Cr (VI)
-
50
Cu
20
50
Zn
500
3000
The basic aims of the 78/659/EEC directive are close to the context of the WFD. This `fish directive'
contains EQSs for Cu and Zn. For copper only Guidance (non-binding) concentrations are formulated
for the dissolved fraction. For zinc only Imperative (mandatory) EQSs for the total concentrations are
defined. The main values as mentioned in Annex I of the directive apply to a hardness corresponding
with 100 mg/l CaCO3. Annex II of the Directive shows the concentrations associated with other
hardness values. Further, two different kinds of waters are further discriminated: Salmonid and
Cyprinid. The various criteria are included in the table below.
10 UN/ECE metal classes:
I
No anthropogenic pollution with inorganic matter.
II
Concentrations are below midpoint between natural and chronically toxic levels.
III
Concentrations are above midpoint between natural and chronically toxic levels.
IV
Excursions beyond chronic criteria concentrations occur, but do not establish chronically toxic conditions in terms of
concentration levels, duration or frequency.
V
Excursions beyond chronic criteria concentrations allow acutely toxic conditions in terms of concentration levels, duration or
frequency.
RODECO Consulting GmbH
Orientation on environmental quality standards for nutrients and other Danube specific priority substances
I 39
78/669/EEC
Cu
Cu
Zn
Zn
dissolved
dissolved
total
total
Salmonid waters
Cyprinid waters
Salmonid waters
Cyprinid waters
(Guide value)
(Guide value)
(Imperative
(Imperative
hardness
[µg/l]
[µg/l]
value)
value)
[mg/l CaCO3]
[µg/l]
[µg/l]
10
-
5
30
300
50
-
22
200
700
100
400
40
300
1000
500
-
112
500
2000
For dissolved copper, the most stringent value applie s to Salmonid waters, while for Zn the most
stringent values apply to Cyprinid waters. Hence, assigning one of both water categories (Salmonid or
Cyprinid) would not suffice when trying to be on the safe side of both metals. Secondly, the directive
explicitly discriminates different concentration criteria for different levels of water hardness. This is
different from the UN criteria that encompass one single concentration for quite a wide hardness
range.
While keeping the constraints of calculations in mind: a dissolved copper concentration of 400 µg/l
would imply a total concentration of 1000 µg/l; 40 µg/l dissolved Cu implies 100 µg/l.
5.4.3 United States
The National Recommended Water Quality Criteria as defined by the US-EPA comprise two sets of
concentrations of the four metals: CMC and CCC levels [EPA, 2002]. Quoting [EPA, 2002]: "The
Criteria Maximum Concentration (CMC) is an estimate of the highest concentration of a material in
surface water to which an aquatic community can be exposed briefly without resulting in an
unacceptable effect. The Criterion Continuous Concentration (CCC) is an estimate of the highest
concentration of a material in surface water to which an aquatic community can be exposed
indefinitely without resulting in an unacceptable effect."
The description of the (lower, more stringent) CCC can be considered as an approximation of `good
status'. Nevertheless, both classes of concentrations are shown in the table below.
Following footnote D in [EPA, 2002]: "Freshwater and saltwater criteria for metals are expressed in
terms of the dissolved metal in the water column", the water quality criteria refer to dissolved
concentrations. The total concentrations were calculated for this report purposes only; the limitations
of such calculations should be kept in mind.
CMC CCC
calculated
total CCC
[µg/l] [µg/l]
[µg/l]
As
340
150
195
Cr (III)
2.0
0.25
-
Cr (VI)
570
74
718
Cu
13
9.0
22
Zn
120
120
516
5.4.4 The Netherlands; Rhine
Also referring to the discussion in section 5.3, it has been decided to mention the Dutch EQSs only.
The calculation exercises, that inferring concentrations/EQSs for total and dissolved concentrations
I 40
UNDP/GEF Danube Regional Project
from suspended solids concentrations only, can be very tricky. In addition, may lead to inconsistent
results. As far as the suspended solids are concerned, the EQSs defined for the Rhine basin and for the
Netherlands compare quite well (except for chromium). It is assumed that the major principles
underlying the EQSs for the RAP and for NL are comparable.
Basically, four times two sets of EQSs can be discriminated in the Dutch system of EQSs. For
respectively total, dissolved, sediment, and suspended solids metal concentrations both NOEL and
MAC values are defined.
The NOELs of the total concentrations of Cu and Zn are similar to the natural background. For As
and Cr, the NOELs are higher.
NL
natural
NOEL MAC
background
Total [µg/l]
[µg/l]
[µg/l]
As
1.0
1.3
32
Cr
1.6
2.4
84
Cu
1.1
1.1
3.8
Zn
12.0
12
40
The NOELs of the dissolved concentrations for all four metals are (slightly) higher than the natural
background.
NL
natural
NOEL MAC
background
Dissolved [µg/l]
[µg/l]
[µg/l]
As
0.8
1.0
25
Cr
0.2
0.3
8.7
Cu
0.4
0.5
1.5
Zn
2.8
2.9
9.4
The NOELs for sediment for all four metals are equal to the natural background.
NL
Sediment
Sediment Sediment Suspended Suspended
natural
solids
solids
Adsorbed background NOEL
MAC
NOEL
MAC
[mg/kg]
[mg/kg]
[mg/kg]
[mg/kg]
[mg/kg]
As
29
29
55
44
83
Cr
100
100
380
150
570
Cu
36
36
73
54
110
Zn
140
140
620
210
930
Generally, the Dutch NOELs are quite close to the defined natural background concentrations.
Therefore, it seems reasonable to interpret them as `high status' criteria. The MAC values could be
qualified as setting the boundaries for `moderate' (or worse) status.
RODECO Consulting GmbH
Orientation on environmental quality standards for nutrients and other Danube specific priority substances
I 41
5.4.5 Canada
The Canadian systems of water quality standards comprises water as well as sediment [Environment
Canada, 2002]. For water, different sets of standards are formulated: community, recreation and
aesthetics, and aquatic life. In the table below, the freshwater standards for aquatic life are mentioned.
The values refer to the total concentration in an unfiltered sample [Environment Canada, 1999a]. For
sediment in freshwaters, two categories are discriminated: ISQG: Interim sediment quality guideline,
and PEL: Probable effect level. The ISQG is more stringent than the PEL. There is no explicit
mentioning of sieving prior to analysis or to a certain composition of the sediment [Environment
Canada, 1999b].
Canada Water
Sediment Sediment
(aquatic life) ISQG
PEL
[µg/l]
[mg/kg]
[mg/kg]
As
5.0
5.9
17
Cr
&
37.3
90
Cu
2 4
35.7
197
Zn
30
123
315
& For water, no value is contained for chromium as such. Values are include for trivalent chromium Cr(III): 8.9 µg/l, and hexavalant
chromium Cr(VI): 1.0 µg/l.
5.4.6 Joint Danube Survey
The heavy metal section 5.2.3 of the JDS Technical Report contains a series of quality targets
[ICPDR, 2002]. They are compiled from various sets of standards, and from different data sources.
For the sake of completeness, the quality targets are included in the table below.
JDS Water Suspended solids Sediment
(total)
[µg/l] [mg/kg]
[mg/kg]
As
-
20
20
Cr
3.1
100
100
Cu
3
60
60
Zn
7
200
200
The quality targets are the same for sediment and suspended solids. In most cases the suspended
solids/sediment targets are similar, either quite close to the of the IRC.
5.4.7 Synthesis of findings for `good status' of metals
First of all it can be concluded that there can be big differences between the various sets of EQSs.
Compare for instance the UN/ECE and Dutch EQSs on the one hand, versus those from the EU and
US on the other hand.
The dissolved US-EPA CCC concentration of Cr is higher than the UN/ECE Class V (total)
concentration, while the dissolved CCC concentration of Zn is equal to the boundary between Class IV
and V (120 µg/l). The dissolved Cu CCC concentration would qualify as UN-ECE Class III. The
calculated total CCC concentrations surpass the UN-ECE class V substantially.
The US-EPA arsenic criteria would qua lify as UN-ECE Class III (dissolved) to IV (calculated total).
The guide concentrations of dissolved Cu in the EU are higher than US-EPA CMC criteria.
I 42
UNDP/GEF Danube Regional Project
The mandatory Zn total concentration for Salmonid waters (300 µg/l) is higher than the dissolved
CMC/CCC (same values) concentration of 120 µg/l, but lower than the calculated total (516 µg/l). The
mandatory Zn concentration is almost three times larger than the concentration marking UN-ECE
class V (>120 µg/l).
The Dutch EQSs were not yet mentioned, but it should be sufficient to mention that the MAC values
for Zn would fit in Class I of the UN-ECE, and for As and Cu in class II. The MAC for Cr (84 µg/l) is
significantly higher than the Class V boundary (16 µg/l), but quite close to the dissolved CCC
concentrations of the US-EPA (74 µg/l).
For the sake of completeness, the descriptions of the CMC, CCC and of Class V are quoted again:
· CMC The Criteria Maximum Concentration is an estimate of the highest concentration of a
material in surface water to which an aquatic community can be exposed briefly
without resulting in an unacceptable effect.
· CCC The Criterion Continuous Concentration is an estimate of the highest concentration of
a material in surface water to which an aquatic community can be exposed indefinitely
without resulting in an unacceptable effect.
· Class V Excursions beyond chronic criteria concentrations allow acutely toxic conditions in
terms of concentration levels, duration or frequency.
What is considered as a critical (MAC, acute toxic) concentration in one system (UN, NL) can be a
recommended (acceptable) concentration in the other system (EU, US). Or, an approximate
`high'/'good status' in one set of criteria would qualify as `moderate' (and worse) in another.
This observation leads to a `stale mate' also as far as the aims of present project are concerned. The
different water quality criteria systems were developed by well-known and respectable bodies. For all
four systems it is mentioned, or can it be inferred, that the concentrations were defined based upon
eco-toxicological risk assessments. It would be imprudent to `choose' one of the EQS systems (either
to compile some averages) without having additional knowledge, arguments, and criteria. Latter
implies a more in-depth screening by a qualif ied ecotoxicologist, which definitely goes beyond the
settings of this project. It is therefore proposed to be taken into consideration as a follow-up activity.
5.4.7.1 Total, dissolved, adsorbed?
Taking into account the approach of the RAP, where target-values for metals were set for suspended
solids only on the one hand, and e.g. the US-EPA approach (where criteria for metals apply to the
dissolved concentrations only) on the other hand, it generally can be advised for the Danube to define
metal EQSs for both the total, dissolved, and adsorbed (suspended solids) concentrations. In this way
one anticipates all possible environmental situations and compartments. A system of EQSs for metal
encompassing both water and solid phases is definitely more watertight, and is not expected to demand
too much extra efforts (compared to defining EQSs for e.g. dissolved concentrations only).
Section 5.3 outlines points of attention for continuation of such activities, like the (im-)possibility for
establishing mathematical relationships between total, dissolved, and adsorbed concentrations. Since
analysing the metals concentrations in suspended solids will become more a common practice in the
Danube basin, information can be gathered which allows for defining factors more specific for
condition in the Danube basin (e.g. Kd, average suspended solids concentration, composition of
suspended solids and sediment).
5.5 Comparison with actual metal concentrations
Because of the stalemate in recommending EQSs for the metals, a proper comparison with measured
data cannot be made. Readers are referred to section 5.2.3 of the JDS Technical report [ICPDR, 2002].
This contains an exhaustive assessment of the heavy metal findings in the JDS.
RODECO Consulting GmbH
Orientation on environmental quality standards for nutrients and other Danube specific priority substances
I 43
6 Conclusions and recommendations
Readers are forwarded to chapter 7 for proper apprehension of the conclusions and
recommendations contained in this chapter.
While aiming at formulating recommendations for EQO/EQSs, it has been considered equally
important to provide background material and arguments to the MLIM and EMIS expert groups for
supporting their further activities. The findings of the underlying study can be summarised as follows.
6.1 Nutrients: Ntot and Ptot
For nutrients the following EQSs are suggested as "good status" values, in line with the WFD. The
related EQO is to avoid eutrophication in the Danube basin.
Ų Ntot:
1.0 1.5 mg N/l
Ų Ptot:
0.02 0.08 mg N/l
The above values are presented as ranges. The present study could not provide additional criteria to
decide which single value to select from within these ranges. Follow-up activities can include:
· consultations with biological experts;
· investigations on the actual occurrences of eutrophication in the Danube basin, combining the
findings with the physico-chemical data available;
· developments in other European river basins during implementation of the WFD.
The proposed EQSs have not taken the Black Sea into account. Ultimately, the EQSs to be set should
both enable a "good status" situation within the Danube Basin itself, as well as in the Black Sea
regions influenced by the discharge of the Danube.
6.2 NH4
From a potential toxicity point of view, a separate EQS has been esta blished for the "good physico-
chemical status" of ammonium:
Ų NH4: =0.2 mg N/l
6.3 CODMn
For the chemical oxygen demand, the following EQS is suggested:
Ų CODMn: =10 mg O2/l
6.4 Metals: As, Cr, Cu, Zn
It has not been feasible to prepare recommendations for EQSs for the metals As, Cr, Cu, and Zn. The
major problem is that there can be huge differences between the values of different sets of water
quality standards. In principle, ecotoxicological research and criteria have been underlying the sets of
standards included in the inventory of this study. It will require further, more in-depth investigations to
find out how such differences can be explained, despite sharing `ecotoxicology' as a shared basis. In
addition, it is useful to keep track in the development in other European river basins, as far as they also
will include these metals in addition to the list of WFD priority pollutants.
I 44
UNDP/GEF Danube Regional Project
7 Epilogue: comments to the draft final report
The draft final report of August 2003 was discussed on 17 and 19 September 2003 during respectively
the 2nd Joint MLIM/EMIS Working Groups meeting and the 31st MLIM meeting. The German and
Austrian representatives afterwards sent their comments and suggestions by e-mail, which contains the
majority of the issues raised during the meetings in September. The comments sent by e-mail are
included in the first section, followed by brief responses by the author.
7.1 Remarks to draft report submitted by e-mail
7.1.1 Germany
The draft report deals with quality standards for nutrients. This is a very complex issue which has been
discussed within the ICPDR-MLIM working group for several years. It should be made clear that
aspects mentioned in this report shall not be interpreted as final solution but can be used as a
contribution to this discussion.
The following items should be added / considered, at least by a few remarks:
- background of the EQS or reference values mentioned in the report (context in which they were
developed, purpose, water types, legal restrictions)
- differentiation between surface water types
- relation between nutrient concentration and eutrophic conditions
- availability of nutrients / heavy metals to organisms (discussion on dissolved fractions by MLIM)
- relation between concentrations and loads
Concerning seasonality of concentration and time specific EQS this approach is not considered as very
helpful because for example phosphate concentrations may be decrease to detection limit when this
nutrient is assimilated in spring. In this case no EQS could be defined for this season. Usually
seasonality is considered by using 90 percentiles of an annual data set which reflects periods of higher
concentrations.
7.1.2 Austria
In general, the current report can serve as an interesting discussion basis for developing quality
objectives for nutrients in the Danube basin. Specific issues within the report need reconsideration.
Therefore and in addition to the comments made during the referring discussion in Bratislava, we
would like to state the following input comments:
· The typespecific approach required by the EU Water Framework Directive is not considered
within the report.
-
One single concentration value for N and P parameters concerning high status,
reference/background conditions, good status and thresholds (in general and such thresholds
demanding immediate action, EQS values) for the entire extent of the Danube and its
tributaries does not follow the typespecific approach. The mentioned values will for sure differ
within the Danube Basin depending on the typological region. Hence, set concentration values
for high status, reference/background conditions, good status and thresholds (in general and
such thresholds demanding immediate action, EQS values) will vary over the extent of the
Danube and in its tributaries.
-
If thresholds/EQSs are suggested they have to follow the typespecific approach and therefore
need value adaptation to the relating condition of the Danube or its tributaries.
-
The recommended EQS values need reconsideration.
· MONERIS is definitely an interesting and useful model for the calculation of diffuse inputs and
origins in river basins. However, MONERIS cannot be used for setting any concentration values
for High Status or other management thresholds.
· Concerning several mentioned N and P concentrations it is unclear what kind of in-stream values
are addressed (e.g. guideline values or threshold values etc.).
RODECO Consulting GmbH
Orientation on environmental quality standards for nutrients and other Danube specific priority substances
I 45
· All values which go back to specifications by MLIM expert group are not average but 90% values
(e.g. see table 2.2-2.5.).
· A correlation between chemical values and their effect on biological coenoses is missing in
general. E.g. concerning historical data.
· Concerning eutrophication, nutrient values related for lakes are regarded. Lake values should not
be used for river related considerations and management due to the different typespecific
characterisation of these two systems. Even in the dammed sections of the Danube such values do
not seem appropriate.
· Values for oligotrophic conditions are considered relevant for indicating High Status and
concentrations of mesotrophic conditions for Good Status. This approach should be reconsidered
as it does not include the typespecific approach.
7.2 Heavy metals
During the presentation of the results on 17 September 2003, it was suggested that one of the options
to proceed with formulating EQSs for the Danube specific priority metals (As, Cr, Cu, and Zn) could
be to apply the methodology of the Expert Advisory Forum on WFD Priority Substances.
As it turned out, this actually already has been done in Austria. The approach for the derivation of
EQSs follows strictly the procedure given in Annex V, 1.2.6 of the Water Framework Directive and
the methodical proposals of the EQS-study commissioned by the EC to the Fraunhofer Institute (FHI)
for Molecular Biology and Applied Ecology. It follows the "added-risk" approach (see also the section
on Typespecific approach below) and focuses on the dissolved phase as the first step in setting out
EQS for metals as applied by FHI.
The report also touches upon the discussion still ongoing in Europe, concerning the issue whether the
added-risk approach refers to the dissolved (i.e. filtered at 0.45 µm) or to the total metal fraction or
suspended matter fraction.
The final document is still being subjected to a national review; hence, the final details will be made
public after completion of the review. It is expected that this document will be useful for the Danube
community concerning the formulation of EQSs for As, Cr, Cu and Zn.
7.3 Typespecific approach
Several of the remarks comment upon the report not having followed the typespecific approach. The
author acknowledges this notice. The present study not having followed the typespecific approach had
two major reasons. One reason has been that several Danube countries still are working on the
typology and reference conditions for surface water bodies. The other and most limiting factor was
simply time constraints (two person-weeks were allocated for both underlying tasks). As was agreed
during the 1st Joint MLIM/EMIS meeting in February 2003, this assignment was first of all to focus on
the main course of the Danube River. (In the MLIM-Working paper on Typology and reference
conditions for surface water bodies of 13 May 2003 it has been recommended to consider the Danube
river itself as a water body of its own.)
The report already included several subsections where limitations in the approach and in the results
were addressed, but a brief review is added here as well.
Generally, taking into account the typespecific approach when formulating environmental quality
standards for physico-chemical parameters implies that at least following issues are taken into
account:
· Natural background conditions. This applies to both nutrients and heavy metals. Differences in
the (geogenic) conditions of a certain (sub)basin already can result in different loading of the
water system by natural sources. There are examples where concentrations in the water
already could exceed existing water quality standards due to natural background loading only!
Partially in this context, the Austrian colleagues pointed-out to the so-called added-risk
approach when formulating EQSs for heavy metals, as recommended by the Fraunhofer
I 46
UNDP/GEF Danube Regional Project
Institute (who is working on establishing water quality standards for the WFD priority
substances). The environmental quality standard is derived from the sum of the background
concentration and a maximum permissible additional concentration derived from
ecotoxicological test data.
With respect to nutrients it further can be argued that specific and unique biological
characteristics of certain water bodies can only exist with relatively high nutrient
concentrations (from natural sources). Vice versa, examples exist of ecosystems with a
delicate balance that easily could be disturbed even by minor increases in a nutrient
concentration.
· Typespecific conditions. This applies especially to the nutrients (for potentially harmful
substances like heavy metals the added- risks for organisms are considered to be more
applicable more generally, although conditions like salinity or hardness are rele vant as well).
One example is for instance the difference between freshwaters and marine waters. Generally,
it is understood that for occurrence of eutrophication phenomena in freshwaters the P-
concentrations are most relevant (P-limited), while for marine waters nitrogen is the most
decisive ingredient (N-limited). This is merely a fist of rule, since there are also examples of
eutrophication occurring in freshwater due to enrichment with nitrogen. The EU-nitrates
directive 91/676/EEC for instance deals with this issue. Further, instead of absolute
concentrations, it can be the ratio between N and P concentrations, which is most relevant.
Unfortunately, there is no strict (mathematical) relationship between absolute concentrations
of nutrients and the actual occurrence of eutrophication. Finding out such peculiarities is a
good example of features of the typespecific approach,
Another already rather generic difference is whether one is dealing with a running freshwater
or a standing freshwater (as annotated by the Austrian colleagues, one also cannot put lakes
and reservoirs together by them sharing to be -rather- standing waters).
More specifically, the composition of the aquatic community in a certain water body will be
relevant when evaluated in the context of its sensitivity in relation to elevated concentration of
nutrients. Here quite sophisticated biological knowledge and expertise needs to be included in
the assessment and the setting of water quality standards.
The present study first of all made an attempt to make operational the WFD `good status' definition
for physico-chemical parameters. Also under the typespecific approach there still will be the need for a
translation key to convert a description like "nutrient concentrations do not exceed the le vels
established as to ensure the functioning of the ecosystem and the achievement of the values specified
above for the biological quality elements" into actual concentrations. As a bridging factor
`eutrophication' was used in the present study. The searches aimed at identifying concentrations of N
and P that can be related to the (risk of) occurrence of eutrophication. Consequently, in existing
systems the concentrations related to oligotrophic and mesotrophic conditions were discriminated (not
implying that to meet good status the whole Danube basin should be considered (to become) a
mesotrophic water).
In order to avoid possible misunderstanding: the recommended EQSs for Ntot and Ptot in section 2.6
were not suggested to apply to the whole Danube basin 11, as already indicated by the points of
attention mentioned in section 2.6. To these points, the typespecific approach and other additional
remarks can be added. As complications already experienced in the present study, following can be
mentioned.
· Defining natural background conditions. Even though meeting `good status' is the major final
WFD requirement, it can be very helpful to be able to formulate also the conditions where
there is undisturbed with virtually no anthropogenic impact (`high status'). It was not so easy
to find references that were dealing with defining natural background concentrations, let alone
that such information could be acquired for specific subbasins in the Danube. Maybe in
national or scientific libraries reports exist to such extent, but these sources were out of reach
11 During the discussions in September 2003 for instance it was mentioned that apples and pears were compared and
combined since for instance lake values seem to be implied for river related situations ...
RODECO Consulting GmbH
Orientation on environmental quality standards for nutrients and other Danube specific priority substances
I 47
of present study. If such data are not explicitly available, then there can be some fallback
options. 1) Historic data sets, where monitoring results are available going back until the
1950/60-ies, or preferably earlier. For reliable assessment of heavy metals such historic
datasets presumably are not suitable (because of analytical constraints). For nutrients, there
may exist significant data, but then again one can expect few or no data at all for organic
nitrogen. 2) Models like MONERIS are capable of calculating natural loads into (subbasins
of) the Danube, which can be used as a basis for estimating natural background
concentrations. 3) Reference areas, for which data about natural background concentrations
exist, and which then are assumed for the corresponding water body in the Danube basin.
· Additional criteria for setting the EQS. The nutrient concentrations presented in section 2.6
are ranges. As mentioned there, more criteria are required to decide whether the range fits to
the specific water body at all, and if so, in which part of the range to seek the corresponding
EQS. Depending on the outcome of the (national and international) typology of the Danube
basin one might consider the development of set of uniform criteria to be applied. Further, it
has been suggested to find out whether there are more datasets available that combine
biological and chemical assessments, as was the case in the Joint Danube Survey. Using the
actual occurrence of eutrophication phenomena as criterion, they than can be associated with
the measured nutrient concentrations.
· The recommended EQS for nutrients seem too stringent. This at least would be the conclusion
when one compares the suggested natural background concentrations (range) with the
suggested EQS-ranges. Whether or not they are too stringent finally will depend on the
features and water quality requirements of the specific water bodies. The author agrees that it
would become quite a difficult a task to realise good status, assuming that the proposed EQSs
indeed would turn out to be applicable for the Danube river itself.
7.4 Background of the values mentioned in the report
Some of the remarks asked for clarification of the status of the values mentioned in the various report
sections. Most of suchlike information is included in most sections when it was available (readers
could consult the various Internet- references for further information).
The status of the various sets of water quality standards can differ. For example, the Dutch MAC-
standards are already used as binding limits under the present water quality management (hence
appropriate action is required when water quality does not comply), while the NOEL-levels are
considered to provide a medium-range perspective of the desired situation. The `Zielvorgaben' of the
Rhine represent agreed water quality targets to which the riparian committed themselves to reach such
quality in the river Rhine. The values proposed by the UN/ECE are first of all suggested as values for
assessment of the water quality, hence in itself they are not EQSs (by e.g. adopting the Class II values,
the same values can become EQSs).
Generally, for the nutrients in chapter 2 in many cases status assessment values are mentioned. These
assessment values were then transferred by present study to setting the provisional range of the
proposed `good status' EQS for Ntot and Ptot. The heavy metals chapter contains more examples of
values which are already existing as EQSs in various countries.
In several cases, it is sometimes a matter of interpretation whether or not one a certain value could be
set equivalent to `high' status, or `good status', `moderate status', etcetera.
In the case of the heavy metals, it is obvious that the differences between values are not a matter of for
instance comparing MAC with NOEL levels. From the descriptions provided in the text of the report,
the intention of the different sets can be interpreted rather straightforward.
In the synthesis of the EQSs for Ptot, Ntot, NH4 and CODMn, it was aimed at proposing values to be
considered as threshold values for `good status' (non-compliance means moderate status or worse).
For Ptot and Ntot, the EQSs are suggested to apply to total concentrations. For the occurrence of
eutrophication, both the dissolved and adsorbed nutrients are relevant (the issue of eutrophication is a
different from the discussion on bioavailablity of dissolved fractions of heavy metals ).
I 48
UNDP/GEF Danube Regional Project
7.5 Compliance testing
Several remarks can be summarised as addressing the issue of compliance testing12. As mentioned just
above, the EQSs suggested for the nutrients and for chemical oxygen demand are considered to
represent threshold values for a WFD `good status'. The author did not elaborate on how to use the
values in the perspective of compliance testing.
· Percentiles. It is quite common to use the 90-percentile (or other peak-shaving methods) of the
annual- monitoring results when comparing the actual water quality with the standards.
Using a statistical method as the 90-percentile implies that one has a sufficient number of data
available, often mentioned to be from 6 to 10 samples as a minimum. In this perspective, it is
interesting to notice that the WFD mentions in Annex V, 1.3.4 a sampling frequency of 3
months (implying four samples a year) for nutrient status.
· Average concentrations/seasonality. The author suggested taking into consideration some
general season-specific features of nutrients that might be used as an advantage for
compliance testing. The reasoning is following. Notably nitrogen concentrations are normally
lower in the period late spring early autumn. This is more or less the same period during
which one could expect eutrophication phenomena actually to happen. The suggestion would
be to apply the nutrient EQSs only for that period during which eutrophication actually can
happen. This implies that compliance then would be checked during the period with overall
lower concentrations in the water body (compared to the winter period). It can be noticed that
such a principle is implemented two of the systems mentioned in chapter 2. The SEPA has
defined an EQS for the May-October average concentration for Ntot and Ptot in lakes. The
Netherlands applies the EQS to the April-September average concentrations of Ntot and Ptot in
standing waters.
This approach seems not to be compatible when putting the Black Sea into the perspective
(compare section 2.6).
Such an approach further cannot apply to NH4 anyway, since for ammonium the potential
toxicity is relevant and therewith the higher (90-percentile) values.
7.6 Concentration and loads
Relationships between concentrations, flows, and loads actually can be rather complicated, and go
beyond the scope of the present study.
The mentioning of loads in the report has merely been made in order to relate the discussion of
formulating EQSs for nutrients in the freshwater part of the Danube to the Black Sea. It is expected
that the EQSs formulated for the Danube itself somehow are to be consistent with reaching the quality
objectives for the (north-western part of the) Black Sea. Assuming the Danube itself as one of the
water bodies, and having formulated EQSs for phosphorous (plus possibly nitrogen) for his water
body, then it is not automatically obvious how this water quality status will affect the water quality in
the (north-western part of the) Black Sea. In order to conduct such assessments, one needs to work
with Danube pollution loads discharged into the Black Sea (and with computer models).
Maybe such an exercise is not necessary. Discussions in the Black Sea community seem to gear
towards agreeing that if the pollution of the Danube would be comparable the situation in the early
1960-ies this would be satisfactory with respect to no eutrophication in the Black Sea. Of course, this
implies that one there are historic data for the nutrient concentrations in the Danube in the early 1960-
ies one can use to agree upon.
12 Part of the confusion may arise from the comparisons in this report with annual averaged TNMN 1995-2000 results.
Comparing with annual average TNMN data in the underlying report merely has been done because of illustration purposes;
details of the TNMN results are included in other reports.
RODECO Consulting GmbH
Orientation on environmental quality standards for nutrients and other Danube specific priority substances
I 49
7.7 Closing remarks
From the previous sections one can conclude that there is still quite some work to be done in order to
formulate EQSs for the Danube specific priority substances; also discussions definitely are not yet
completed. Nevertheless, the author expects that the information and the experiences gathered during
this study indeed will facilitate the expert groups in continuing and structuring their activities
concerning this topic.
I 50
UNDP/GEF Danube Regional Project
References
Adamkovį, Juliana; Hamchevici, Carmen; Litheraty, Peter; Makovinskį, Jarmila; Rauchüchl, Alfred;
Wolf, Birgit (2003) "Five-years Report on Water Quality in Danube River Basin Based on
TransNational Monitoring Network - 1996-2000". UNDP-GEF Danube Regional Project,
ICPRD, Vienna.
Behrendt, Horst (2003) personal communication.
Breukel, R.M.A., (1993) "Natuurlijke achtergrondgehalten van de Nederlandse rivieren, een keuze uit
de literatuurgegevens." (Dutch: Natural background contents of Dutch rivers, a selection from
literature data.) RIZA werkdocument 93.193x, Lelystad, the Netherlands.
EEA (2001)
"Environmental Signals 2001" European Environmental Agency, Copenhagen,
Denmark.
EEIC (2000)
"State of the Environment in Estonia" http://nfp-ee.eionet.eu.int/SoE/index_en.htm.
EMECS (2003) "Features of Japanese Environmental Water Quality Standards". International Center
for the Environmental Management of Enclosed Coastal Seas.
http://www.emecs.or.jp/01cdrom/section_3_e/sec3_a_ro_b_6_e.html.
EPA (1999)
"1999 Update of Ambient Water Quality Criteria for Ammonia", United States
Environmental Protection Agency, EPA-822-R-99-014, December 1999.
EPA (2002)
"National Recommended Water Quality Criteria: 2002". United States Environmental
Protection Agency, EPA-822-R-02-047, November 2002.
EPA (2003)
"Ecoregional Nutrient Criteria" http://www.epa.gov/ost/standards/nutrient.html.
Environment Canada (1999a) "A protocol for the derivation of Water Quality Guidelines for the
Protection of Aquatic Life." Environment Canada, Guidelines and Standards Division; Hull,
Canada.
Environment Canada (1999b) "A protocol for the derivation Canadian Sediment Quality Guidelines
for the Protection of Aquatic Life." Environment Canada, Guidelines and Standards Division;
Hull, Canada.
Environment Canada (2002)
"Summary of Existing Canadian Environmental Quality Guidelines"
(document can be downloaded from
http://www.ec.gc.ca/ceqgrcqe/English/download/default.cfm).
Haskoning (1994)
"Pollutant loads into the Danube from the Danube tributaries, surface water
quality of the Danube tributaries, and identified major point pollution sources in the Danube
basin". Draft, February 1994, Haskoning, Nijmegen, The Netherlands.
RODECO Consulting GmbH
Orientation on environmental quality standards for nutrients and other Danube specific priority substances
I 51
ICPDR (2002) "Joint Danube Survey. Technical Report of the International Commission for the
Protection of the Danube River." International Commission for the Protection of the Danube
River Basin, Vienna, September 2002.
IKSR (1992) "Tätigkeitsbericht 1991. Anlage 1.3.3 Konzept zur Ausfüllung des Punktes A.2 des
ASPR (1. Zielvorgaben)." Internationale Kommission zum Schutze des Rheins gegen
Verunreinigung, 1992, Koblenz, Germany.
IKSR (1993a) "Zusammenstellung von Information über die geogenen Stoffgehalte von Sedimenten
und Schwebstoffen" Internationale Kommission zum Schutze des Rheins gegen
Verunreinigung, Ps 45/92 rev. 14.01.93, Koblenz, Germany.
IKSR (1993b) "Vereinbarungen der IKSR für Messprogramme under Sonderuntersuchungen in den
Teilbereichen Wasser, Schwebstoff, Sedimente und Organismen. Teil E: Grundprinzipien zur
Messtechnischen Überprüfung der Zielvorgaben " Internationale Kommission zum Schutze
des Rheins gegen Verunreinigung, P 30E/93 rev. 15.12.93, Koblenz, Germany.
Kroiss, H; Zessner, M.; Lampert, C. (2002)
"Nutrient Management in the Danube Basin and Its
Impact on the Black Sea" Institute for Water Quality and Waste Management, Vienna
University of Technology, Wien, Austria. http://danubs.tuwien.ac.at/Publications/8.doc.
Meybeck, M.; Chapman, D.V.; Helmer, R. (editors)(1989)
"GEMS Global Monitoring System.
Global freshwater quality; a first assessment" WHO/UNEP. Basil Blackwell Ltd., Oxford,
UK.
NDWR, 1999 "Dictionary: Technical Water, Water Quality, Environmental, and Water-Related
Terms." Nevada Division of Water Resources, http://water.nv.gov/Water%20planning/dict-
1/wwindex.htm.
Pope, L.M. (2002)
"Significant Findings of Water-Quality Studies and Implications for Cheney
Reservoir Watershed, South-Central Kansas, 19962001"
http://ks.water.usgs.gov/Kansas/pubs/fact-sheets/fs.009-02.pdf.
Reynolds, P (2001)
"Environmental Quality Objective for Bug River Basin" Published in the
report series of the Tacis "Bug Transboundary Water Quality Monitoring and Assessment
between Belarus and Poland" project. ICWS, the Netherlands.
Riet, Valentijn van `t (1998)
"An input-output study on nutrient loads in the Danube River basin"
Landbouwuniversiteit Wageningen, AOWK-afstudeerverslag nr. 019/98, Wageningen, The
Netherlands.
RIZA, (1989) "Kansen voor waterorganismen. Een ecologische onderbouwing voor
kwaliteitsdoeltellingen voor water en water bodem. Deel 1: Resultaten en berekeningen."
DBW/RIZA nota nr. 89.016a, april 1989, Lelystad, The Netherlands.
Schreiber, Heide; Constantinescu, Lucian Theodor; Cvitanic, Irena; Drumea, Dumitru; Jabucar, Dalila;
Juran, Stanislav; Pataki, Beata; Snishko, Sergej; Zessner, Matthias; Behrendt, Horst (2003)
"Harmonised Inventory of Point and Diffuse Emissions of Nitrogen and Phosphorus for a
Transboundary River Basin", Durchführende Institution Leibniz-Institut für Gewässerökologie
und Binnenfischerei im Forschungsverbund Berlin, Research Report 200 22 232, Berlin.
I 52
UNDP/GEF Danube Regional Project
SEPA (2002) Environmental Quality Criteria for Lakes and Watercourses. Nutrients /
Eutrophication.
http://www.internat.naturvardsverket.se/documents/legal/assess/assedoc/lakes.htm
UK/EA (2001) "Managing aquatic eutrophication" The Environment Agency, United Kingdom.
http://www.environment-agency.gov.uk/commondata/105385/eutrophication.pdf.
UN/ECE (1992)"ECE Standards Statistical Classification of Surface Water Quality for the
Maintenance of Aquatic Life" United Nation, Economic and Social Council, Statistical
Commission and Economic Commission for Europe. Conference of European Statisticians.
CES/733, 13 April 1992.
US Dept. (2002)
"Circular 1207". U.S. Department of the Interior, U.S. Geological Survey,
Center for Coastal Geology,
http://sflwww.er.usgs.gov/publications/circular/1207/nutrients.html.
Veldstra, A.W.F., Zuurdeeg, B.W. (1989)
"Nutriėnten en PCA's in natuurlijk Rijnwater"
(Dutch, "Nutrients and PAHs in natural Rhine water"), GEOCHEM-research,
Rijksuniversiteit Utrecht, the Netherlands.
Vituki (1997) "Water Quality Targets and Objectives for Surface Waters in the Danube Basin",
Phare project EU/AR/203/91, final report, Water Resources Research Centre Plc. VITUKI,
Budapest, Hungary, September 1997.
NIMRD ()
"Report on the ecological indicators of pollution in the Black Sea, Romania. Danube
River Pollution Reduction Programme and the Black Sea Environmental Programme."
Romanian Marine Research Institute, Constanta.
VW, 2000
"Gewijzigde versie Bijlage A: Normen 4e Nota Waterhuishouding". (Dutch: Adjusted
version Appendix A: Standards 4th Note Watermanagement") Ministerie voor Verkeer en
Waterstaat. Staatscourant, 16 juni 2000, nr. 114, page 18. Staatsuitgeverij, Den Haag, The
Netherlands.
RODECO Consulting GmbH