February 2007
PROVISION OF TECHNICAL SUPPORT ON
DANUBE NUTRIENTS
FINAL REPORT
AUTHORS
PREPARED BY:
Vienna University of Technology
Institute for Water Quality, Resources and Waste Management
AUTHORS:
Helmut Kroiss
Matthias Zessner
Christoph Lampert
Vienna University of Technology
Institute for Water Quality, Resources and
Waste Management
Karlsplatz 13
1040 Vienna
PREFACE
The Vienna University of Technology, Institute for water Quality, Resources and Waste
Management (IWA) has been contracted by the UNDP-GEF Danube Regional Project to carry out
activities within the DRP's objective of "Reinforcement of monitoring, evaluation and information
systems to control transboundary pollution, and to reduce nutrients and harmful substances".
The main objective of this activity is to provide the UNDP/GEF DRP and the ICPDR Secretariat
technically robust analyses of the nutrient state in the Danube River Basin and the Black Sea
North West Shelf. These analyses are built on the earlier work of daNUbs project using readily
available data where possible.
Three separate parts of work have been carried out (chronological order):
1.
Support of Stefan Speck for a study on cost efficient mitigation of nutrient
emissions
2.
Comparison of current levels of nutrients in the Danube River with the 1997 levels
and an assessment of the dissolved oxygen levels in the NW shelf of the Black Sea
3.
Basic considerations on the introduction if phosphate free detergents in the
Danubian countries
Further IWA participated in the Stakeholder Seminar "Phosphates in detergents: Latest
developments in the Danube River Basin" in Bucharest on the 25 January 2007. The results of
scenario calculations in respect to the use of P-containing detergents and the full
implementation of the Urban Waste Water Treatment Directive were presented.
The objective was to raise awareness of the impact of the nutrients from laundry detergents and
to discuss options to minimise their environmental impacts.
Danube nutrients
page 5
TABLE OF CONTENTS
1.
INTRODUCTION ................................................................................................ 10
2.
CURRENT NUTRIENT LEVELS IN THE DANUBE ....................................................... 11
2.1.
Nitrogen ...................................................................................................... 11
2.1.1.
Dissolved Inorganic Nitrogen (DIN) .............................................................. 11
2.1.2.
Ammonium ............................................................................................... 14
2.2.
Phosphorus .................................................................................................. 14
2.3.
Updated figures with trends since 1988 ............................................................ 18
3.
Dissolved oxygen levels in the NW-shelf of the Black Sea........................................ 19
4.
Basic considerations on the introduction of phosphate free detergents ...................... 22
4.1.
Scenario development.................................................................................... 22
4.1.1.
Scenario development basic assumptions ...................................................... 23
4.1.2.
Scenario results......................................................................................... 24
4.1.3.
Reliability and Uncertainties of Scenario calculations ....................................... 26
4.2.
Aspects of the management of the limited resource phosphate............................. 27
4.3.
Costs of sludge management and P-removal ..................................................... 28
4.4.
Different sources - different forms of P ............................................................. 30
5.
Issues for innovative mechanisms for cost efficient mitigation of nutrient emissions.... 32
6.
SUMMARY of MAIN RESULTS............................................................................... 35
LIST OF ANNEXES
ANNEX 1: Results of scenario calculations
ANNEX 2: Definition of technical terms
LIST OF TABLES
Table 1: Mean of Scenario calculations for emissions from municipal wastet water treatment
plants in tP/a .........................................................................................................39
LIST OF PICTURES AND GRAPHS
Figure 1: Mean P-discharge in different scenarios in tP/a. ....................................................9
Figure 2: TIN and NH4-N-loads and Danube discharges at Reni from 1996-2005 ...................11
Figure 3: TIN-concentrations and temperature in the Danube River at Reni ..........................12
Figure 4: TIN-loads and river discharges of the Danube River at Reni ..................................12
Figure 5: Yearly TIN loads and Danube discharges at Reni from 1996-2005..........................13
Figure 6: NH4-N-loads and river discharges of the Danube River at Reni ..............................14
Figure 7: TP and PO4-P-loads and Danube discharges at Reni form 1996-2005......................15
Figure 8: TP-loads and river discharges of the Danube River at Reni....................................15
Figure 9: TP-loads and river discharges of the Danube River at Reni....................................16
Figure 10: Yearly TP-loads and Danube discharges at Reni form 1996-2005 .........................17
Figure 11: PO4-P-loads and river discharges of the Danube River at Reni .............................17
UNDP/GEF DANUBE REGIONAL PROJECT
page 6
Figure 12: River Danube annual total phosphorus loads (corrected for annual discharge) to the
Black Sea ..............................................................................................................18
Figure 13 : River Danube annual inorganic nitrogen loads (corrected for annual discharge) to
the Black Sea.........................................................................................................18
Figure 14: Monitoring locations in the Western Black Sea Shelf (WBS): ................................19
Figure 15: Oxygen saturation at bottom of WBS, station Sf. Georghe...................................20
Figure 16: Oxygen saturation at bottom of WBS, station Zaton ...........................................20
Figure 17: Oxygen saturation at bottom of WBS, station Constanta .....................................21
Figure 18: Mean P-discharge in different scenarios in tP/a. .................................................25
Figure 19: P-emissions from WWTPs of the 3 variants considered........................................25
Figure 20: Amount of P in various P-flows in comparison with the P-fertilizer consumption in
Austria 2004 ..........................................................................................................28
Figure 21: Increase of sludge dry matter due to P-precipitation in treatment plants above
10.000 pe or due to the replacement of P-containing laundry detergents by Zeolite in t/a
(mean of the variants are used)................................................................................29
Figure 22: Additional need for precipitants of 3 variants in % of the "Zero laundry detergent
Scenario" (0.3 g P/inh.d from detergents) ..................................................................30
Figure 23: Emission to the river system and discharge to the Black Sea ...............................32
Figure 24: P-flow at high flow: Danube at Vienna..............................................................34
Figure 25: P-emissions in different settlement categories and different variants ....................40
Figure 26: P-discharge of Variant 1: The effluent concentration in area A is 1 mgP/l, in Area B 2
mgP/l. The infiltration rate amounts to 100 l/inhabitant. In the area C biological treatment is
applied..................................................................................................................41
Figure 27: P-discharge of Variant 2: The effluent concentration in area A is 1 mgP/l, in Area B 2
mgP/l. The infiltration rate amounts to 200 l/inhabitant. In the area C biological treatment is
applied..................................................................................................................41
Figure 28: P-discharge of Variant 3: 80% P-reduction in relation to the inflow load. In the area
C biological treatment is applied. ..............................................................................42
Figure 29: Emissions from WWTPs assuming "average" values for the percentage of the
population in settlements from 2.000 10.000 and below 2.000 for Romania (mean of the 3
Variants). ..............................................................................................................42
VIENNA UNIVERSITY OF TECHNOLOGY
Danube nutrients
page 7
ABBREVIATIONS
CEE
Central and Eastern Europe
ICPDR
International Commission for the Protection of the Danube River
IWA
Institute for Water Quality, Resources and Waste Management, Vienna
University of Technology
Q
Discharge of the river
MQ Mean
discharge
STTP Sodium
tripolyphosphate
T Temperature
TIN
Total Inorganic Nitrogen
TNMN
Transnational monitoring network
UWWTD
Urban Waste Water Treatment Directive
Var Variant
WWTP
Waste water treatment plant
UNDP/GEF DANUBE REGIONAL PROJECT
Executive Summary
page 8
EXECUTIVE SUMMARY
Main objective of the services carried out was to provide the UNDP/GEF DRP and the ICPDR
Secretariat technically robust analyses of the nutrient state in the Danube River Basin and the
Black Sea North West Shelf.
The analyses are built on the earlier work of daNUbs project using readily available data where
possible.
Three separate parts of work have been carried out (chronological order):
> Comparison of current levels of nutrients in the Danube River with the 1997 levels and
an assessment of the dissolved oxygen levels in the NW shelf of the Black Sea
> Support of Stefan Speck for a study on cost efficient mitigation of nutrient emissions
> Basic considerations on the introduction if phosphate free detergents in the Danubian
countries
To derive cost efficient measures it has to be taken into account that emissions are modified via
their transport from the area of emission to the area of discharge (denitrification, adsorption).
Emissions from waste water treatment plants are mainly in the form of dissolved Phosphate
which is immediately highly available for algae production. P-emissions via erosion are mainly in
particulate form which is only partly available for algae-production. Particulate P- emissions
provide a potential P-source which can be partly mobilised in the case of anaerobic conditions in
the waters which are a result of eutrophication.
A mere focus on the reduction of the nutrient-discharge to the WBS could lead to a deterioration
of smaller rivers or local groundwater. A comprehensive strategy as outlined in the Water
Framework Directive including the quality of surface waters as well as ground water should be
aimed at.
Transported nutrient loads in River Danube are highly influenced by physical factors as water
discharge and temperature. The influence of these physical factors on yearly loads may
outweigh anthropogenic impacts by far. Relatively high nutrient loads in 2005 are a result of the
relatively high flow situation of this year.
Eliminating the two effects temperature and annual discharge a slight decrease in the Dissolved
inorganic Nitrogen-loads transported in the Danube can be detected between 1996 and 2005.
TP-loads are heavily influenced by high flow events. A comparison of different time periods has
to consider high flow events above average. Between 1997 and 2004/2005 no significant
difference in total P-loads can be detected. The PO4-P loads slightly decreased in the period
2004 2005 as compared to the period 1996-1998 (about 10 % of the yearly load).
The level of oxygen saturation at the bottom of WBS has significantly improved between the
80ies and 1996. No dramatic changes in the period between 1996 and 2003 can be interpreted
from existing data. The data base is not appropriate to demonstrate changes between 1996 and
2003 in detail.
Scenario calculations have been carried out to show the effects of different consumption of P-
containing detergents on the P-emissions from waste water treatment plants assuming full
VIENNA UNIVERSITY OF TECHNOLOGY
Danube nutrients
page 9
implementation of the Urban Waste Water Directive and the whole Danube Basin is considered
as sensitive area.
If the average consumption of P-containing detergents exceeds 0.65 gP/inh.d (including
automatic dish washing detergents) the emission will be higher as in the year 2004.
55 to 70% of the P-emission will stem from settlements between 2.000 and 10.000 inhabitants
as for these areas no P-removal is required by the UWWTD.
40000
35000
30000
/
a
discharge 2000
t
P 25000
in
e 20000
r
g
ha 15000
c
i
s
d 10000
5000
0
> 100.000 inh.
10.000 -
2000 - 10.000
< 2000 inh.
Total
100.000
0.3 gP/i.d
1 gP/i.d
1.75 gP/i.d
2.5 gP/i.d
Figure 1: Mean P-discharge in different scenarios in tP/a.
Sewage sludge represents a considerable source of P.
The amounts of additional sludge production due to the replacement of P containing detergents
by e.g Zeolites or due to the precipitation of P are similar (increase by 10 - 20%).
The costs of precipitants for P-removal compared to the total costs (investment costs plus
operation costs) for sewer development and waste water treatment are very small.
UNDP/GEF DANUBE REGIONAL PROJECT
Introduction
page 10
1. INTRODUCTION
The Vienna University Institute for Water Quality Resources and Waste management was
contracted to provide the UNDP/GEF DRP and the ICPDR Secretariat with technically robust
analyses of the nutrient state in the Danube River Basin and the Black Sea North West Shelf.
These analyses should be built on the successful earlier work of daNUbs project using readily
available data where possible.
The following main activities were carried out within this assignment:
> Comparison of current levels of nutrients in the Danube River with the 1997 levels
> To provide as up-to-date as possible assessment of the dissolved oxygen levels in the
NW shelf of the Black Sea
> Basic considerations on the introduction of phosphate free detergents in the Danubian
countries including scenario calculations considering the introduction of P-free or
reduced-P detergents and the full implementation of the Urban Waste Water
Treatment Directive
> Support of Stefan Speck for a study on cost efficient mitigation of nutrient emissions
Further IWA participated
> in the Stakeholder Seminar "Phosphates in detergents: Latest developments in the
Danube River Basin" in Bucharest on the 25 January 2007. The objective was to raise
awareness of the impact of the nutrients from laundry detergents and to discuss
options to minimise their environmental impacts. The results of scenario calculations in
respect to the use of P-containing detergents and the full implementation of the Urban
Waste Water Treatment Directive were presented by IWA.
> in the Regional Conference on Nutrient Pollution Control in the Danube-Black Sea
Basin held on 36 October 2006 in Chisinau, Moldova Objective was to facilitate the
sharing of lessons and experiences in the Danube-Black Sea Strategic Partnership
Program. A special focus of the Conference was to develop an indicator framework for
investments that could be used across similar Global Environment Facility-sponsored
Strategic Partnerships being developed and implemented in the Mediterranean, Africa,
and East Asia.
VIENNA UNIVERSITY OF TECHNOLOGY
Danube nutrients
page 11
2. CURRENT NUTRIENT LEVELS IN THE DANUBE
Sources:
Data from 1996 2002: TNMN database
Data from 2003 2005: Romanian data from station Reni
2.1. Nitrogen
Figure 2 shows the development of daily discharge and total inorganic nitrogen loads (TIN =
NO3-N + NO2-N + NH4-N) as well NH4-N-loads at measuring days from 1996 to 2005. The high
fluctuation of river discharge and the high fluctuation of nitrogen loads are evident. In addition
to anthropogenic pressure (discharges from municipal and industrial point sources, agricultural
activities and air pollution by burning processes) nitrogen loads in the Danube River depend to a
high degree on physical factors.
16000
4000
Q
NH4-N - load
TIN-load
12000
3000
)
/s
)
3
d
m
8000
2000
(t/
Q (
N
4000
1000
0
0
01.01.96 31.12.96
31.12.97 01.01.99 01.01.00
01.01.01 01.01.02 01.01.03 02.01.04
01.01.05 02.01.06
Figure 2: TIN and NH4-N-loads and Danube discharges at Reni from 1996-2005
2.1.1.
Dissolved Inorganic Nitrogen (DIN)
A clear relation can be shown between the water temperature of the river and the TIN
concentrations (influence of denitrification in the river, see Figure 3) and between the river
discharge and the river load of TIN (influence of nitrogen discharge via groundwater and surface
runoff, see Figure 4). The short term influence of these physical factors on TIN loads in river
Danube can outweigh anthropogenic influences by far. Thus, the evaluation of changes in
anthropogenic pressure based on the comparison of yearly loads can be misleading.
UNDP/GEF DANUBE REGIONAL PROJECT
Current nutrient levels in the Danube
page 12
5
2004-2005
4.5
1996-1998
4
Trend 1996 - 1998
3.5
Trend 2004-2005
3
/
l
)
g
2.5
R2 = 0.43
I
N (m
T
2
R2 = 0.52
1.5
1
0.5
0
0
5
10
15
20
25
30
T (°C)
Figure 3: TIN-concentrations and temperature in the Danube River at Reni
3000
2004-2005
2500
1996-1998
Trend 2004-2005
R2 = 0.66
2000
Trend 1996-1998
t
/
d)
R2 = 0.63
( 1500
N
TI
1000
500
0
0
2000
4000
6000
8000
10000
12000
14000
16000
Q (m3/s)
Figure 4: TIN-loads and river discharges of the Danube River at Reni
Based on the method of load calculation agreed within ICPDR an increasing TIN load between
2003 and 2005 is detected. This increase is due to the increasing water discharge during this
VIENNA UNIVERSITY OF TECHNOLOGY
Danube nutrients
page 13
period. If the influence of Q and T on the calculation of the development of TIN loads over the
years is eliminated by calculation based on the functions in Figure 3 and Figure 4, the
development of loads shows a slightly decreasing tendency.
Calculative elimination of Q was done based on
TIN-loadQ,i = TIN-load ICPDR,i . MQ1996-2005/MQi with
TIN-loadQ,i .....
TIN-load of the year i with calculative elimination of the influence of
fluctuations of MQ
TIN-load ICPDR,i ....
TIN-load of a year i calculated with the method agreed upon at the
ICPDR
MQ1996-2005 ....
Mean flow of the period 1996-2005
MQi ....
Mean flow of the year i
Calculative elimination of T was done based on the relation shown in Figure 3 with the following
equation:
'c14,4; i = ci + (Ti 14,4) . 0,057
'c14,4; i ....
TIN concentration at the time i adjusted to a temperature of 14,4°C
14,4 °C ....
Mean temperature of the period 1996-2005
ci ....
Measured TIN concentration at the time i
Ti ....
Measured temperature at the time i
Load calculations have been repeated based on the "normalised" concentration c14,4; i.
600
12000
500
10000
400
8000
a
)
t/3
)
1
0
/s3
( 300
d
6000
m
(
-
loa
MQ
TIN 200
TIN-load - ICPDR mehtod
4000
TIN-load, influence of Q calculative eliminated
100
2000
TIN-load, influence of Q and T calculative eliminated
Mean flow of the year
0
0
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
Figure 5: Yearly TIN loads and Danube discharges at Reni from 1996-2005
Despite the fact, that due to high Danube flows the transported TIN-loads have been relatively
high in 2005, the Figure 3 to Figure 5 indicate that for comparable flow and temperature
UNDP/GEF DANUBE REGIONAL PROJECT
Current nutrient levels in the Danube
page 14
situations a slight decrease of nitrogen levels can be detected from the period 1996-1998 to the
period 2004-2005. The TIN concentrations at higher temperatures as well as the TIN-loads at
higher flows are slightly lower in the period 2004-2005. After calculative elimination of the
influence of differences in Q and T between the periods the TIN-load in the period 2004-2005 is
about 10 % lower than in the period 1996-1998, which still can not be seen as a significant
trend. Changes of anthropogenic pressure at this level of difference only can be evaluated based
on emission assessment.
2.1.2.
Ammonium
If the Ammonium loads at different days are plotted against the flow (Figure 6), again a slight
decrease of loads in the period 20042005 compared to the period 1996-1998 can be detected.
Expressed as yearly loads this decrease is about 10 % of the yearly load.
500
2004-2005
450
1996-1998
400
Linear (2004-2005)
350
Linear (1996-1998)
) 300
t
/
d
250
-
N (
4
200
NH
150
100
50
0
0
2000
4000
6000
8000
10000
12000
14000
16000
Q (m3/s)
Figure 6: NH4-N-loads and river discharges of the Danube River at Reni
2.2. Phosphorus
Similar as the TIN-loads, the TP loads show a high fluctuation (Figure 7). In contrary to the TIN-
loads there is no linear but a disproportionate relation between loads and flow (Figure 8). High
flow situations lead to a mobilization of suspended solids which highly influences the transport
of TP. The TP load is significantly higher at increasing water levels compared with falling water
levels. The TP loads at high flow situations are not directly related to anthropogenic emissions.
Yearly TP-loads are highly influenced by TP transport at high flow and thus strongly depend on
the number and intensity of high flow situations.
VIENNA UNIVERSITY OF TECHNOLOGY
Danube nutrients
page 15
15000
500
Q
PO4-P - load
12000
TP-load
400
)
9000
300
/s3
d)
m
(t/
P
Q (
6000
200
3000
100
0
0
01.01.96 31.12.96 31.12.97 01.01.99 01.01.00 01.01.01 01.01.02 01.01.03 02.01.04 01.01.05 02.01.06
Figure 7: TP and PO4-P-loads and Danube discharges at Reni form 1996-2005
500
2004-2005
450
1996-1998
400
350
Trend 2004-2005
300
Trend 1996-1998
t
/
d)
( 250
TP
R2 = 0.38
200
150
100
50
R2 = 0.19
0
0
2000
4000
6000
8000
10000
12000
14000
16000
Q (m3/s)
Figure 8: TP-loads and river discharges of the Danube River at Reni
TP loads of the period 1996-1998 compared to the TP loads of the period 2004-2005 at
corresponding flow situations show no significant differences (figure 8). However the picture for
1999- 2001 is completely different. TP loads do not rise in the same relation to the increasing
flow as in the other periods. As already indicated during the daNUbs project (2005) the TP-
UNDP/GEF DANUBE REGIONAL PROJECT
Current nutrient levels in the Danube
page 16
measurements of the TNMN-dataset at Reni for the period beginning with 2000 show
significantly lower values as compared to other stations and other data sources. In this project
the hypothesis was that these values are measured from filtered sample only.
250
2004-2005
1999-2001
200
1996-1998
Trend 1999-2001
Trend 2004-2005
150
Trend 1996-1998
)
t
/
d
(
TP 100
50
0
0
2000
4000
6000
8000
10000
12000
14000
16000
Q (m3/s)
Figure 9: TP-loads and river discharges of the Danube River at Reni
The yearly loads calculated based on the method agreed upon at the ICPDR show a sharp
increase in 2005. This increase is due to an increase of suspended solid transport at high flow
situations of this year. To eliminate this influence of high flow events for a comparison of the
TP-levels in 2004-2005 to the levels of 1997, the load calculation (Figure 10) was done a second
time based on data from flow situations with less than 7000 m3/s only. Based on these
calculations it can be seen that there is no significant difference in loads between 1997 and the
values of the period 2004-2005. Very low TP loads in the period 2000 2001 probably are
based on measurements representing the TP-concentration of filtered samples only (instead of
unfiltered samples in the years before and afterwards).
VIENNA UNIVERSITY OF TECHNOLOGY
Danube nutrients
page 17
45
12000
40
10000
35
30
8000
a
)
t/
)
3
25
/s
1
0
3
6000
m
a
d ( 20
-
l
o
MQ (
TP 15
4000
10
TP-load, ICPDR method
2000
5
TP-load, influence of high flow events excluded
mean flow of the year
0
0
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
Figure 10: Yearly TP-loads and Danube discharges at Reni form 1996-2005
If the PO4-P loads at different days are plotted against the flow (Figure 11), a slight decrease of
loads in the period 2004 2005 as compared to the period 1996-1998 can be detected.
Expressed as yearly loads this decrease is about 10 % of the yearly load.
120
2004-2005
100
1996-1998
trend 2004-2005
80
Trend 1996-1997
)
t
/
d
60
-
P (
4
PO
40
R2 = 0.13
20
R2 = 0.37
0
0
2000
4000
6000
8000
10000
12000
14000
16000
Q (m3/s)
Figure 11: PO4-P-loads and river discharges of the Danube River at Reni
UNDP/GEF DANUBE REGIONAL PROJECT
Current nutrient levels in the Danube
page 18
2.3. Updated figures with trends since 1988
80
60
)
t/y
(k 40
P
T
20
0
9
88
89
90
91
92
93
94
95
96
97
98
99
00
01
02
03
04
05
95
-
1
19
19
19
19
19
19
19
19
19
19
19
19
20
20
20
20
20
20
48
19
TP (Q corr.) kt/y
TP kt/y Almazow
Figure 12: River Danube annual total phosphorus loads (corrected for annual
discharge) to the Black Sea
800
600
t
/
y
)
k 400
N (
DI
200
0
59
88
89
90
91
92
93
94
95
96
97
98
99
00
01
02
03
04
05
19
19
19
19
19
19
19
19
19
19
19
19
19
20
20
20
20
20
20
48-
19
DIN (Q corr.) kt/y
DIN kt/y Almazow
linear trend
Figure 13 : River Danube annual inorganic nitrogen loads (corrected for annual
discharge) to the Black Sea
VIENNA UNIVERSITY OF TECHNOLOGY
Danube nutrients
page 19
3.
DISSOLVED OXYGEN LEVELS IN THE NW-SHELF OF THE
BLACK SEA
Source: Cosiasu et al. (2003-2005) Deliverables: D7.1 Romanian report; D7.3 Romanian annual
reports 2001-2002; and D7.6 "Summary report on field and laboratory work in 2001-2003 in
comparison with previous observations in the Western Black Sea" from the project "Nutrient
Management in the Danube Basin and its Impact on the Black Sea" supported under contract
EVK1-CT-2000-00051 by the Energy, Environment and Sustainable Development (EESD
Programme of the 5th EU Framework Programme, http://danubs.tuwien.ac.at)
Figure 14: Monitoring locations in the Western Black Sea Shelf (WBS):
SG ... Sf. Georghe; Z ... Zaton, Ct ... Constanta, Mg ... Mangalia
UNDP/GEF DANUBE REGIONAL PROJECT
Dissolved Oxygen levels in the NW Shelf of the Black Sea
page 20
Oxygen saturation at bottom of WBS
station Sf. Georghe
160
140
120
100
80
60
y
g
en saturation [%]
40
Ox
distance to coast: 6km
distance to coast: 18km
20
distance to coast: 30km
0
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
Figure 15: Oxygen saturation at bottom of WBS, station Sf. Georghe
Oxygen saturation at bottom of WBS
station Zaton
160
140
120
]
[% 100
tion
ra
80
tu
60
y
g
en sa
40
distance to coast: 10km
Ox
distance to coast: 14km
distance to coast: 24km
20
distance to coast: 32km
0
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
Figure 16: Oxygen saturation at bottom of WBS, station Zaton
VIENNA UNIVERSITY OF TECHNOLOGY
Danube nutrients
page 21
Oxygen saturation at bottom of WBS
station Constanta
140
120
] 100
[%
n
80
tio
r
a
tu
a
60
s
distance to coast: 2km
40
distance to coast: 12km
x
y
g
e
n
O
distance to coast: 22km
20
distance to coast: 46km
distance to coast: 70km
0
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
Figure 17: Oxygen saturation at bottom of WBS, station Constanta
Figure 15 to Figure 17 show the development of oxygen saturation at the bottom of the Western
Black Sea shelf area (WBS) of the transects St. George, Zaton and Constanta from 1996 to
2003. These figures indicate an improvement of oxygen saturation in this period. Going more in
detail it becomes clear that the measurement from 1996 was done in September, representing
the season at which in the years 2001-2003 the lowest oxygen saturation appeared as well. If
only the autumn values are compared only for the stations very close to the coast (2 resp. 6 or
10 km distance) an increase can be seen in oxygen saturation between 1996 and 2003.
Especially the oxygen saturation at Constanta station has not improved compared to 1996.
Nevertheless, the improvement since the 80ies, where large zones of anoxia (bottom oxygen
saturation = 0 %) have been detected, is obvious.
UNDP/GEF DANUBE REGIONAL PROJECT
Basic Considerations on the Introduction of P-free Detergents
page 22
4. BASIC CONSIDERATIONS ON THE INTRODUCTION OF
PHOSPHATE FREE DETERGENTS
Eutrophication is of major concern in the Danube Region and especially in the receiving Sea, the
Western Black Sea. The ecological situation in the Black Sea has improved considerably in the
last decade (reduced eutrophication, disappearance of anoxic conditions, regeneration of zoo-
benthos and phytoplankton). However the improvement was only partly due to the effect of
measures of environmental policy, like nutrient removal at waste water treatment plants
(WWTPs) or the ban of P-containing laundry detergents. To a considerable part the reduction is
caused by the economic crises in several countries in Central and Eastern Europe.
There are two major developments that endanger the improvements in the Western Black Sea
observed, i.e. which will lead to an increase of nutrient emissions:
> The economy in these countries will redevelop in the coming years
> The (full) implementation of the Urban Waste Water Treatment Directive (UWWTD).
The challenge towards policy nowadays is to enable economic development without increasing
nutrient emissions again (as in the 70ies and 80ies) above "critical loads" for the Western Black
Sea. This means: efforts have to be taken to provide space for the anticipated increase of
nutrient emissions. Otherwise the ecology of the Black Sea will deteriorate again.
The introduction of P-free laundry detergents is considered to be a fast and efficient measure to
reduce nutrient emissions into surface waters.
The following 3 basic considerations are evaluated more in detail in the following chapters:
1.
Scenario development for nutrient emissions via waste water treatment plants in
respect to the amounts of detergents used
2.
Aspects of the management of the limited resource phosphate
3.
Costs of waste water treatment: P-removal, sludge management
4.1. Scenario development
Already in the daNUbs project several scenarios have been developed. Some main differences to
the scenarios developed below shall be depicted:
> In the daNUbs scenarios also an increase of the economy was assumed.
> In the scenarios developed now, a further distinction was made between the sizes of
the settlements. In total 4 categories were defined: A = settlements > 100.000
inhabitants; B = settlements 10.000 - 100.000 inhabitants; C = settlements 2.000
10.000 inhabitants; D = settlements < 2.000 inhabitants. The classification is based
on the figures given in the study ("Nutrient balances for Danube countries", PHARE
EU/AR/102A/91). The former Yugoslavia was not included in that study. For Serbia
relevant data was retrieved from the ICPDR webpage (http://www.icpdr.org/icpdr-
pages/serbia.htm), for Bosnia and Croatia it was assumed to have the same
distribution as Serbia. For Romania statistical data for the year 2004 was used.
VIENNA UNIVERSITY OF TECHNOLOGY
Danube nutrients
page 23
4.1.1. Scenario development basic assumptions
The scenarios developed now are based on the following basic assumptions:
> Connection to sewer systems: All areas except those below 2000 inhabitants are
completely (100%) connected to sewer system and corresponding waste water
treatment plants. For Germany in addition 77% and for Austria 36% of these areas
(reflecting the situation in 2004) are connected.
> All areas connected to WWTPs are considered to be "sensitive" areas according the
UWWTD. This means if some regions would be declared as non-sensitive areas, the P-
loads emitted by the treatment plants in these regions would be considerably higher.
> The P-removal efficiency of treatment plants is according the UWWTD:
o > 100.000 pe: effluent concentration 1 mgP/l
o 10.000 100.000 pe: effluent concentration 2 mgP/l
o Optional for both categories mentioned: 80% reduction in relation to the inflow
load
o Less than 10.000 pe: secondary treatment; For secondary treatment a removal
of 0.6 gP/pe was assumed.
> The amount of sewer infiltration water influences the P-concentration in the raw waste
water and as a consequence the efforts required to meet the effluent quality standards
differ; calculation were carried out with 100 l (low) and 200 l (high) sewer infiltration
water per inhabitant.
> The amount of detergents consumed per inhabitant and day were varied as follows: In
Western countries like Austria, Germany or Suisse, the maximum consumption of P-
containing laundry detergents amounted up to 3 g P/inh.d. It has to be recognized that
the composition of detergents has changed in the last decades. For instance the total
amount of Sodium tripolyphosphate (STPP) contained in washing powders has been
reduced from 50% to about 25% (or even less). Therefore "modern" P-containing
laundry washing powders use less STPP per washing. Depending on the hardness of
the washing water 4 to 13 kg [Fox et al. 2002] of washing powders are consumed per
inhabitant. Assuming a consumption of 4 13 kg washing powder with an STPP
concentration of 25% per inhabitant would mean a specific P-emission of 0.7 2.2
gP/inh.d.
In the last years a considerable increase in the use of P-containing detergents in
automatic dishwasher products was observed in Germany and Austria amounting to
about 0.3 g P/inh.d. It is almost out of discussion that it is currently not possible to
replace STPP in these products. Therefore the "Zero-Laundry P-detergent scenario"
would mean a P-detergent consumption of 0.3 g P/inh.d. This amount is included in all
other Scenarios as well.
The following assumptions have been made:
o 0.3 gPdet/inh.d: this is the amount used in Germany or Austria in dish washing
products etc., but no P-containing laundry detergents
o 1.0 gPdet/inh.d: 0.7 gP/inh.d in laundry detergents + 0.3 gP/inh.d dish washing
products; for D and A the amounts of 0.3 gP/inh.d were used
o 1.75 gPdet/inh.d; 1.45 gP/inh.d in laundry detergents + 0.3 gP/inh.d dish
washing products; for D and A the amounts of 0.3 gP/inh.d were used
o 2.5 gPdet/inh.d: 2.2 gP/inh.d in laundry detergents + 0.3 gP/inh.d dish washing
products; for D and A, the amounts of 0.3 gP/inh.d were used
The Scenario with 1.75 gP/inh. probably is the worst case. The Scenario with 2.5
gP/inh.d is highly unlikely, as (i) not all areas in the Danube Basin have very hard water
(meaning that the washing powder consumption is increased) and (ii) there will be a
share of P-free detergents in the future.
UNDP/GEF DANUBE REGIONAL PROJECT
Basic Considerations on the Introduction of P-free Detergents
page 24
> The contribution from industry was assumed to be on the same level as in 2000
> The specific emission per inhabitant excluding detergents is 1.65 gP/inh.d
> The water consumption of 1 pe: 150 l
> For the 4 categories of settlements (A, B, C, D) assumptions for the amount of
industrial pe (peind) were made as follows: A: 1.2 pend/inhabitant.day, B and C: 1
pend/inh.d, D: 0.2 pend/inh.d; These assumptions are based on recent studies in
Austria reflecting well developed economic activities. In respect to the economic
transition process with lower economic intensity in the CEE countries the following
reduction for the amount of peind per inhabitant.d have been done for the calculations:
A, D: 0%, CZ, SK, HU, SL: minus 25%, all others: minus 50%
> P-Emissions from industry: 1.1 gP/peind.d
According these assumptions 18% of the population is not connected to a sewage treatment
plant (about 14.5 Mio. inhabitants).
4.1.2. Scenario results
As a base for comparison for the scenario results: The current emissions via waste water
treatment plants amount to about 24 kt P.
In total 32 different scenarios were evaluated:
4 different areas (A, B, C, D), 4 different assumptions on the use of detergents (0.3, 1.0, 1.75,
2.5 g P in detergents per inhabitant and day) and 3 different removal efficiencies for the Areas
A and B (2 different sewer infiltration rates (100l/inh., 200l/inh. and 80% removal rate).
The following results depict the mean P-emissions for the four different areas and the 4 different
consumptions of P-containing detergents. The results of the other scenarios are included in the
Annex.
The following main results were obtained:
> Only in the Scenarios with 0.3 g Pdet the emissions of P from point sources can be
lower than in 2000.
> The highest P- loads are emitted in the Area C (settlements between 2.000 and 10.000
inhabitants): 54 to 69% of the total P-emissions.
> The introduction of P-free laundry detergents in the Danubian countries would save
emissions of 5 kt P/a compared to the 1gP/inh.d Scenario and 10.5 kt P/a compared to
the 1.75 gP/inh.d Scenario.
> For treatment plants with more than 100.000 pe the emissions increase with
increasing consumption of P-containing detergents by up to 4 kt P.
> Doubling the sewer infiltration water (from 100 to 200 l/inh.d) increases the discharge
of P by 2 kt /a.
VIENNA UNIVERSITY OF TECHNOLOGY
Danube nutrients
page 25
40000
35000
30000
/
a
discharge 2000
t
P 25000
in
e 20000
r
g
ha 15000
c
i
s
d 10000
5000
0
> 100.000 inh.
10.000 -
2000 - 10.000
< 2000 inh.
Total
100.000
0.3 gP/i.d
1 gP/i.d
1.75 gP/i.d
2.5 gP/i.d
Figure 18: Mean P-discharge in different scenarios in tP/a.
The following figure shows the results of the 3 variants with different sewer infiltration rates
(Var 1 and Var 2) and the calculations based on the 80% removal rate (Var 3)
If the average consumption of P-containing detergents exceeds 0.65 gP/inh.d (including
automatic dish washing detergents) the emission will be higher as in the year 2004.
Total P-emissions wwtps
45000
40000
35000
/
a 30000
P
t 25000
i
n
n
o 20000
ssi 15000
emi 10000
5000
0
0
0,5
1
1,5
2
2,5
gP/inh.d in detergents
Var 1
Var 2
Var 3
discharge 2000
Figure 19: P-emissions from WWTPs of the 3 variants considered
UNDP/GEF DANUBE REGIONAL PROJECT
Basic Considerations on the Introduction of P-free Detergents
page 26
Storm water overflow: Based on experiences in Austria the total P-load emitted via storm
water over flow is less than 3% of the total P-load in the raw waste water. The emission of P via
storm water overflow is relatively important in areas with P-removal. Assuming a P-removal of
80% would mean that the contribution of storm water overflow amounts to 15% of the P-
emissions.
Assumption: 3% of the total P-load is discharged directly to the surface waters:
Depending on the Scenario considered in total 2 to 3 kt P will be discharged untreated into the
receiving waters via storm water overflow.
The P-output from households and industry amounts from 77 kt P (in the 0.3 g P/inh. Scenario)
to 128 kt P (2.5 g P/inh.d). Out of this about 15% are discharged partly to septic tanks and
partly directly into the river system causing additional P-loads in the rivers which are not
included in the Scenario calculations.
The overall removal efficiency of waste water treatment plants is about 65%.
4.1.3. Reliability and Uncertainties of Scenario calculations
In the scenario calculations it was assumed that the whole Danube Basin is considered as a
sensitive area. If parts would be defined as non-sensitive areas the requirements to waste water
treatment would be lower and the resulting emissions considerably higher.
>
Essential for the scenario calculations is the share of the population living in the
different settlement categories, especially in those below 2000 inhabitants (where
no sewer system and no treatment is required according the UWWTD) and in those
from 2000 to 10.000 inhabitants (biological treatment required in sensitive areas).
Romania is due to its size a key country in the Danube Basin. The data and statistics
evaluated for Romania for the year 2004 differ from those for the other countries
(e.g. population in settlements below 2000 inhabitants: Serbia: 27% Romania in
2004: 4.2%). As this data represents the official data base and as the
Implementation Plan for Directive 91/271/EEC concerning urban waste water
treatment for Romania is based on this data it was used it for the scenario
calculations.
Using averages of the other Danubian countries for Romania would lower the
emissions to surface waters by 3 to 7 kt per year. In this case the emissions in the
1g P/inh.d-Scenario would amount to the emissions in 2000 (see Annex Figure 29).
>
No change in the population was assumed. There are indications that the population
will decrease in several Danubian countries in the coming years. Probably the
decrease depends on the economic development in these countries. Furthermore it
would not be possible to distinct in which areas (categories of settlements) the
decrease will take place probably people from larger settlements are more mobile
to go abroad, people from small settlements will move to the cities etc.
>
No development of industrial activities (of industry connected to municipal waste
water treatment plants) was assumed.
>
No estimations on other industrial and agricultural point sources have been carried
out.
>
Probably also several settlements less than 2000 inh. will be (partly) sewered or will
discharge partly untreated waste water into the river systems.
VIENNA UNIVERSITY OF TECHNOLOGY
Danube nutrients
page 27
>
According the Implementation Plan for Directive 91/271/EEC concerning urban
waste water treatment for Romania there will be a time lag between the number of
inhabitant connected to sewers and the treatment of the collected waste water (e.g.
2010: collection system to be provided for 60.8% of the total pe, waste water
treatment to be provided for 50.5% of the total pe). For the year 2018 no time lag is
foreseen (100 % of the waste water collected will be treated).
4.2. Aspects of the management of the limited resource
phosphate
Phosphate rock is a limited resource and need to be managed properly. The phosphate-reserves
will last according to different estimations between 88 [Global 2000, 1976] and 500 years
[Finck, 1992], and those having low Cadmium concentrations are even more limited (Semi
Island Kola: 1 mg Cd/kg DM, Taiba / Senegal: 68 111 mg Cd/kg DM) [Sauerbeck and Rietz,
1980]. Additionally, the production of mineral N- and P-fertilizers demands (fossil) energy input.
Sewage sludge contains considerable amounts of nutrients that can be reintegrated into the
nutrient cycle. Together with the nutrients sewage sludges contain potentially hazardous
substances having different origins. The heavy metal contents of sewage sludge can be highly
influenced by diffuse sources (corrosion of roofings, etc.) [Zessner and Lampert, 2002]. These
substances might be accumulated on the long run in the environment if their quantity is not
properly considered. Furthermore the use of precipitants (side products of industry) for P-
removal is an additional source of heavy metals (esp. Ni, Cr but also Cd, Hg and Pb). Therefore
only precipitants with low concentrations of heavy metals shall be employed.
Sludge can be
>
used directly in agriculture: only if the concentrations of harmful substances in the
sludge are acceptable (aspect of soil protection); low costs
>
used after combustion: partly a separation of atmophilic elements (like Cd and Hg)
can be gained; organic hazard compounds are destroyed, no hygienic problems;
conditioning needed before use; high costs
Material balances show:
>
Less than 3% of the annual P-input into soils is eroded and therefore lost.
>
A considerable part of the P-input into the soil remains in the soil (increase in the P-
stock of soils) - this is not a loss of P and shall be considered as an interim storage.
>
About 70% of the P-emissions from households are included in the waste water, the
remaining 30% in solid wastes. Using P-containing detergents increases the share of
waste water.
As a conclusion: relevant losses of P only (can) occur via waste water and the final disposal of
sludge (in landfills, non agricultural areas).
The emissions to the environment (underground, direct discharges) from households in the
Danube Basin not connected to sewer systems and WWTPs are in the same order of magnitude
as the P-fertilizer consumption in Austria 2004.
The P-load of the sewage sludge in the Scenario 1.75 gP/inh.d is almost 4 times higher as the
total P-fertilizer consumption in Austria in 2004.
UNDP/GEF DANUBE REGIONAL PROJECT
Basic Considerations on the Introduction of P-free Detergents
page 28
Resource Potential of waste water implementing UWWD
80000
70000
60000
50000
/
a 40000
t P
30000
20000
10000
0
not conn.; em. to env.
em.point sources
P load sludge
fertilizer consumption
Austria 2004
0,3 g P det
1,0 g P det
1,75 g P det
2,5 g Pdet
Figure 20: Amount of P in various P-flows in comparison with the P-fertilizer
consumption in Austria 2004
4.3. Costs of sludge management and P-removal
In the following rough estimations on the additional amounts of dry matter of sewage sludge
produced due to the replacement of P-containing detergents respectively due to P-precipitation
in the WWTPs above 10.000 pe are given.
No detailed information will be provided on
>
the additional sludge volume,
>
the costs of dewatering,
>
costs of storage, etc..
For the sludge production in treatment plants without P-removal an amount of 40g dm/pe.d was
assumed. Implementing the UWWTD will lead to an annual total (biological) sludge production
of about 1.65 Mio t dm. As the implementation of the UWWTD requires P-removal in all plants >
10.000 pe additional amounts of sludge will be produced. The amount of additional sludge
produced in these plants is based on the following assumptions:
Calculation of the P-amount to be precipitated:
>
Removal of 0.6 gP/pe due to biological treatment
>
Total load minus P-load removed by biological treatment minus P-load in the
effluent = P-load to be precipitated
Calculation of the specific amount of additional dm/kg P:
>
Precipitant per kg P: 1.8 kg Fe/kg P, 0.87 kg Al/kg P
VIENNA UNIVERSITY OF TECHNOLOGY
Danube nutrients
page 29
>
Beta-value: 1.8 (Beta value: molar ratio of precipitant (Fe, Al) : P)
>
Sludge production: 2.5 kg dm/kg Fe (Beta = 1,5), 4 kg dm/kg Al (Beta = 1.5)
These assumptions lead to an additional dm production of 9.7 kg dm/kg P using Fe, and 7.5 kg
dm/kg P using Al.
A replacement of P-containing detergents by e.g. Zeolites would increase the production of
sewage sludge. The Zeolite consumption in Germany amounted in 1999 to 4.5 g/inh.d. 61 mio
people using the same amount (which should be comparable to the "reliable" worst case of 1.75
g P/inh.d. used in detergents) produces almost the same amount of sewage sludge as the P-
precipitation in all plants above 10.000 pe.
The amounts of additional sludge production due to the replacement of P-containing detergents
by e.g. Zeolites or due to the precipitation of P are similar (increase by 10-20%).
This means that the costs for sludge management would be similar for these two options.
350000
300000
250000
ar
e
r
y 200000
e
es p 150000
n
n
to 100000
50000
0
Zeolite
0,3 g
1 g P/inh.d
1,75 g
2,5
Zeolite+0,3
P/inh.d
P/inh.d
gP/inh.d
g P/inh.d
Figure 21: Increase of sludge dry matter due to P-precipitation in treatment plants
above 10.000 pe or due to the replacement of P-containing laundry detergents by
Zeolite in t/a (mean of the variants are used)
The main difference exists for the costs of precipitants for P-removal.
The following diagram depicts the additional need of precipitants in the 3 variants considered (
two different sewer infiltration rates (Var 1 and Var 2) and calculations based on the 80%
removal rate (Var 3)) compared to the "zero-laundry P-detergent scenario" of 0.3 gP/inh.d.
UNDP/GEF DANUBE REGIONAL PROJECT
Basic Considerations on the Introduction of P-free Detergents
page 30
120
100
80
%
60
40
20
0
1(2) mgP/l 100 l 1(2) mgP/l 200l 80% removal
s.i.r.
s.i.r.
1 g P/inh.d
1,75 g P/inh.d
2,5 gP/inh.d
Figure 22: Additional need for precipitants of 3 variants in % of the "Zero laundry
detergent Scenario" (0.3 g P/inh.d from detergents)
The mean amount of precipitants increase (compared to Scenario: 0.3 g P/inh.d in dish washing
detergents) by about 30% in the 1.0 gP/inh.d Scenario and by about 60% in the 1.75 gP/inh.d
Scenario. In the latter Scenario the annual operation costs of the treatment plants would
increase slightly by about 4 %.
In respect that
> the costs for sewer development are higher as those for the waste water treatment
and
> the operation costs amount to about 40% of the total costs of the waste water
treatment
the increase of the total costs of waste water management (collection + treatment) due to P-
precipitation is very small.
4.4. Different sources - different forms of P
Relevant P-emissions in 2000 stem from diffuse sources, mainly from the erosion of soils.
Therefore it is also possible to reduce P-emissions in the agricultural sector.
However there is a big difference in P- emissions from waste water treatment plants and from
erosion:
Emissions from waste water treatment plants are mainly in the form of dissolved Phosphate
which is immediately highly available for algae production. On the contrary P-emissions via
erosion are mainly in particulate form which is only partly available for algae-production. These
emissions provide a potential P-source which can be partly mobilised in the case of anaerobic
conditions in the waters which are a result of eutrophication.
VIENNA UNIVERSITY OF TECHNOLOGY
Danube nutrients
page 31
The different forms of P and consequently the different availability for algae growth shall be
considered in the innovative mechanisms for cost efficient mitigation of nutrient emissions when
measures on emission reductions from different sectors are compared.
To keep the mobilisation of the particulate P-source low, the discharge of dissolved P-forms has
to be kept low in order to avoid eutrophication.
UNDP/GEF DANUBE REGIONAL PROJECT
Issues for innovative mechanisms for cost efficient mitigation of nutrient emissions
page 32
5. ISSUES FOR INNOVATIVE MECHANISMS FOR COST
EFFICIENT MITIGATION OF NUTRIENT EMISSIONS
In September 2006 a document was established to support Stefan Speck in his work on
innovative mechanisms for cost efficient mitigation of nutrient emissions from diffuse sources in
large catchments. In the following section relevant issues to be tackled by innovative
mechanisms are depicted. Definitions of technical terms included in the document can be found
in the ANNEX.
A mere focus on the reduction of the nutrient-discharge to the WBS could lead to a deterioration
of rivers or local groundwater. A comprehensive strategy as outlined in the Water Framework
Directive including the quality of surface waters as well as ground water should be aimed at.
A trading system for nutrients has to be based on an agreement of the understanding of the
system (quantification of nutrient fluxes) and on an agreement on indicators as benchmark for
nutrient management (e.g. surpluses on soil, cattle density, emissions to surface waters, loads
to the Black Sea).
Emission from agriculture in the CEE countries probably will increase. The strategy should
enable economic growth and simultaneous ensure a "good" status of the environment.
The cost effectiveness of measures is only related to direct costs. Impacts e.g. on employment
or costs of the health system are not included.
Management of nutrients is only possible if the sources of the nutrients as well as their fate in
the catchment are known. For this purpose the flows, transformations and storage processes in
terms of mass loads have to be identified and described. These processes take place as well in
the soil, the groundwater as in the surface water.
The Danube river and its main tributaries play a minor role for N removal by denitrification. In
respect to P, the Iron Gate I is still a major point sink (up to 40% of the total P-load entering
the Iron Gate reservoir is deposited).
The Danube Delta does not play a dominant role in nutrient retention as > 90% of the nutrient
load of the Danube passes the three main channels directly to the Black Sea.
N
P
Si
0%
20%
40%
60%
80%
100%
ret. "small waters"
ret. large rivers
ret. delta
Black Sea
Figure 23: Emission to the river system and discharge to the Black Sea
VIENNA UNIVERSITY OF TECHNOLOGY
Danube nutrients
page 33
Differences in the hydrology and hydrogeology of the catchments highly affect the amount of
the total N emissions.
The quantification of the retention (denitrification) of nitrogen in soils and groundwater is more
important for the assessment of nitrogen emissions via groundwater as the surplus in
soils. In respect to emissions to surface water in comparison between regions the denitrification
in groundwater may outweigh the influence of the surplus in soils.
Monitoring data- handle with care:
> Load calculations based on monitoring data from diffuse sources, especially via the
groundwater do not reflect the current nutrient management. Depending on the
residence time of the groundwater the current emission is the effect of management
several years or even decades ago. This phenomenon has also to be considered for
success control: measures taken now will show effects in the future.
> Impact of flow conditions:
In respect to transport of P from the catchment to the receiving Sea, it can be
concluded that there is no immediate influence of high flow or flood events in
upstream parts of the Basin on the transport of P from the catchment to the receiving
Sea. Particle-bound P is mobilised from the catchment (erosion) and the river bottom
to a high extent at high flow events and transported at peak discharges downstream
where retention by sedimentation of particles takes place. On the one hand this
retention is a transport to floodplains. In this case it can be considered as more or less
long term retention. On the other hand sedimentation takes place in the riverbed, as
soon as the tractive effort of the river drops. In this second case the P-pool in the
sediments of the sedimentation area will be increased. If anaerobic conditions in the
sediment appear, part of the P will be transformed to soluble ortho-phosphate and will
continuously contribute to the P transport to the receiving Sea. Part of the P-retained
in the river sediment will be mobilised by resuspension at the next bigger high flow
event. All together these alternating processes of suspension, transport, export to
floodplains or sedimentation in the river bed with partly solution and partly
resuspension at the next event decrease the share of the P transport during high flow
events on the total loads transported in the more downstream parts of a catchment as
compared to the more upstream parts. In the year of occurrence of an extreme flood
event the P-transport of this year is dominated by the flood event. As average over
many years the contribution of high flow events to the total P-transport still may be
significant in smaller catchments. In a large catchment (e.g. river Danube) much
smaller contributions of flood events on the total P-transport can be expected as
average over many years.
For the monitoring of P-loads this means that flood events have to be specifically
addressed in tributaries anyway. In a large river the importance of event oriented load
monitoring depends on the time scale considered. For calculations of yearly loads
monitoring during flood events is still decisive. If average loads over 5 years and more
are taken into consideration, monitoring at flood events is less decisive unless the
probability of events increases significantly due to change of landuse practices in the
catchment or due to climate change.
UNDP/GEF DANUBE REGIONAL PROJECT
Issues for innovative mechanisms for cost efficient mitigation of nutrient emissions
page 34
7000
6000
5000
4000
t P 3000
2000
1000
0
average
6 days,
5 days,
year
July 1997 August
2002
Figure 24: P-flow at high flow: Danube at Vienna
In general: Elevated concentrations (phosphate, nitrate, ammonia etc.) due to nutrient
emissions affect the ground and surface water quality mainly in regions with low groundwater
recharge rates and low river discharge (small rivers) as the dilution capacity is low. At the same
time the retention of P and the removal of N by denitrification are high. These regions
contribute only little to the total nutrient discharge to the Black Sea. In these areas emission
reduction (at waste water treatment plants or best agricultural practice) is effective mainly for
the protection of the local/regional water quality. In such regions the influence of
agricultural practice on N discharges via groundwater, differs completely depending
on the distance (expressed in groundwater residence time) of an agricultural field
related to surface waters.
In regions with high groundwater recharge and high river discharge nutrient concentrations can
be low, while the loads transported in the rivers are comparatively high (nutrient retention and
losses during transport are low). Emission reduction in these regions effectively influences water
quality of the Danube, the Delta and the Black Sea.
VIENNA UNIVERSITY OF TECHNOLOGY
Danube nutrients
page 35
6. SUMMARY OF MAIN RESULTS
Results of the current level of nutrients in the Danube and dissolved oxygen in the NW shelf
> Transported nutrient loads in River Danube are highly influenced by physical factors as
water discharge and temperature. The influence of these physical factors on yearly
loads may outweigh anthropogenic impacts by far. Relatively high nutrient loads in
2005 are a result of the comparatively high flow situation of this year.
> If nutrient levels are compared at comparable flow situations, constant values up to a
slight decrease can be detected between 1996-1998 and 2004-2005. Still this
decrease is not significant. Changes in the emissions within this period can only be
assessed based on emission calculations, as has been done till 2000 within the daNUbs
project.
> As already indicated in the daNUbs-project the quality of phosphorus data at the Reni
station is questionable.
> The level of oxygen saturation at the bottom of WBS has significantly improved
between the 80ies and 1996. No dramatic changes in the period between 1996 and
2003 can be interpreted from existing data. The data base is not appropriate to
demonstrate changes between 1996 and 2003 in detail. Only for the period 2001-2003
a good database exists from the daNUbs project.
Results of Scenario calculations on the effects of different consumption of P-containing
detergents on the P-emissions from waste water treatment plants assuming full implementation
of the Urban Waste Water Directive and the whole Danube Basin is considered as sensitive area.
> If the average consumption of P-containing detergents exceeds 0.65 gP/inh.d
(including automatic dish washing detergents) the emission of P from WWTPs after full
implementation of the Urban Waste Water Treatment Directive will be higher as in the
year 2004 (emission in 2004: ca. 24 ktP).
> 54 to 69% of the P-emission will stem from settlements between 2.000 and 10.000
inhabitants as for these areas no P-removal is required by the UWWTD.
> The introduction of P-free laundry detergents in the Danubian countries would save 5
kt P/a compared to the 1gP/inh.d Scenario and 10.5 kt P/a compared to the 1.75
gP/inh.d Scenario.
> For treatment plants with more than 100.000 pe the emissions increase with
increasing consumption of P-containing detergents by up to 4 kt P.
> In total 2 to 3 kt P will be discharged untreated into the receiving waters via storm
water overflow.
> The amounts of additional sludge production due to the replacement of P containing
detergents by e.g. Zeolites or due to the precipitation of P are similar (increase by 10-
20%). This means that the costs for sludge management would be similar for these
two options.
> The costs of precipitants for P-removal compared to the total costs (investment costs
plus operation costs) for sewer development and waste water treatment are very
small.
UNDP/GEF DANUBE REGIONAL PROJECT
References
page 36
REFERENCES
Finck A. (1992). Dünger und Düngung. 2. Aufl., VCH-Verlagsgesellschaft mbH, Weinheim.
Fox K.K., Cassani G., Facchi A., Schröder F.R., Poelloth C., Holt M.S. (2002): Measured
Variations in boron loads reaching European sewage treatment works. Chemosphere 47, 499
505
Sauerbeck D. and Rietz E. (1980). Zur Cadmiumbelastung von Mineraldüngern in Abhängigkeit
von Rohstoff- und Herstellungsverfahren. Landw. Forschung, Sdhft. 37, pp. 685.
Zessner M. and Lampert C. (2002). The use of regional material balances in water quality
management. Urban water, 4, pp. 73-83.
VIENNA UNIVERSITY OF TECHNOLOGY
Danube nutrients Annexes
page 37
ANNEXES
ANNEX 1
Results of scenario calculations
ANNEX 2
Definition of technical terms
UNDP/GEF DANUBE REGIONAL PROJECT
ANNEX 1: Results of Scenario calculations
page 38
ANNEX 1
RESULTS OF SCENARIO CALCULATIONS
VIENNA UNIVERSITY OF TECHNOLOGY
Danube nutrients
page 39
RESULTS OF SCENARIO CALCULATIONS
Table 1: Mean of Scenario calculations for emissions from municipal waste water
treatment plants in tP/a
consumption of P in detergents per inh.d
emission in t P/a
0.3 gP/i.d
1 gP/i.d
1.75 gP/i.d
2.5 gP/i.d
> 100.000 inh.
3730
4015
4318
4621
10.000 - 100.000
4926
5211
5517
5822
2.000 - 10.000
11366
15844
20636
25428
< 2.000 inh.
1172
1188
1188
1188
Total 21.194
26.258
31.659
37.060
UNDP/GEF DANUBE REGIONAL PROJECT
ANNEX 1: Results of Scenario calculations
page 40
1:
0,3 gP/inh.d in detergents
The effluent
45000
40000
concentration in
35000
area A is 1 mgP/l,
/a 30000
in Area B 2 mgP/l.
t
P
discharge 2000
25000
1
e
in
The infiltration rate
g
2
ar 20000
3
amounts to 100
i
sch 15000
d
l/inhabitant.
10000
5000
In the area C and
0
biological treatment
Area A
Area B
Area C
Area D
total
is applied.
1,0 gP/inh.d in detergents
45000
40000
35000
2.
/
a 30000
t
P
discharge 2000
The effluent
25000
1
e
in
2
a
r
g 20000
concentration in
3
i
sch 15000
area A is 1 mgP/l,
d
10000
in Area B 2 mgP/l.
5000
The infiltration rate
0
Area A
Area B
Area C
Area D
total
amounts to 200
l/inhabitant.
In the area C
1,75 gP/inh.d in detergents
biological treatment
45000
is applied.
40000
35000
30000
/
a
discharge 2000
P
t 25000
1
i
n
e
2
arg 20000
3
3:
i
sch
d 15000
80% P-reduction in
10000
relation to the
5000
inflow load
0
Area A
Area B
Area C
Area D
total
In the area C
biological treatment
is applied.
2,5 gP/inh.d in detergents
45000
40000
35000
/
a 30000
discharge 2000
t
P
1
25000
e in
2
a
r
g 20000
3
i
sch 15000
d
10000
5000
0
Area A
Area B
Area C
Area D
total
Figure 25: P-emissions in different settlement categories and different variants
VIENNA UNIVERSITY OF TECHNOLOGY
Danube nutrients
page 41
The following diagrams show the different P-emissions depending on the use of P-containing
laundry detergents.
40000
35000
30000
/
a
discharge 2000
t
P 25000
in
e 20000
r
g
ha
c 15000
s
di 10000
5000
0
> 100.000 inh.
10.000 -
2000 - 10.000
< 2000 inh.
Total
100.000
0.3 gPdet/inh.d
1 gPdet/inh.d
1.75 gPdet/inh.d
2.5 gPdet/inh.d
Figure 26: P-discharge of Variant 1: The effluent concentration in area A is 1 mgP/l, in
Area B 2 mgP/l. The infiltration rate amounts to 100 l/inhabitant. In the area C
biological treatment is applied.
40000
35000
30000
discharge 2000
/
a 25000
P
t
n
e
i 20000
r
g
s
c
ha
15000
di
10000
5000
0
> 100.000 inh.
10.000 - 100.000
2000 - 10.000
< 2000 inh.
Total
0.3 gPdet/inh.d
1 gPdet/inh.d
1.75 gPdet/inh.d
2.5 gPdet/inh.d
Figure 27: P-discharge of Variant 2: The effluent concentration in area A is 1 mgP/l, in
Area B 2 mgP/l. The infiltration rate amounts to 200 l/inhabitant. In the area C
biological treatment is applied.
UNDP/GEF DANUBE REGIONAL PROJECT
ANNEX 1: Results of Scenario calculations
page 42
40000
35000
30000
/a
discharge 2000
25000
t
P
in
e 20000
r
g
a
h
c 15000
i
s
d 10000
5000
0
> 100.000 inh.
10.000 - 100.000
2000 - 10.000
< 2000 inh.
total
0.3 gPdet/inh.d
1 gPdet/inh.d
1,75 gP/inh.d
2.5 gPdet/inh.d
Figure 28: P-discharge of Variant 3: 80% P-reduction in relation to the inflow load. In
the area C biological treatment is applied.
40000
35000
30000
/a
P
discharge 2000
t 25000
e
in 20000
a
r
g
15000
i
s
c
h
d 10000
5000
0
> 100.000 inh.
10.000 - 100.000
2000 - 10.000
< 2000 inh.
Total
0.3 gP/i.d
1 gP/i.d
1.75 gP/i.d
2.5 gP/i.d
Figure 29: Emissions from WWTPs assuming "average" values for the percentage of
the population in settlements from 2.000 10.000 and below 2.000 for Romania
(mean of the 3 Variants).
VIENNA UNIVERSITY OF TECHNOLOGY
Danube Nutrients Annexes
page 43
ANNEX 2
DEFINITION OF TECHNICAL TERMS
UNDP/GEF DANUBE REGIONAL PROJECT
ANNEX 2: Definition of technical terms
page 44
DEFINITION OF TECHNICAL TERMS
Definitions of key technical terms prepared for Stefan Speck in order to support the work on key
issues for innovative mechanisms for cost efficient mitigation of nutrient emissions from diffuse
sources in large catchments.
Denitrification: The conversion of nitrite and nitrate nitrogen (after nitrification) to inert
nitrogen gas (N2). This process may occur in all parts of the water cycle and leads to a "loss" of
nitrogen from the water system. The process requires that little or no oxygen be present in the
system and that an organic food source be provided to foster growth of another type of
bacteria. The resultant nitrogen gas is released to the atmosphere. In addition to N2 also N2O
can be emitted. While N2 is no harm for the environment, N2O is a greenhouse gas and may also
return by deposition to the soil water system.
Emission: Refers to pollution being released or discharged into the environment from natural or
man-made sources.
Erosion: The disruption and movement of soil particles by wind, water, or ice, either occurring
naturally or as a result of land use.
Eutrophication: The fertilization of surface waters by nutrients that were previously scarce.
Eutrophication through nutrient and sediment inflow is a natural aging process by which warm
shallow lakes evolve to dry land. Human activities are greatly accelerating the process. The
most visible consequence is the proliferation of algae. This algae eventually die and decompose,
which reduce the amount of dissolved oxygen in the water.
Nonpoint Source (diffuse Source): A diffuse source of pollution that cannot be attributed to
a clearly identifiable, specific physical location or a defined discharge channel. This includes the
nutrients that runoff the ground from any land use - croplands, feedlots, lawns, parking lots,
streets, forests, etc. - and enter waterways. It also includes nutrients that enter through air
pollution, through the groundwater, or from septic systems.
Point Source: A source of pollution that can be attributed to a specific physical location; an
identifiable, end of pipe "point". The vast majority of point source discharges for nutrients are
from wastewater treatment plants, although some come from industries.
Nitrogen: N is used primarily by plants and animals to synthesize protein. Nitrogen enters the
ecosystem in several chemical forms and also occurs in other dissolved or particulate forms,
such as tissues of living and dead organisms. In water systems usually the dissolved forms are
dominating.
Phosphorus: A key nutrient in the ecosystems; phosphorus occurs in dissolved organic and
inorganic forms, often attached to particles of sediment. This nutrient is a vital component in
the process of converting sunlight into usable energy forms for the production of food and fiber.
It is also essential to cellular growth and reproduction for organisms such as phytoplankton and
bacteria. Phosphates, the inorganic form are preferred, but organisms will use other forms of
phosphorus when phosphates are unavailable.
VIENNA UNIVERSITY OF TECHNOLOGY
Danube Nutrients Annexes
page 45
Limiting nutrient: A nutrient, whose concentration in the environment of an organism,
determines the growth and productivity of that organism.
Nutrient load: The term "nutrient load" refers to the mass or quantity of nutrients that pass
out of a catchment or sub-catchment via a stream per unit of time. Nutrient load comprises
direct and indirect input of waterborne and airborne nutrients from point sources and diffuse
sources, as a result of land-based activities (driving forces) within the drainage area. This total
input is caused by human activities as well as from natural sources.
The size of this input (load) can be indicated in different ways, for example as:
> tonnes of N or P /year = total annual input from all sources in all countries within the
drainage area
> tonnes of N or P/year/source = e.g. tonnes of N from agriculture in all countries within
the drainage area
> tonnes of N or P/year/country = e.g. tonnes of N from all sources in Austria
specific load:
> quantity per capita and year = e.g. kg N per person and year or
> quantity per ha and year = e.g. kg N per ha and year
Nutrient loads represent the emissions and inputs of nutrients from all sources (diffuse and
point sources) upstream and include the effects of nutrient transport losses (e.g. from
assimilation, chemical breakdown and storage due to sedimentation).
Nutrient loads can be measured by monitoring the outputs from catchments and sub-
catchments. A nutrient load model can then be calibrated against this monitoring data.
However, there are difficulties in measuring nutrient emissions if these are not to include any
effect of transport loss and therefore, the results from studies may not be true nutrient
emissions.
It seems that the major difference between nutrient emissions reported in the literature and
nutrient loads that can be measured by monitoring outputs from catchments is a matter of
scale. Scale refers to the size of the area of land used for measurement. The effect of transport
losses in varying areas of land may explain, in part, the wide variation in reported nutrient
generation rates (= specific loads) for the same landuse. In general: the larger the scale the
better is the fit between emission estimates and the nutrient load calculated on the base of
monitoring data.
UNDP/GEF DANUBE REGIONAL PROJECT
ANNEX 2: Definition of technical terms
page 46
VIENNA UNIVERSITY OF TECHNOLOGY