The EFSA Journal (2008) 842, 1-28
Technical Guidance
for assessing the safety of feed additives for the environment
Prepared by the Panel on Additives and Products or Substances used in
Animal Feed
(Question No EFSA-Q-2008-408)
Adopted on 22 October 2008
FOREWORD
This document provides guidance on how to conduct and report studies concerning the assessment of
the safety of feed additives for the environment.
This technical guidance is an update of the previous opinion (Opinion of the Scientific Panel on
Additives and Products or Substances used in Animal Feed on the development of an approach for the
environmental risk assessment of additives, products and substances used in animal feed, EFSA, 2007)
and supersedes it.
Consideration of the environmental impact of additives is important since administration of additives
typically occurs over long periods, often involves large groups of animals and the constitutive active
substance(s) may be excreted to a considerable extent either as the parent compound or its metabolites.
To determine the environmental impact of additives, a stepwise approach shall be followed. All
additives have to be assessed through Phase I to identify those additives which do not need further
testing. For the other additives a second phase (Phase II) assessment is needed to provide additional
information, based upon which further studies may be considered necessary. These studies shall be
conducted according to Council Directive 67/548/EEC.1
1
OJ 196, 16.8.1967, p. 1, as last amended by Commission Directive 2004/73/EC (OJ L 152, 30.4.2004, p. 1 and corrigendum
in OJ L 216, 16.6.2004, p. 3)
The EFSA Journal (2008) 842, 1-28
Technical guidance – Environmental Risk Assessment
TABLE OF CONTENTS
Foreword ..................................................................................................................................................... 1
Table of Contents ........................................................................................................................................ 2
1. Phase I assessment .............................................................................................................................. 3
1.1.
Additives for terrestrial animals ................................................................................................ 3
1.1.1. Calculation of PEC in soil (PECsoil) ...................................................................................... 4
1.1.2. Calculation of the PEC in groundwater (PECgw)................................................................... 5
1.2.
Additives for aquatic animals .................................................................................................... 6
1.2.1. Calculation of the PEC in the sediment (PECsed) for sea cages ............................................ 7
1.2.2. Calculation of the PEC in surface water from aquaculture (PECswaq) in raceway / pond /
tanks and recirculation systems .......................................................................................................... 7
2. Phase II assessment ............................................................................................................................ 8
2.1.
Phase II A................................................................................................................................... 9
2.1.1. Data requirements................................................................................................................ 10
2.1.1.1. Physical-Chemical properties studies ......................................................................... 10
2.1.1.2. Environmental fate studies ......................................................................................... 10
2.1.1.3. Toxicity tests............................................................................................................... 11
2.1.2. PEC refinement for soil ....................................................................................................... 12
2.1.2.1. Refinement based on metabolism ............................................................................... 12
2.1.2.2. Refinement based on degradation in manure.............................................................. 12
2.1.2.3. Refined PEC in soil for persistent compounds ........................................................... 13
2.1.2.4. Refinement based on degradation in soil .................................................................... 14
2.1.3. PEC in surface water (PECsw) and fresh water sediment (PECfw sed) for additives used in
terrestrial animals ............................................................................................................................. 15
2.1.3.1. PEC in surface water (PECsw)..................................................................................... 15
2.1.3.2. PEC in fresh water sediment (PECfw sed) ................................................................... 15
2.1.4. Advanced models for calculating PEC ................................................................................ 15
2.1.4.1. Groundwater ............................................................................................................... 16
2.1.4.2. Surface Water ............................................................................................................. 17
2.1.4.3. Interpretation of results from FOCUS ........................................................................ 18
2.1.5. PEC refinement for marine sediment (aquaculture) ............................................................ 18
2.2.
Phase IIB .................................................................................................................................. 19
2.2.1. Toxicity tests: Terrestrial compartment .............................................................................. 19
2.2.2. Toxicity tests: Fresh water compartment ............................................................................ 19
2.2.3. Toxicity tests: Sediment aquaculture .................................................................................. 19
3. PNEC derivation............................................................................................................................... 19
3.1.
Surface water ........................................................................................................................... 19
3.2.
Terrestrial environment ........................................................................................................... 20
3.2.1. Terrestrial plants.................................................................................................................. 20
3.2.2. Earthworms ......................................................................................................................... 20
3.2.3. Micro-organisms.................................................................................................................. 20
3.2.3.1. Normalisation ............................................................................................................. 21
3.3.
Sediment .................................................................................................................................. 21
3.3.1. Marine sediment .................................................................................................................. 21
3.3.2. Fresh water sediment ........................................................................................................... 21
Glossary (definition of terms) ................................................................................................................... 23
References ................................................................................................................................................. 25
Appendix ................................................................................................................................................... 26
The EFSA Journal (2008) 842, 2-28
Technical guidance – Environmental Risk Assessment
1.
Phase I assessment
The purpose of Phase I assessment is to determine if a significant environmental effect of the additive is
likely and whether a Phase II assessment is necessary (see quick check).
Exemption from Phase II assessment may be made on one of two criteria, unless there is scientificallybased evidence for concern:
The chemical nature and the biological effect of the additive and its conditions of use indicate
that impact will be negligible, i.e., where the additive is:
o
a physiological or natural substance (e.g. vitamin, protein, carotenoids) that will not
result in a substantial increase of the concentration in the environment, considering the
intrinsic toxicity of the excreted substance(s)
o
intended for non food-producing animals only
The worst case Predicted Environmental Concentration (PEC) is too low to be of concern (see
Sections 1.1. and 1.2). The PEC shall be evaluated for each compartment of concern (see
Sections 1.1. and 1.2), assuming that 100% of the dose ingested is excreted as the parent
compound.
Quick-check – Environmental Risk Assessment: Phase I
Is the additive a physiological/natural
substance/agent of established safety?
Yes
No
Is the additive intended for non-food
producing animals only?
No environmental
concern
No
Stop in Phase I
Additives used in terrestrial species
Is the PEC in ground water <0.1µg L-1
and PEC in soil <10 µg kg-1?
Yes
Additives used in aquatic species
Is the PEC in surface water <0.1µg L-1
and PEC in sediment <10 µg kg-1?
No
Phase II required
1.1.
Additives for terrestrial animals
When excreta from livestock are applied on land, the use of feed additives can lead to contamination of
soil, ground water and surface water (via drainage and run-off).
The EFSA Journal (2008) 842, 3-28
Technical guidance – Environmental Risk Assessment
The PECs used in Phase I would arise considering all excreted compounds being spread on land and
other specified assumptions (see Sections 1.1.1 and 1.1.2) which reflect in summary worst case
conditions.
PEC for soil (PECsoil) (default: 5 cm depth) is less than 10 μg kg-1 and
If
PEC for groundwater (PECgw) is less than 0.1 μg L-1
the substance is considered not to pose a risk for the environment, and therefore no further assessment
is necessary.
1.1.1. Calculation of PEC in soil (PECsoil)
The amount of manure/slurry containing the feed additives allowed to be spread on land depends on the
nitrogen content of the manure and the nitrogen immission standard. Based on the data on feed intake
and nitrogen content in manure, the maximum amount of parent compound per kg nitrogen excreted can
be calculated by multiplying the concentration of the additive in feed with the feed consumption and
dividing it by the corresponding nitrogen excretion. In Table 1, the feed intake and corresponding
nitrogen excretion is given for the relevant target animals (ERM/AB-DLO, 1999). Other data can be
used if justified.
For a worst case estimation of the concentration in soil, the following conditions are assumed:
The additive is continuously applied to the feed of the target animal;
Total intake of the active substance is considered to be excreted as parent compound;
The current annual nitrogen immission standard for slurry/manure spread on land is applied
(EU Nitrate Directive 91/676/EEC2);
There is no dissipation of the parent compound during storage and spreading of slurry/manure;
The standard assumption, when slurry/manure is spread on land, is that additive is applied to
soil up to 5 cm depth; an exemption can be made for poultry manure which is commonly
applied to arable land and incorporated into 20 cm soil depth.
Table 1.
Default values for feed intake and nitrogen excretion (ERM/AB-DLO, 1999)
Animals
Pigs for fattening
Sows (with piglets)
Dairy cows3
Cattle for fattening
Veal calves
Sheep-goats4
Lambs for fattening
Broilers
Laying hens
Turkeys
1
2
3
4
2
Feed intake (kg animal-1 year-1)1
N excretion (kg animal-1 year-1)2
800
1300
6700
3070
770
740
160
33
50
90
10.1
26.3
114
53
11
10.2
1.7
0.36
0.56
1.1
Standardised complete feed with 88 % of dry matter
Considering nitrogen losses during storage
Dairy cows with a milk yield of 7000 kg fat corrected milk year-1
Data for sheep and goats (body weight 70 kg) were derived from the SCAN report on zinc taking into account a 10 %
N loss
OJ L375, 31.12.1991
The EFSA Journal (2008) 842, 4-28
Technical guidance – Environmental Risk Assessment
The following equations should be used to calculate PEC in manure and soil:
Cadd FI total
N excreted
PEC manure
PEC soildw
RHOsoil
PEC manure Q
CONV area field DEPTH field
where:
Symbol
Parameter
Default
Value*
Input
Cadd
FItotal
Nexcreted
RHOsoil
DEPTHfield
Concentration of the additive in feed
Total feed intake (DM) per year
Total N excretion per year
Bulk density of soil
Mixing depth with soil
CONVarea field
Conversion factor for the area of the agricultural field
Q
Annual nitrogen immission standard
Intermediate results
PECmanure
Concentration of the additive (parent compound) in
manure expressed per amount nitrogen
Output
PECsoil
Concentration of the additive (parent compound) in soil
*
1500
20
5
10000
170
Unit
mg kg-1 complete feed
kg feed
kg N
kg m-3
cm (arable land)
cm (grassland)
m2 ha-1
kg N ha-1
mg kg-1 N
mg kg-1 soildw
The use of the indicated default values in the equations is recommended. Reasons for any deviations from these values
should be given by the applicant.
1.1.2. Calculation of the PEC in groundwater (PECgw)
The PEC groundwater (PECgw) is calculated using the approach described in the EU Technical
Guidance Document for new and existing substances (EU TGD) where PECgw is set equal to PEC
porewater (PECporewater). In this screening model, partitioning depends on equilibrium sorption to solids,
no saturation at binding places and steady-state conditions. This model provides a worst case estimate
of the groundwater concentrations as movement, dilution, desorption, transformation, weather or crops
are not considered. Soil is defined through compartment volumes for solids, water and air, dry bulk
density and texture (mineral and organic fraction). The soil depth for calculation of the PECsoil used for
calculating the PECgw is set at 20 cm.
The model calculation of the concentration in groundwater is as follows:
PEC manure
PEC soil ww
RHOsoil
K air
Cadd FI total
N excreted
PEC manure Q
CONV area field DEPTH field
water
VP MOLW
SOL R TEMP
Kp soil
Focsoil Koc
The EFSA Journal (2008) 842, 5-28
Technical guidance – Environmental Risk Assessment
K soil
water
Fairsoil K air
water
PEC porewater
Fwater
soil
Fsolidsoil
Kpsoil
RHOsolid
1000
PEC soilww RHOsoil
Ksoil water 1000
where:
Symbol
Input
RHOsoil
DEPTHfield
RHOsolid
Fairsoil
Fwater-soil
Fsolidsoil
Focsoil
TEMP
R
VP
MOLW
SOL
FItotal
Cadd
Nexcreted
Q
CONVarea field
DEPTHfield
Parameter
Bulk density of fresh soil
Mixing depth with soil
Density of soil solids
Fraction air in soil
Fraction water in soil
Fraction solids in soil
Weight fraction organic carbon in soil
Temperature at air-water interface
Gas constant
Vapour pressure
Molar mass
Water solubility
Total feed intake (DM) per year
Concentration of the additive in feed
Total N excretion per year
Annual nitrogen immission standard
Conversion factor for the area of the agricultural
field
Mixing depth with soil
Koc‡
Organic carbon normalised partition coefficient
Intermediate results
Ksoil-water
Partition coefficient solids and water in soil (v/v)
Kpsoil
Partition coefficient solids and water in soil (v/w)
Kair-water
Partition coefficient air and water in soil
Output
PECmanure
Concentration of the additive (parent compound) in
manure expressed per amount nitrogen
PECsoilww
Concentration of the additive (parent compound) in
soil
Concentration of the additive (parent compound) in
PECporewater
porewater
*
‡
1.2.
Default
Value*
1700
20
2500
0.2
0.2
0.6
0.02
285
8.314
170
10000
20
5
Unit
kg m-3
cm
kgdwt m-3
m3 m-3
m3 m-3
m3 m-3
kg kg-1
K
Pa m3 mol-1 K-1
Pa
g mol-1
mg L-1
kg feed
mg kg-1 complete feed
kg N
kg N ha-1
m2 ha-1
cm (arable land)
cm (grassland)
dm3 kg-1
m3 m-3
dm3 kg-1
m3 m-3
mg kg-1 N
mg kg-1 soilww
mg L-1
The use of the indicated default values in the equations is recommended. Reasons for any deviations from these
values should be given by the applicant.
Where no measured Koc value is available, in the Phase I assessment estimation techniques can be used based on
correlation with the Kow or water solubility given in OECD guideline 106 (Soil Adsorption/Desorption).
Additives for aquatic animals
Feed additives used in aquaculture can result in contamination of sediment and water. The
compartment of concern for the environmental risk assessment for fish farmed in cages is assumed to
be the sediment. For fish farmed in land-based systems the effluent flowing to surface water is
considered to pose the major environmental risk.
The PECs used in Phase I would arise considering all excreted compounds being dispersed to sediment
and water and other specified assumptions (see Sections 1.2.1 and 1.2.2) which reflect in summary
worst case conditions.
The EFSA Journal (2008) 842, 6-28
Technical guidance – Environmental Risk Assessment
If
PEC for sediment (PECsed) (default: 20 cm depth) is less than 10 μg kg-1 wet weight and
PEC for surface water (PECsw) is less than 0.1 μg L-1
the substance is considered not to pose a risk for the environment, and therefore no further assessment
is necessary.
The method to calculate the PEC in sediment and water varies for the different European fish
production systems: sea cages versus land-based aquaculture (ponds, tanks and recirculation systems).
In aquaculture operations involving the use of sea cages, benthic organisms are considered to be most at
risk whereas waterborne exposure of pelagic organisms presents the main risk from land-based fish
farms.
1.2.1. Calculation of the PEC in the sediment (PECsed) for sea cages
The PEC can be calculated as follows:
PC faeces
PECsed
Cadd CF
PCfaeces kdep Tproduction
RHOsed DEPTH sed
where:
Symbol
Input
Cadd
CF
kdep
Tproduction
RHOsed
DEPTHsed
Output
PCfaeces
PECsed
Parameter
Default value*
Concentration additive in feed
Conversion factor (kg feed to kg carbon in faeces)
Maximum deposition rate of faeces
Number of production days
Fresh bulk density sediment
Mixing depth in sediment
‡
15.1
0.01¥
365
1300
20
Concentration of the additive (parent compound) in faeces
Highest initial concentration of additive in sediment
Unit
mg kg-1complete feed
kg kg-1 carbon
kg carbon m-2 day-1
day
kgwwt m-3
cm
mg kg-1 carbon
mgc kg-1 ww **
* The use of the indicated default values in the equations is recommended. Reasons for any deviations from these values
should be given by the applicant.
‡
Concentration in feed (Cadd) given in mg kg-1 feed has to be converted in mg kg-1 C feed (2.06). Subsequently, mg kg-1 C
feed is converted to into mg kg-1 C faeces (7.3), hence the total conversion is 2.06 * 7.3=15.1
¥
According to Hansen et al., 1991 ; Karakassis et al., 2002 ; Corner et al., 2006 ; Holmer et al., 2006 ; Kutti et al., 2007
1.2.2. Calculation of the PEC in surface water from aquaculture (PECswaq) in raceway / pond /
tanks and recirculation systems
The PEC can be calculated as follows:
PEC swaq
Cadd FR (1 Fret )
Flow DF
The EFSA Journal (2008) 842, 7-28
Technical guidance – Environmental Risk Assessment
where:
Symbol
Input
Cadd
FR
Fret
Flow
DF
Output
PECswaq
Parameter
Unit
Concentration of the additive additive in feed
Feed Ration
Fraction of retention in the system
Water flow rate through the system
Dilution Factor
mgc kg-1 complete feed
kg feed kg fish-1 day-1
0*
L kg-1 fish day-1
10
Highest initial concentration of additive (parent
compound) in surface water
mgc L-1
* In Phase I it is assumed that there is no retention in the system and the value is set to 0. For Phase II assessment the
retention value can be adjusted according to data.
For Feed Ration and Water Flow Rate, the following default settings are proposed for some fish
species commonly farmed in Europe:
Fish types
Salmon
Rainbow trout
Seabass/Seabream
Turbot
Feed Ration
(kg feed kg fish-1 day-1)
0.011
0.02
0.013
0.013
Water Flow Rate
(L kg-1 fish day-1)
1400
14002
4003
7203
1
Bailey, 2003
http://www.fao.org/fishery/culturedspecies/Oncorhynchus_mykiss/en#tcNC008F
3
Hussenot et al., 1998
2
2.
Phase II assessment
The aim of Phase II is to assess the potential for additives to affect non-target species in the
environment, including both aquatic and terrestrial species or to reach groundwater at unacceptable
levels. It is not practical to evaluate the effects of additives on every species in the environment that
may be exposed to the additive following its administration to the target species. The taxonomic levels
tested are intended to serve as surrogates or indicators for the range of species present in the
environment. For example, earthworms could be used to represent soil invertebrates, chlorella for
aquatic plants, and rainbow trout for aquatic vertebrates.
The Phase II assessment is based on a risk quotient approach, where the calculated PEC and Predicted
No Effect Concentration (PNEC) values for each compartment shall be compared. The PNEC is
determined from experimentally determined endpoints divided by an appropriate assessment (safety)
factor. The more data are available, the lower is the assessment factor applied. The PNEC value shall
be calculated for each compartment of concern. How to derive PNECs is detailed in Section 3.
The Phase II assessment is based on a tiered approach (Figure 1).
The first tier, Phase IIA, makes use of a limited number of fate and effect studies to produce a
conservative assessment of risk based on exposure and effects in the environmental compartment of
concern.
To start, a comparison should be made between the initial PEC and the PNEC (the initial PEC should
also consider the potential accumulation in soil, see Section 2.1.2.3):
If the ratio of the PEC to the PNEC is lower than 1 no further assessment is required, unless
bioaccumulation is expected (for further details see Section 2.1);
If the PEC/PNEC is > 1, a more refined PEC can be calculated based on data not considered in
Phase I.
If the refined PEC/PNEC ratio predicts an unacceptable risk (ratio > 1), the applicant shall progress
to Phase IIB to refine the environmental risk assessment (for further details see Section 2.2).
The EFSA Journal (2008) 842, 8-28
Technical guidance – Environmental Risk Assessment
PEC/PNEC < 1
Compare initial PEC with PNEC
PEC/PNEC ≥ 1
PEC refinement based on degradation/
metabolism and/or more sophisticated
exposure models
PECR/PNEC < 1
STOP
PECR/PNEC ≥ 1
PNEC refinement based on
additional (chronic) toxicity
PEC or PECR/PNECR < 1
PECR/ PNECR ≥ 1
Potential risk
Figure 1.
2.1.
Phase II decision tree for the environmental risk assessment of soil and aquatic
compartment for terrestrial animals (PECR and PNECR are PEC refined and PNEC
refined, respectively)
Phase II A
In addition to the compartments considered in Phase I, the PEC for surface water has to be calculated
considering runoff and drainage.
Based on data not considered in Phase I, a more refined PEC can be calculated for each environmental
compartment of concern. In ascertaining the refined PEC, account shall be taken of:
The concentration of active substance/metabolites of concern in manure/fish faeces following
administration of the additive to animals at the proposed dose level. This calculation shall
include consideration of dosage rates and amount of excreta produced;
The potential degradation of the excreted active substance/metabolites of concern during
normal manure processing practice and storage prior to its application to land;
The adsorption/desorption of the active substance/metabolites of concern onto soil or sediment
in case of aquaculture, preferentially determined by studies in soil/sediment;
Degradation in soil and water/sediment systems;
Other factors such as hydrolysis, photolysis, evaporation, etc.
The highest value for the PEC obtained by these calculations for each environmental compartment of
concern should be adopted for Phase II risk assessment purposes.
If a high persistence in soil/sediment is anticipated (time to degradation of 90% of original
concentration of the compound (DT90) > 1 year), the potential for accumulation should be considered.
The concentrations of additive (or metabolite) producing serious adverse effects for various trophic
levels in the environmental compartments of concern shall be determined. These tests are mostly acute
tests and shall follow OECD guidelines or similar well-established guidelines. Studies for the
terrestrial environment should include: toxicity to earthworms, three terrestrial plants and, soil microorganisms (e.g., effects on nitrogen fixation) (see Section 2.1.1.3). Studies for the freshwater
environment shall include: toxicity to fish, Daphnia magna, algae and a sediment dwelling organism.
The EFSA Journal (2008) 842, 9-28
Technical guidance – Environmental Risk Assessment
In case of sea cages, three species of different taxa of sediment dwelling organisms shall be studied
(see Section 2.1.1.3).
Calculation of the PNEC value shall be carried out for each compartment of concern. The PNEC is
normally derived from the lowest toxicity value observed in the above tests and dividing by a safety
factor of at least 100 depending on the endpoint and number of test species used.
The potential for bioaccumulation can be estimated from the value of the n-octanol/water partition
coefficient, Log Kow. Values ≥ 3 indicate that the substance may bioaccumulate. In order to assess the
risk for secondary poisoning it shall be considered whether to carry out a bioaccumulation factor
(BCF) study at Phase IIB.
2.1.1. Data requirements
2.1.1.1.
Physical-Chemical properties studies
In order to evaluate the environmental fate and toxicity data of the feed additive, some basic physicalchemical properties, required under Commission Regulation (EC) No 429/2008, are needed. These
properties are described in and should be conducted according to OECD Guidelines 105 (Water
Solubility), 112 (Dissociation Constants in Water), 101 (UV-Visible Absorption Spectrum), 104
(Vapour Pressure) and 107/117/123 (n-Octanol/Water Partition Coefficient).
Some precautions must be taken regarding the use of the shake-flask method (OECD 107) or the HPLC
method (OECD 117) to determine log Kow for very lipophilic compounds. These are outlined in the
Globally Harmonized System of Classification and Labelling of Chemicals:
“The shake-flask method is recommended when the log Kow value falls within the range from –2 to 4.
The shake-flask method applies only to essential pure substances soluble in water and n-octanol. For
highly lipophilic substances, which slowly dissolve in water, data obtained by employing a slowstirring method are generally more reliable. Furthermore, the experimental difficulties, associated with
the formation of microdroplets during the shake-flask experiment, can to some degree be overcome by
a slow-stirring method where water, octanol, and test compound are equilibrated in a gently stirred
reactor. With the slow-stirring method (OECD 123) a precise and accurate determination of Kow of
compounds with log Kow of up to 8.2 is allowed. As for the shake-flask method, the slow-stirring method
applies only to essentially pure substances soluble in water and n-octanol. The HPLC method, which is
performed on analytical columns, is recommended when the log Kow value falls within the range 0 to 6.
The HPLC method is less sensitive to the presence of impurities in the test compound compared to the
shake-flask method.”
It should also be emphasised that the log Kow for ionisable substances should be measured on the nonionised form at environmentally relevant pHs.
2.1.1.2.
Environmental fate studies
Biodegradation studies should be performed in soil for additives intended for use in terrestrial species
and in aquatic systems for additives intended for aquatic animals. The Soil Adsorption/Desorption test
should be used for additives for both terrestrial and aquatic species as long as there is no validated test
for sediment.
OECD Guidelines 106/121(Soil Adsorption/Desorption), 307 (Soil Biodegradation (route and rate) or
308 (Degradation in Aquatic Systems) should be followed.
Adsorption/desorption studies should report both the Koc and Kd values for a range of soils. The OECD
121 guideline to determine the log Koc by means of HPLC should be used with care. For polar
compounds especially, the method is not fully validated and may provide unreliable K oc values. Also
log Koc values > 5.6 should not be considered to be reliable. For this reason, the OECD 106 test method
is recommended, especially for ionisable substances.
Since feed additives can be large molecules with several functional groups and a tendency to speciate
into ionic species around environmental pH values, other soil components with polar and/or charged
The EFSA Journal (2008) 842, 10-28
Technical guidance – Environmental Risk Assessment
surfaces may also act as sorbents. Additionally, sorption behaviour is known to be strongly pHdependent. If this is confirmed by further studies on the sorption behaviour of additives, models need to
be adapted to account for additional sorbents and pH-dependence of sorption.
For feed additives used in mariculture it may be appropriate to do the Degradation in Aquatic Systems
study (308) under saltwater conditions.
2.1.1.3.
Toxicity tests
Terrestrial compartment
One nitrogen transformation test on soil micro-organism (28 days), one acute toxicity test on
earthworms and one growth test in three different terrestrial plants are required.
Tests required are described in and should be conducted according to OECD Guidelines 216 (Soil
Microorganisms, Nitrogen Transformation Test (28 days)), 208 (Terrestrial Plants, Growth Test) and
207 (Earthworm, Acute Toxicity Test).
The earthworm acute toxicity test is considered to be relatively insensitive. Therefore, it is advisable to
perform an Earthworm Subacute/Reproduction test (OECD 220/222) already at this stage of the
assessment, in particular when the substance is persistent in soil.
Fresh water compartment
For feed additives to be used in terrestrial animals and freshwater aquaculture, one study each on Algal
Growth Inhibition, Daphnia Immobilization and Fish Acute Toxicity and Sediment-Water Chironomid
Toxicity Test are required.
OECD Guidelines 201 (Algal Growth Inhibition), 202 (Daphnia Immobilization), 203 (Fish Acute
Toxicity) or 218 (Sediment-Water Chironomid Toxicity Test) should be followed.
The composition of the sediment used for the tests depends on the requirements of the test species and
should therefore follow that in the respective test methods. The use of artificial sediment is
recommended. However, if there is experience with special natural sediments, these can also be used
for the test as long as the properties of the sediment are described in detail.
The organic carbon content of sediment may influence bioavailability and consequently the toxicity of
the test substance. Therefore, for comparison of sediment tests, the organic carbon content of the test
sediment should be within a certain range. The OECD guideline 218 (Sediment-Water Chironomid
Toxicity Test) for the test with Chironomus using spiked sediment recommends an organic carbon
content of the test sediment of 2 % (+/- 0.5 %). For the risk characterisation it is recommended to
normalise the PNEC to the organic carbon content used in the calculation of the PEC in sediment.
Marine compartment
For feed additives used in mariculture, it is recommended to test three marine sediment species. At
present, no internationally accepted, i.e. ISO or OECD, guidelines are available. However, there are
relevant guidelines available from the American Society for Testing of Materials (ASTM) for toxicity
in salt water systems which can be considered appropriate. Those include Standard Guide for
Conducting Life-Cycle Toxicity Tests with Saltwater Mysids (E1191-03a), Standard Test Method for
Measuring the Toxicity of Sediment-Associated Contaminants with Estuarine and Marine Invertebrates
(E1367-03), Standard Guide for Conducting Sediment Toxicity Tests with Marine and Estuarine
Polychaetous Annelids (E1611-00) and Standard Guide for Conducting Renewal Microplate-Based
Life-Cycle Toxicity Tests with a Marine Meiobenthic Copepod (E2317-04).
The US EPA has also published a number of useful harmonised guidelines for assessing environmental
effects, including salt-water tests. Those can be obtained from the webpage of the Office of Prevention,
Pesticides and Toxic Substances (OPPTS). A harmonised guideline within Series 850 (Ecological
Effect Test Guidelines) is also available.
The EFSA Journal (2008) 842, 11-28
Technical guidance – Environmental Risk Assessment
2.1.2. PEC refinement for soil
2.1.2.1.
Refinement based on metabolism
The PECsoil can be refined (PECsoil refined) by determining the actual composition of the excreted residue.
Metabolites representing less than 10 % of the administered dose can be subtracted from the total dose
administered. In addition, the biological activity of metabolites compared to the parent compound can
be considered. This procedure will result in the calculation of the fraction of the administered dose still
considered to be active.
The PECsoil calculated in Phase I and used initially in Phase IIA can be refined as shown:
PEC soil refined
PEC soil initial Fa
where:
Symbol
PECsoil refined
PECsoil
Fa*
Parameter
Refined concentration of the additive (parent compound) in soil
Concentration of the additive (parent compound) in soil in Phase I
Fraction of the dose considered to be active
Unit
g kg-1soil
g kg-1soil
-
* [value between 0 and 1]
2.1.2.2.
Refinement based on degradation in manure
There are at present no guidelines for degradation studies in manure. Should studies be conducted, they
should satisfy the following criteria:
The test should preferably be carried out using labelled material although unlabelled test
compound may be used if justified;
It is important, when using unlabelled compound, that 100 % of the dose can be accounted for;
The test should be carried out in the manure of the target species;
The relevant temperature for the test manure is 20 °C for pigs, 10 °C for cattle and 25 °C for
chickens and horses (if for the risk assessment other temperatures have to be used, then the
degradation rates can be converted to the temperature needed using the Arrhenius-equation);
The manure from pigs and cattle should be incubated wet/anaerobic; manure from chickens
should be incubated dry/aerobic.
Additional guidance is given by Van Vlaardingen et al. (2001).
In order to determine the residues of feed additives at the end of the storage time for manure, it should
be noted that for most animal categories (broilers and calves for fattening, for instance), the manure
production increases with the age of the animal. Concurrently, most feed additives will be excreted at
the end of the production cycle. Any calculation method to refine the PEC in manure should take this
fact into account. The period of administration of the additive and its proximity to the end of the
production cycle should be taken into consideration in refining the PEC manure.
As the storage capacity shows a large variation among the different EU Member States, it is
recommended to set the storage capacity/time equal to the production period of the target animal up to
three months, unless the number of cycles is more than four per year. In this case the storage time is set
equal to the period of the cycle.
If degradation is to be considered in Phase II, the PECmanure should be calculated for a storage time
similar to one animal production cycle and, by doing, so the amount of manure is also set equal to the
amount produced in that storage period, which fills the annual nitrogen quota of 170 kg N ha-1. It is also
necessary to consider that the animals could be given a feed additive at a particular period. If animals
are given a feed additive at the beginning of the storage period there will be more time for the active
ingredient to degrade than if they were given the additive at the end of the storage period. For this
The EFSA Journal (2008) 842, 12-28
Technical guidance – Environmental Risk Assessment
reason, the time for degradation of the active substance is taken to be half the storage time of the
manure.
To calculate the PECsoil by taking into account the degradation during storage, the following equations
can be used:
Cadd FI total
e
N excreted
PEC manure
kT st / 2
ln 2
DT50
k
PEC manure Q
PEC soil
RHOsoil CONVarea field DEPTH field
where:
Symbol
Parameter
Default Unit
Value*
Input
Cadd
Concentration of the additive in feed
FItotal
Nexcreted
RHOsoil
DEPTHfield
Total feed intake (DM) per year
Total N excretion per year
Bulk density of soil
Mixing depth with soil
CONVarea field Conversion factor for the area of the agricultural field
Q
Annual nitrogen immission standard
DT50
Half-life of the additive in manure
K
Rate constant
Tst
Length of time manure is stored
Intermediate results
PECmanure
Concentration of the additive (parent compound) in
manure expressed per amount nitrogen
Output
PECsoil
Highest concentration of the additive (parent compound)
in the soil
1500
20
5
10000
170
mg kg feed-1 complete
feed
kg feed
kg N
kg m-3
cm (arable land)
cm (grassland)
m2 ha-1
kg N ha-1
day
day
mg kg-1 N
mg kg-1 soil
* The use of the indicated default values in the equations is recommended. Reasons for any deviations from these values
should be given by the applicant.
2.1.2.3.
Refined PEC in soil for persistent compounds
If a high persistence in soil is anticipated (DT90 > 1 year), the potential for residues to accumulate in
soil should be considered. In those cases, the PECsoil plateau at steady state should be calculated at the start
of Phase IIA as follows:
PEC soil 1 year
Fd
PEC soil initial e
PEC soil initial
0.693 365
DT50
PEC soil1 year
PEC soil initial
PEC soil plateau
PEC soil initial
Fd
The EFSA Journal (2008) 842, 13-28
Technical guidance – Environmental Risk Assessment
where:
Symbol
Input
DT50
PECsoil initial
Intermediate results
Fd
Output
PECsoil 1 year
PECsoil plateau
Parameter
Unit
Half-life of additive (parent compound) in soil
Concentration of the additive (parent compound)
immediately after spreading
day
g kg-1
Fraction of additive (parent compound) degraded in 1 year
-
Concentration of the additive (parent compound) 1 year
after spreading
PECsoil at plateau concentration
g kg-1soil
g kg-1 soil
The PEC in soil can be refined based on either information related to the metabolism of the substance
in the target animals or degradation in manure or soil.
2.1.2.4.
Refinement based on degradation in soil
Refinement of PECsoil based on soil degradation data is possible when it is realistic to assume that
manure is spread in more than one spreading event. In that case, the concentration calculated after the
last spreading event should be taken.
In the case of arable land, manure/slurry is usually applied to fulfil the permissible limit during a single,
annual application event. This partly reflects the fact that the presence of a crop will prevent
applications of manure/slurry throughout much of the year.
In the case of grassland, it is more typical to make a number of applications of manure/slurry
throughout the year, with the total amount of nitrogen applied adding up to equal the annual permissible
limit. It is up to the applicant to provide information to support the number of spreading events which
have been taken to occur on grassland.
The following formula can be used to calculate the PECsoil after the last spreading event:
( Nspreading )
PEC soil refined
PEC soil single event
Frs
e
k
1 Frs
1 Frs
k Tinterval spreading
ln 2
DT50
where:
Symbol
Input
PECsoil single-event
Parameter
Unit
Concentration of the additive (parent compound) in soil immediately after
spreading
Nspreading
Number of spreading events
Tinterval spreading
Time between spreading events
DT50
Half-life of additive (parent compound) in soil
k
Rate constant
Intermediate results
Frs
Fraction remaining in soil after time T interval spreading
Output
PECsoil refined
Refined Concentration of the additive (parent compound) in soil after last
spreading event
g kg-1soil
day
day
g kg-1
The EFSA Journal (2008) 842, 14-28
Technical guidance – Environmental Risk Assessment
2.1.3. PEC in surface water (PECsw) and fresh water sediment (PECfw sed) for additives used in
terrestrial animals
2.1.3.1.
PEC in surface water (PECsw)
As first estimate for the surface water concentration resulting from run-off or drainage, it is assumed
that one part run-off water will be diluted by two parts receiving water (Note for guidance:
environmental risk assessment for veterinary medicinal products other than GMO-containing and
immunological products). The concentration in run-off water is calculated using the formula given for
pore water but using soil depth with which the initial PEC soil has been calculated. The PECsw
corresponds to one third of the PEC for run-off water.
2.1.3.2.
PEC in fresh water sediment (PECfw sed)
Concentrations in sediment can be determined by the concentrations in water and the sediment-water
partitioning coefficient, using the following equations:
PEC fwsed
K sed water
RHOsed
K sed
Fwatersed
water
CONVsed
Kp sed
PEC surface water 1000 CONVsed
Fsolidsed
Kp sed
RHOsolid
1000
RHOsed
Fsolidsed RHOsolid
Focsediment K oc
where:
Symbol
Input
RHOsed
RHOsolid
PECsurfacewater
CONVsed
Fwatersed
1000
Fsolidsed
Koc
Focsed
Ksed-water
Kpsed
Output
PECfw sed
Parameter
Default Value* Unit
Wet bulk density of sediment
Bulk density of solids
Concentration of the additive (parent compound) in surface
water
Conversion factor for sediment concentrations: wwt to dwt‡
Volume fraction of water in sediment
Conversion for litre to m3
Volume Fraction of solids in sediment
Organic carbon partition coefficient
Weight fraction organic carbon in sediment
Sediment-water partition coefficient
Partition coefficient solids and water in sediment (v/w)
1300
2500
Concentration of the additive (parent compound) in sediment
0.8
0.2
0.05
kgwwt m-3
kgdwt m-3
µg L-1
kgwwt kgdwt-1
m3 m-3
L m-3
M3 m-3
L kg-1
kg kg-1
M3 m-3
L kg-1
g kgdwt-1
* The use of the indicated default values in the equations is recommended. Reasons for any deviations from these values
should be given by the applicant.
‡
If the PECfw sed has to be expressed on a wet weight basis, the expression CONVsed is omitted from first equation.
2.1.4. Advanced models for calculating PEC
The simple equations described above provide worst case estimates of the likely concentrations of the
additive in groundwater (see Section 1.2.2) and surface waters (see Section 2.1.3.1). If Risk Quotient
(RQ) values for surface water organisms are > 1 and/or the PECgw is > 0.1 μg L-1 then it is advisable to
use a more advanced model to predict the movement of the additive to groundwater and surface waters.
The EFSA Journal (2008) 842, 15-28
Technical guidance – Environmental Risk Assessment
More sophisticated models have been developed by the FOCUS (Forum for the Coordination of
Pesticide Fate Models and Their Use) group. Justification for using these models is given in the EFSA
opinion on the development of an approach for the environmental risk assessment of additives, products
and substances used in animal feed.
2.1.4.1.
Groundwater
Groundwater calculations developed by FOCUS involve the simulation of the leaching behaviour of
agrochemicals using a set of four models (PEARL, PELMO, PRZM and MACRO) in a series of up to
nine geographic settings with various combinations of crops, soils and climate. Groundwater
concentrations are estimated by determining the annual average concentrations in shallow groundwater
(1m soil depth) for a period of 20 consecutive years, rank ordering the annual average values and then
selecting the 80th percentile value.
When using the FOCUS models, a simple first step of this assessment can be based on a realistic worst
case FOCUS scenario. For reasons given in the EFSA opinion, it seems most appropriate to base such a
leaching assessment on the FOCUS Okehampton scenario using PEARL.
In order to simplify the first step in the refined exposure assessment, calculations were performed with
FOCUS_PEARL v3.0 applying a dose of 1 kg ha-1 on 3 October every year over a 20-year period. The
dose was incorporated into the top 20 cm of soil. The crop was winter cereal. All substance properties
except KOM and DT50 were equal to the model substance D as defined by FOCUS. Runs were carried out
with 90 KOM - DT50 combinations covering FOCUS leaching concentrations ranging from 0.001 to
about100 μg L-1. The results were fitted to a metamodel to be able to estimate leaching concentrations
without running a FOCUS scenario (EFSA opinion). Based on this analysis, the following inequalities
can be used for the first-tier leaching assessments of feed additives.
Table 2.
Requirements for the KOM (= Koc/1.7) as a function of the FOCUS leaching
concentration
CFOCUS (μg L-1)
Requirement for the KOM
0.01
0.1
1
10
KOM > -5.9 + 9.1 DT50
KOM > -5.9 + 6.5 DT50
KOM > -5.9 + 3.8 DT50
KOM > -5.9 + 1.2 DT50
Note that these relationships are based on a dose of 1 kg ha-1. In the event that the actual dose is
substantially lower or higher then a less or more stringent relationship should be used in proportion to
the dose (e.g. when the dose is < 0.1 kg ha-1, the relationship KOM > -5.9 + 3.8 DT50 can be used to
ensure the leaching concentrations is < 0.1 µg L-1).
If it is not possible to exclude the likelihood that groundwater concentration is > 0.1 μg L-1 based on the
metamodel, then it is necessary to run the PEARL model using the scenarios recommended in the
EFSA opinion. Table 3 indicates which scenarios have to be run for the specific target animals, taking
into account the indicated considerations.
Table 3.
Proposed FOCUS scenarios for PECgw calculation of feed additives
Target animal
Bovine
Ovine
Swine
Avian
FOCUS GW
N: Jokioinen
S: Sevilla, Piacenza
C: Okehampton
S: Sevilla, Thiva
N: Jokioinen
S: Piacenza
N: Jokioinen
S: Piacenza
N: Northern/Scandinavian; C: Central; S: Southern/Mediterranean
The EFSA Journal (2008) 842, 16-28
Technical guidance – Environmental Risk Assessment
Settings of the FOCUS model for groundwater
As explained above, application to arable land is most typically carried out in the early autumn. In order
to standardise, the exposure assessments timing of application to soil is assumed to coincide with
drilling of winter cereals (in the absence of pure grassland scenario) as these crops are typically grown
throughout Europe and represent a significant input of manures on a total mass basis across Europe.
The DT50 values should be the average values from the experimental data. In Section 2.1.1.2, guidance
is given to select the most appropriate Koc value.
It is assumed that manure will be applied at a rate of 170 kg N ha-1 in one spreading event. As the input
in FOCUS is expressed in kg ha-1, the PECsoil has to be converted to kg ha-1 before running the FOCUS
model. Recommended input parameters on the application of FOCUS model is presented in the
Appendix.
2.1.4.2.
Surface Water
The surface water and sediment calculations developed by FOCUS include three progressively refined
tiers of evaluation, ranging from initial spreadsheet-based evaluations of potential aquatic
concentrations to more detailed mechanistic calculations of drift, runoff, erosion and field drainage
loaded into a series of small water bodies. The surface water and sediment calculations are performed
using an overall calculation shell called SWASH which controls models that simulate runoff and
erosion (PRZM), leaching to field drains (MACRO), spray drift (internal in SWASH) and finally
aquatic fate in ditches, ponds and streams (TOXSWA). Those simulations provide detailed assessments
of potential aquatic concentrations in a range of water body types in up to ten separate geographic and
climatic settings.
Detailed explanations of the FOCUS models as well as the modelling scenarios, key assumptions,
required modelling inputs and model outputs are provided in the respective FOCUS modelling reports
(EFSA opinion). The FOCUS surface water and groundwater models have been placed on a website
(viso.ei.jrc.it/focus/index.htm) where they can be freely downloaded.
Based on the EFSA opinion, the runoff and drainage scenarios given in Table 4 were identified as
potential „base-set‟ scenarios:
Table 4.
Proposed FOCUS SW scenarios for PECsw calculation of feed additives
Target animal
Bovine
Ovine
Swine
Avian
FOCUS SW scenario (Drainage)
D4
D6
D4, D3
D5, D3
FOCUS SW scenario (runoff)
R1, R3
R4
R1, R3
R1, R3
This selection covers not only the areas identified by FOCUS but also several areas in the member
states that joined the EU after May 2005 and is supported by study carried within ERAPharm Project
(Schneider et al., 2007).
Settings of the FOCUS model for surface water
As proposed for groundwater, the application of manure to arable and grass land is considered to
coincide with the drilling of cereals in autumn (in the absence of a pure grassland scenario). The DT50
values should be the average values from the experimental data. In section 2.1.1.2, guidance is given to
select the most appropriate Koc value. In order to select the most appropriate date, the FOCUS PAT tool
should be used. As a realistic worst case, it is assumed that manure will be applied at a rate of 170 kg N
ha-1 in one spreading event. Without information on the degradation in a water/sediment, the
degradation rate is set to zero. As mentioned for groundwater as the input in FOCUS is expressed in kg
ha-1, the PECsoil has to be converted to kg ha-1 before running the FOCUS model. Recommended input
parameters on the application of FOCUS model are presented in the Appendix.
The EFSA Journal (2008) 842, 17-28
Technical guidance – Environmental Risk Assessment
2.1.4.3.
Interpretation of results from FOCUS
In FOCUS groundwater models the 80th percentile value for a 20-year period is presented.
The results for surface water are presented as the maximum predicted PECsw and the time of occurrence
of the peak. The decline in concentrations after the peak is presented graphically.
In principal, PECgw and PECsw values for all scenarios should be compared with the limit concentration
of 0.1 µg L-1 for groundwater and the PNEC values for the aquatic organisms, respectively. For
scenarios where the trigger value for groundwater is exceeded or the RQ values are
1, further
consideration of the results in relation to the proposed use of the product will be needed. It is up to the
applicant to present further assessment which may involve more modelling, more studies or relevant
arguments as to why exceeding the trigger value for groundwater or the RQ for aquatic organisms is not
indicative of an unacceptable risk.
For further guidance to investigate leaching to groundwater under field conditions, the reader is
referred to the OECD Guidance Document (EFSA opinion).
2.1.5. PEC refinement for marine sediment (aquaculture)
A more refined PEC can be modelled for marine sediment taking into account the degradation in
sediment:
PEC sed refined
PEC faeces K dep
RHOsed DEPTH sed
Frs
kdegsed
CONVsed
e
T
1 Frs production
CONV sed
1 Frs
kdegsed
ln 2
DT 50degsed
RHOsed
Fsolidsed RHOsolid
where:
Symbol
Input
Tproduction
kdep
DT50degsed
Parameter
Number of production days
Maximum deposition rate of faeces
Half-life time of additive (parent compound) in
sediment
RHOsed
Fresh bulk density sediment
DEPTHsed
Mixing depth in sediment
Intermediate results
kdegsed
Reaction rate for transformation in sediment
Frs
Fraction remaining in sediment
PECfaeces
Concentration of the additive (parent compound) in
faeces
Output
PECsed refined
Refined concentration of the additive (parent
compound) in sediment
Default
Value*
Unit
365
0.01
day
kg carbon m-2 day-1
day
1300
20
Kgwwt m-3
cm
day-1
mg kg carbon
mgc kg w wt-1
* The use of the indicated default values in the equations is recommended. Reasons for any deviations from these
values should be given by the applicant.
The metabolic fate of the additive in fish and other processes that may change its bioavailability can
also be considered for PEC refinement, where data are available.
The EFSA Journal (2008) 842, 18-28
Technical guidance – Environmental Risk Assessment
2.2.
Phase IIB
For those additives where, following Phase IIA assessment, an environmental risk cannot be excluded,
more information is required on the effects on biological species in the relevant environmental
compartment(s). In this situation, further tests are needed to determine the chronic and more specific
effects on appropriate microbial, plant, and animal species. This additional information will allow the
application of a lower safety factor.
Suitable additional ecotoxicological tests are described in a number of publications, e.g., in OECD
guidelines. Careful choice of such tests is necessary to ensure that they are appropriate to the situation
in which the additive and/or its metabolites may be released and dispersed in the environment. The
refinement of the effect assessment for soil (PNECsoil) may be based on studies on the chronic effects on
earthworms, additional studies on soil microflora and a number of relevant plant species, studies on
grassland invertebrates (including insects) and feral birds.
The refinement of the effect assessment for water/sediment may be based on chronic toxicity tests on
the most sensitive aquatic/benthic organisms identified in Phase IIA assessment.
Bioaccumulation studies, if necessary, shall be performed according to OECD guideline 305
(Bioconcentration: Flow-through Fish Test).
2.2.1. Toxicity tests: Terrestrial compartment
In order to refine the effect assessment for the soil compartment, studies based on the OECD
Guidelines 216 (Soil Microorganisms, Nitrogen Transformation Test, 100 days), 208 (Terrestrial
Plants, Growth Test, Additional species) and 220/222 (Earthworm Reproduction Test) are required.
2.2.2. Toxicity tests: Fresh water compartment
In order to refine the effect assessment for the freshwater compartment, studies based on the OECD
Guidelines 211 (Daphnia magna Reproduction), 210 (Fish, Early-life Stage) and 225 (Sediment-Water
Lumbriculus Toxicity Test) are required.
2.2.3. Toxicity tests: Sediment aquaculture
In order to refine the effect assessment for the marine compartment, tests can be selected that were not
completed during Phase IIA. Other appropriate tests are acceptable. For the assessment of feed
additives used in marine aquaculture, emphasis should still be on sediment dwelling organisms.
3.
PNEC derivation
PNECs are based on the experimental effect end-points of the toxicity studies applying appropriate
assessment factors depending on the type of toxicity data submitted.
3.1.
Surface water
In Phase IIA, the assessment of PNEC is normally based on acute toxicity data. In this case, an
assessment factor of 1000 will be applied to the lowest LC50/EC50 of the relevant available toxicity data
(see notes to Table V.1, EFSA opinion). If for the same species more than one LC50/EC50 value is
available, the geometric mean is used. The Algal Growth Inhibition Test (OECD 201) of the base-set is,
in principle, a multi-generation test. However, for the purposes of applying the appropriate assessment
factors, this EC50 is treated as a short-term toxicity value.
When a risk cannot be excluded based on Phase IIA assessment, additional long-term tests with a
relevant test organism can be conducted at Phase IIB to reduce the uncertainty and, therefore also the
assessment factor (Table 5). For specific comments on the use of assessment factors in relation to the
available data set and justification for changing the assessment factor, reference is made to the EU
TGD for existing and new chemicals and to EFSA opinion.
The EFSA Journal (2008) 842, 19-28
Technical guidance – Environmental Risk Assessment
Table 5.
Assessment factors to derive a PNECaquatic
Available data
Assessment factor
At least one short-term L(E)C50 from each of three trophic
levels of the base-set (fish, Daphnia and algae)
One long-term NOEC (either fish or Daphnia)
Two long-term NOECs from species representing two trophic
levels (fish and/or Daphnia and/or algae)
Long-term NOECs from at least three species (normally fish,
Daphnia and algae) representing three trophic levels
Species sensitivity distribution (SSD) method
1000 (a)
Field data or model ecosystems
100 (b)
50 (c)
10 (d)
5-1
(to be fully justified case by case) (e)
Reviewed on a case by case basis (f)
(a), (b), (c), (d), (e), (f), see notes to Table V.1, EFSA opinion
The refined PNEC based on chronic toxicity data has to be compared with the chronic exposure levels.
These can only be determined using the FOCUS models. For this purpose the time weighted average
PEC should be set equal to chronic exposure time of the most sensitive species tested.
3.2.
Terrestrial environment
For the derivation of the PNEC for terrestrial organisms, the same effect assessment is followed as
performed for veterinary drugs (Guideline on environmental impact assessment for veterinary
medicinal products Phase II), which means that separate assessment factors are applied to every
taxonomic group. The lowest PNEC determines the PNEC for the terrestrial compartment.
3.2.1. Terrestrial plants
At Phase IIA, information on the toxicity to three terrestrial plant species is required. The PNEC for
plants is then derived by applying an assessment factor of 100 to the lowest EC50 value. If a risk is
identified, at Phase IIB the study should be repeated on two additional species from the most sensitive
species category, in addition to repeating the study on the most sensitive species. Subsequently, the
PNEC is derived by applying an assessment factor of 10 to the lowest NOEC value.
3.2.2. Earthworms
At phase IIA, the effect assessment for earthworm can be based on an acute toxicity study. If the feed
additive is not persistent (DT50 < 60 days), the PNEC is derived by applying an assessment factor of
100 to the LC50 value. In cases where the additive is persistent (DT50 > 60 days), an assessment factor
of 1000 is applied.
If based on the acute toxicity a risk cannot be excluded, at Phase IIB an earthworm
subacute/reproduction test (OECD 220/222) should be conducted. When such a study is available, the
PNEC is derived by applying a safety of 10 to the NOEC value.
3.2.3. Micro-organisms
The Soil Microorganisms, Nitrogen Transformation Test (OECD 216) should be conducted at 1X and
10X the PEC. At phase IIA, this study is conducted during a period of 28 days. If, on day 28,
differences between treated and untreated soils are equal to or greater than 25 %, at Phase IIB
measurements have to be continued to a maximum of 100 days. When the difference in the rates of
nitrate formation between the maximum PEC and control is equal to or less than 25 % at any time point
after day 28, the product can be evaluated as having no long-term influence on nitrogen transformation
in soils.
The EFSA Journal (2008) 842, 20-28
Technical guidance – Environmental Risk Assessment
3.2.3.1.
Normalisation
Natural soils used in ecotoxicological tests could differ in characteristics such as organic matter and
clay content, soil pH and soil moisture content. The bioavailability of the test compound, and therefore
the toxicity observed, could be influenced by those soil properties. This means that results from
different test soils cannot be compared directly. If possible, data should be normalised using
relationships that describe the bioavailability of chemicals in soils. If there is evidence that the
bioavailability of the compound is related to the organic matrix, results are converted to a standard soil,
which is defined as a soil with an organic matter content of 3.4 % or an organic carbon content of 2.0
%.
3.3.
Sediment
3.3.1. Marine sediment
If results are only available from short-term tests with sediment dwelling organisms, an assessment
factor of 1000 is applied to the lowest value. However, results from long-term tests with sub-lethal
endpoints, such as reproduction, growth, emergence, sediment avoidance and burrowing activity, are
regarded as most relevant due to the generally long-term exposure of benthic organisms to sedimentbound substances. In such cases, an assessment factor of 10 is applied to the lowest NOEC value. For
additional guidance, reference is made to the EU TGD for existing and new chemicals and to the EFSA
opinion.
3.3.2. Fresh water sediment
As the available Regulation (EC) No 429/2008 investigates long-term effects, the relevant endpoints are
expressed as NOEC values. For deriving the PNEC, the assessment factors of 10 is applied to the
lowest NOEC value.
Additional indication of the PNEC for sediment organisms can be obtained from equilibrium
partitioning of the additive and data on toxicity of the additive during waterborne exposures. However,
this is not a substitute for effects testing of sediment dwelling organisms, unless there is a sufficient
margin of safety (PEC/PNEC < 0.1). The equilibrium partitioning is based on the following equation:
PNEC sed
K sed
water
K sed water
RHOsed
Fwatersed
CONVsed
Kp sed
PNEC surface water 1000 CONV sed
Fsolidsed
Kp sed
RHOsolid
1000
RHOsed
Fsolidsed RHOsolid
Focsed
K oc
The EFSA Journal (2008) 842, 21-28
Technical guidance – Environmental Risk Assessment
where:
*
Symbol
Input
Ksed-water
RHOsed
RHOsolid
PNECsurfacewater
CONVsed
Parameter
Fwatersed
1000
Fsolidsed
Kpsed
Koc
Focsed
Output
PNECsed
Volume fraction of water in sediment
Conversion for litre to m3
Volume Fraction of solids in sediment
Partition coefficient solids and water in sediment (v/w)
Organic carbon partition coefficient
Weight fraction organic carbon in sediment
Default Value* Unit
Sediment-water partition coefficient
Bulk density of sediment
Bulk density of solids
Predicted No Effect Concentration for aquatic organisms
Conversion factor for sediment concentrations: wwt to dwt
1300
2500
0.8
0.2
0.05
Predicted No Effect Concentration for sediment dwelling
organisms
m3 m-3
kgwwt m-3
kgdwt m-3
µg L-1
kgwwt
kgdwt-1
m3 m-3
L m-3
M3 m-3
L kg-1
L kg-1
kg kg-1
g kgdwt-1
The use of the indicated default values in the equations is recommended. Reasons for any deviations from these values
should be given by the applicant.
If the PNECsed has to be expressed on a wet weight basis, the expression CONVsed is omitted from the
first equation. For substances with a log Kow > 5, the PEC/PNEC has to be < 0.01 in order to take into
account the possible uptake via ingestion of sediment.
The EFSA Journal (2008) 842, 22-28
Technical guidance – Environmental Risk Assessment
GLOSSARY (DEFINITION OF TERMS)
BCF
Cadd
CF
CONVarea field
CONVsed
CONVsed
DEPTHfield
DEPTHsed
DF
DT50
DT50degsed
DT90
EC50
=
=
=
=
=
=
=
=
=
=
=
=
=
Fa
Fairsoil
Fd
FItotal
Flow
Focsed
Focsoil
FOCUS
FR
Fret
Frs
Frs
Fsolidsed
Fsolidsoil
Fwatersed
Fwater-soil
k
Kair-water
Kd
kdegsed
kdep
Koc
Kow
Kpsed
Kpsoil
Ksed-water
Ksoil-water
LC50
MOLW
Nexcreted
NOEC
Nspreading
OECD
PAT
PCfaeces
PECfaeces
PECfw sed
PECmanure
PECporewater
PECsed
PECsed refined
PECsoil
PECsoil 1 year
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
Bioconcentration Factor
Concentration of the additive (parent compound) in feed
Conversion factor (kg feed to kg carbon in faeces)
Conversion factor for the area of the agricultural field
Conversion factor for sediment concentrations: wwt to dwt
Conversion factor for sediment concentrations: wwt to dwt
Mixing depth with soil
Mixing depth in sediment
Dilution Factor
Half-life of additive
Half-life time of additive (parent compound) in sediment
Time to degradation of 90% of original concentration of the compound in the tested soils
The concentration of a test substance which results in 50% of the test animals being adversely
affected, i.e., both mortality and sub-lethal effects
Fraction of the dose considered to be active
Fraction air in soil
Fraction of additive (parent compound) degraded in 1 year
Total feed intake (DM) per year
Water flow rate through the system
Weight fraction organic carbon in sediment
Weight fraction organic carbon in soil
The FOrum for Co-ordination of pesticide fate models and their USe
Feed Ration
Fraction of retention in the system
Fraction remaining in soil after time T interval spreading
Fraction remaining in sediment
Volume Fraction of solids in sediment
Fraction solids in soil
Volume fraction of water in sediment
Fraction water in soil
Rate constant
Partition coefficient air and water in soil
Sorption/desorption coefficient
Reaction rate for transformation in sediment
Maximum deposition rate of faeces
Sorption/desorption coefficient, normalized to organic carbon content
n-Octanol/water partitioning coefficient
Partition coefficient solids and water in sediment (v/w)
Partition coefficient solids and water in soil (v/w)
Sediment-water partition coefficient
Partition coefficient solids and water in soil (v/v)
The concentration of a test substance which results in a 50% mortality of the test species
Molar mass
Total N excretion per year
No-observed effect concentration, i.e., the test concentration at which no adverse effect occurs
Number of spreading events
Organization for Economic Co-operation and Development
Pesticide Application Timer
Concentration of the additive (parent compound) in faeces
Predicted concentration of the additive (parent compound) in faeces
Concentration of the additive (parent compound) in fresh water sediment
Concentration of the additive (parent compound) in manure expressed per amount nitrogen
Concentration of the additive (parent compound) in porewater
Concentration of additive (parent compound) in sediment
Refined concentration of the additive (parent compound) in sediment
Concentration of the additive (parent compound) in soil
Concentration of the additive (parent compound) 1 year after spreading
The EFSA Journal (2008) 842, 23-28
Technical guidance – Environmental Risk Assessment
PECsoil initial
PECsoil plateau
PECsoil refined
PECsoil single-
=
=
=
=
Concentration of the additive (parent compound) immediately after spreading
PECsoil at plateau concentration
Refined concentration of the additive (parent compound) in soil
Concentration of the additive (parent compound) in soil immediately after spreading
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
Concentration of the additive (parent compound) in surface water
Highest initial concentration of additive (parent compound) in surface water
Predicted No Effect Concentration for sediment dwelling organisms
Predicted No Effect Concentration for aquatic organisms
Pesticide Root Zone Model
Annual nitrogen immission standard
Gas constant
Fresh/Wet bulk density sediment
Bulk density of soil
Bulk density of solids
Water solubility
Surface WAter Scenarios Help
Temperature at air-water interface
Time between spreading events
TOXic substances in Surface WAters
Number of production days
Length of time manure is stored
Vapour pressure
event
PECsurfacewater
PECswaq
PNECsed
PNECsurfacewater
PRZM
Q
R
RHOsed
RHOsoil
RHOsolid
SOL
SWASH
TEMP
Tinterval spreading
TOXSWA
Tproduction
Tst
VP
The EFSA Journal (2008) 842, 24-28
Technical guidance – Environmental Risk Assessment
REFERENCES
Bailey, J., 2003. Energy requirements and feeding behaviour of salmonids in culture. Doctoral diss.
Dept. of Aquaculture, SLU. Acta Universitatis agriculturae Sueciae. Silvestria, 264, 28
Corner, R.A., Brooker, A.J., Telfer, T.C. and Ross, L.G., 2006. A fully integrated GIS-based model of
particulate waste distribution from marine fish-cage sites. Aquaculture, 258(1-4), 299-311
EFSA, 2007. Opinion of the Scientific Panel on Additives and Products or Substances used in Animal
Feed on the development of an approach for the environmental risk assessment of additives,
products and substances used in animal feed. <http://www.efsa.europa.eu/EFSA/efsa_locale1178620753812_1178626084619.htm>
ERM/AB-DLO, 1999. Establishment of Criteria for the Assessment of Nitrogen Content in Animal
Manures, European Commission, Final Report November 1999
Hansen, P. K., Pittman, K., Ervik, A., 1991. Organic waste from marine fish farms-effects on the sea
bed. In: Marine Aquaculture and Environment. Ed.: T. Makinen. Copenhagen, Denmark: Nordic
Council of Ministers. pp. 105-119
Holmer, M., Diaz-Almela, E., Duarte, C.M., Tsepakis, M., Danovaro, R., 2006. Sedimentation of
organic matter from fish faros in oligotrophic Mediterrranean assessed through bula and stable
isotope analysis. Aquaculture. 26(2-4), 268-280
Hussenot, J., Lefebvre, S., Brossard, N., 1998. Open-air treatment of wastewater from land-based
marine fiih farms in extensive and intensive systems: current technology and future perspectives.
Aquat. Living Resour. 11, 297-304
Karakassis, I., Tsapakis, M., Smith, C.J. and Rumohr, H., 2002. Fish farming impact in the
Mediterranean studied through sediment profiling imagery. Mar. Ecol. Prog. Ser. 227, 125-133
Kutti, T., Kupka Hansen, P., Ervik, A., Høisæter, T., Johannessen, P., 2007. Effects of organic effluents
from a salmon farm on a fjord system II. Temporal and spatial patterns in infauna community
composition. Aquaculture 262, 355-366
Schneider, M.K., Stamm, C., Fenner, K., 2007. Selecting scenarios to assess exposure of surface waters
to veterinary medicines in Europe. Environ Sci Technol. 41(13), 4669-76
Van Vlaardingen, P.L.A., De Knecht, J.A., Janssen, P.A.H., 2001. Degradation of veterinary drugs in
manure. In: Luttik R, Van Raaij MTM, Factsheets for the (eco)toxicological risk assessment strategy
of the National Institute of Public Health and the Environment (RIVM). Bilthoven, The Netherlands:
RIVM, pp. 95-102
The EFSA Journal (2008) 842, 25-28
Technical guidance – Environmental Risk Assessment
APPENDIX
Application of FOCUS models
1. GROUNDWATER
Input parameters PEARL
1.1 Scenario:
Location:

pick one
Crop Calendar:

WCEREALS
Irrigation:

irrigation scenarios are considered for Chateaudun,
Piacenza, Sevilla, Thiva; No irrigation in the other cases.
Tillage:

No tillage

Repeat interval for application events (a):
1
1.2 Simulation Control:

01/01/1901

31/12/1926
Stop criterion (kg ha ):

default zero
Repeat hydrology:

no tick
Start date:
Stop date:
-1
Although the total time is 26 years, the protocol on the reactive tracer will be for only 20 years.
1.3 Output Control:
Summary report:

pick FOCUS report

Run SWAP and then PEARL only
No additional changes.
1.4 Swap Hydrological Method:
Option Hydrology:
No additional changes.
1.5 Substance:
General
Molar mass (g/mol):

enter value
Saturated vapour pressure (Pa):

enter value
Molar enthalpy of vapourisation (kJ/mol):

95 (default pesticides)
Solubility in water (mg/l):

enter value
Molar enthalpy of dissolution (kJ/mol):

27 (default pesticides)
Freundlich sorption
KOM:

enter value (KOM = KOC / 1.724)
The EFSA Journal (2008) 842, 26-28
Technical guidance – Environmental Risk Assessment
No additional changes.
Transformation

Half-life (d):
enter value
No additional changes.
Diffusion
No changes, use default settings from pesticides.
Crop
Wash-off factor (m-1):

≥ 10-6, even if there is no wash-off.
Coefficient for uptake by plant:

no uptake
1.6 Application
Advice should be given, which application form is most appropriate for feed additives. Since for feed
additives, either arable land or grassland without harvest are considered, absolute application seems
more appropriate than relative application.
As the input in FOCUS is expressed in kg ha-1, the PEC soil has to be converted using the following
equation (see also formula in Section 1.1.1):
ApplRate
PEC soil DEPTH field ROH soil
100000
Absolute applications
Application type:

either incorporation or application to the soil surface
Date:

enter date of application (pre-emergence)
Dosage (kg ha-1):

enter value
Depth (m):

default 20 cm (realistic worst case)
7. Deposition
No deposition
2. SURFACE WATER
SWASH
2.1 Actions/ Create view and edit substances
General:
Enter information on chemical properties (molar mass, vapour pressure, solubility in water,
metabolism).
For molar enthalpy of vaporisation and dissolution and diffusion coefficients in water and air the
default values from pesticides may be used.
The EFSA Journal (2008) 842, 27-28
Technical guidance – Environmental Risk Assessment
Maybe a short comment regarding the applicability of the default values especially to macromolecules
should be inserted, since these properties are generally assumed to be substance specific.
Sorption:
Enter either KOM or KOC, the other value will be calculated internally.
Enter Freundlich exponent. (The corresponding Freundlich exponent for soil or sediment is internally
calculated from the given KOM or KOC value and the fraction of organic matter in the soil of the chosen
scenario.)
Ref. concentration in the liquid phase [g m-³]: This refers to the concentration at which the sorption
parameters were determined. If it was at 1 g m-³, then the default value of 1 is correct. In case the
concentration was significantly different from 1 g m-³, the appropriate value should be inserted. This is
then used for internal correction of the Freundlich parameters.
Uptake and wash-off:
Do not assume any plant/ root uptake or wash off. Hence, set all parameters zero.
Transformation:
Enter DT50 in water, soil and sediment and the respective temperatures.
If you assume no transformation in the crop (or no data are available), set a large DT50 in crop (e.g.
103).
Effect of temperature: Use default value from pesticides if now data are available.
Specifications on transformation in soil: Use default values from pesticides for the dependence of
transformation on soil moisture/ water content.
2. 2 Focus wizard
Use Wcereals for crops selection. Although more realistic, a pure grassland scenario is not available.
Root uptake zero has to be set to zero (in the window “uptake and wash off”).
2.3 User defined wizard
Selected crop according to the chosen crop above.
Accept selected water body types.
Accept appropriate scenarios.
2.4 View projects and define applications
View and edit application: Enter number of applications, as well as the application mode (granular
application is the closest scenario to manure spreading). For run-off scenarios the depth of
incorporation is also required.
The EFSA Journal (2008) 842, 28-28