HC5S SSD Workshop Programme

User Manual: HC5S

Open the PDF directly: View PDF PDF.
Page Count: 56

Organised by
W
ORKSHOP
P
ROGRAMME
Estimating toxicity thresholds
for aquatic ecological communities
from sensitivity distributions
11-13 February 2014, Amsterdam
And
EUROPEAN CENTRE FOR ECOTOXICOLOGY AND TOXICOLOGY OF CHEMICALS
www.ecetoc.org
ORGANISING COMMITTEE
Scott Belanger
Procter & Gamble
Peter Craig
University of Durham
Scott Dyer
Procter & Gamble
Malyka Galay Burgos
ECETOC
Mick Hamer
Syngenta
Andy Hart
FERA
Stuart Marshall
Unilever
Paul Whitehouse
Environment Agency
CONTENTS
......................................................................................................................... 1 INTRODUCTION
................................................................... 2 PROGRAMME DAY 1: TUESDAY 11 FEBRUARY
............................................................. 4 PROGRAMME DAY 2: WEDNESDAY 12 FEBRUARY
................................................................ 6 PROGRAMME DAY 3: THURSDAY 13 FEBRUARY
................................................. 8 SYNDICATE SESSION 1: ECOLOGICAL CONSIDERATIONS
............................................... 10 SYNDICATE SESSION 2: STATISTICAL CONSIDERATIONS
............................................. 12 SYNDICATE SESSION 3: REGULATORY CONSIDERATIONS
........................................................................................................... 14 ABSTRACTS AND CVS
........................................................................................................ 52 LIST OF PARTICIPANTS
.............................................................................................................................. 54 LOGISTICS
Page 1
INTRODUCTION
AIM
The workshop will discuss and report current thinking on when and how species sensitivity
distributions, SSDs, should be used and how the methodology might be further developed.
The workshop will consider three key aspects:
1) What is the ecological relevance of an SSD?
• Are we making ecologically relevant assessments? Are regulatory protection goals
explicit and clear? Are they set in relation to environmental quality? How do prospective
and retrospective approaches differ?
• Are all species of equal importance, or are there keystone species that are more
important than others? If so, how might these be accounted for?
• Is a generic PNEC derived from an SSD overly simplistic in terms of ecological
representativeness? Should we develop representative assemblages/communities
(archetypes) to represent different typologies? Should protection goals account for local
community composition?
• How does aquatic community sensitivity vary with species composition? (summary of
and developments since Pellston Classic workshop 2001- Ecological Variability:
Separating Natural from Anthropogenic Causes of Ecosystem impairment)
• How can knowledge of chemical MoA help construct SSDs for HC5 estimation?
2) What SSD statistical models are available for deriving toxic thresholds (HC5/PNEC) for
aquatic communities?
• Review current tools and key (statistical) methodology, including assumptions about
distributions of sensitivity, use of hierarchical models, interspecies correlations. Identify
where there are important differences and what the implications of these could be.
• As sensitivity to chemical stress seems to be related to taxonomic closeness, how could
this be used in the construction and interpretation of SSDs?
• Do models that utilise prior knowledge, e.g. aquatic toxicity data sets on many species,
provide advantages over other methods?
• Are current modelling success criteria, such as those identified in the REACH TGD,
sufficient, overly prescriptive or insufficient?
3) Regulatory application
• Would the methods reviewed in this workshop be accepted for use in regulatory
assessments under current guidance? If not, what steps would be needed to facilitate
their acceptance in the future?
• Should current guidance on the use of SSDs be revised in the light of the issues and
approaches discussed in this workshop?
• What implications are there for the interpretation of SSDs and HC5s in risk assessment
and risk management?
Page 2
PROGRAMME DAY 1: TUESDAY 11 FEBRUARY
08:00 - 09.00 Registration and coffee
09:00 - 09.10 Welcome and introductory remarks Organising Committee
09:10 - 09.40 Sense, simplicity and successes of SSDs in environmental
protection, assessment and management Leo Posthuma
RIVM, The Netherlands
What is the ecological relevance of an SSD?
Chair: Scott Belanger
P&G, USA
09:40 - 10:10 Ecological limitations of SSDs Lorraine Maltby
University of Sheffield, UK
10:10 - 10:40 How do species traits influence sensitivity and herewith
species sensitivity distributions? Paul van den Brink
Alterra, The Netherlands
10:40 - 11:00 Coffee break
11:00 - 11:30 Field validation of species sensitivity distributions Adam Peters
WCA Environment, UK
11:30 - 12:00 Derivation of toxicity thresholds for LAS – integration
of QSARs, SSDs, mesocosms, and field data Scott Belanger
P&G, USA
12:00 – 12:30 Field-based species sensitivity distribution and community
sensitivity distribution as alternative ways for field validation
of the PNECs derived from laboratory based approaches
Kenneth Leung
University of Hong Kong
12:30 - 13:30 Lunch
Page 3
13:30 - 15:00 SYNDICATE SESSION 1: ECOLOGICAL CONSIDERATIONS
Chair: Scott Belanger
P&G, USA
Group: 1A 1B 1C 1D
Moderator: L Maltby L Posthuma S Duquesne K Solomon
Rapporteur: M Hamer P Whitehouse S Dyer S Marshall
• Are we making ecologically relevant assessments? Are regulatory protection goals explicit
and clear? Are they set in relation to environmental quality? How do prospective and
retrospective approaches differ?
• Are all species of equal importance, or are there keystone species that are more important
than others?
• Is a generic PNEC derived from an SSD overly simplistic in terms of ecological
representativeness or should we develop representative assemblages/communities
(archetypes) to represent different typologies? Should protection goals account for local
community composition?
• How does aquatic community sensitivity vary with species composition?
• How can knowledge of chemical MoA help construct SSDs for HC5 estimation?
• What are the research needs?
15:00 - 16:00 Plenary feedback & discussion with panel
Chair: Scott Belanger and Mick Hamer
Breakouts report back (5-10 minutes each)
Identify key points, consensus and research needs
16:00 - 16:30 Coffee break
What SSD statistical models are available for deriving toxic thresholds (HC5/PNEC) for aquatic
communities? Chair: Peter Craig
16:30 - 16:50 HC5 estimation in SSDs revisited Tom Aldenberg
RIVM, The Netherlands
16:50 - 17:10 Assessment factors for deriving PNECs: food for thought Ad Ragas
Radboud University, The Netherlands
17:10 - 17:30 Weight of evidence approaches for deriving HC5s Sandrine Andres
INERIS, France
17:30 – 17:50 Sample size in PNEC derivation Scott Dyer
P&G, USA
17:50 – 18:10 How to extrapolate across 100,000+ substances, sites and
species with SSDs? Jan Hendriks
Radboud University, The Netherlands
Close of first day
19:30 Dinner
Page 4
PROGRAMME DAY 2: WEDNESDAY 12 FEBRUARY
What SSD statistical models are available for deriving toxic thresholds (HC5/PNEC) for aquatic
communities?
Chair: Andy Hart
FERA, UK
09:00 - 09:30 Interspecies correlation estimation (ICE) models predict
supplemental toxicity data for SSDs Sandy Raimondo
US EPA, USA
09:30 - 10:00 HC5s from taxonomically structured hierarchical species
sensitivity distributions Peter Craig
University of Durham, UK
10:00 - 10.30 Coffee break
10:30 - 12:00 Demonstration of the web-based interspecies correlation estimation
(web-ICE) modelling application
Peter Craig/ Mace Baron/Sandy Raimondo
12:00 - 13:00 Lunch
13:00 - 14:00 Case studies Session
Stuart Marshall, Mick Hamer, Scott Belanger and Peter Craig
Two case studies will be described and discussed using a surfactant LAS and a pesticide, chlorpyrifos.
For each chemical, HC5s will be derived with available data using a range of SSD methods/tools.
Different ecological scenarios will be assessed: stream, pond, marine.
14:00 - 15:00 SYNDICATE SESSION 2: STATISTICAL CONSIDERATIONS
Chair: Andy Hart
FERA, UK
Group: 2A 2B 2C 2D
Moderator: K Leung R Wenning A Ragas P Chapman
Rapporteur: P Craig JP Gosling M Barron S Raimondo
• Review current tools and key (statistical) methodology, including assumptions about
distributions of sensitivity, use of hierarchical models, interspecies correlations. Identify where
there are important differences and what the implications of these could be.
• As sensitivity to chemical stress seems to be related to taxonomic closeness, how could this
be used in the construction and interpretation of SSDs?
• Do models based on prior knowledge provide advantages over other methods?
• What are the research needs?
Page 5
15:30 - 16:00 Coffee break
16:00 - 17:00 Plenary: feedback & discussion with panel
Chair: Andy Hart/Peter Craig
Breakouts report back (5-10 minutes each)
Identify key points, consensus and research needs
Regulatory Applications
Chair: Mace Barron
US EPA, USA
17:00 - 17:30 Regulatory application of SSDs in European regulations
Paul Whitehouse
Environment Agency, England
17:30 – 18:00 Regulatory use of SSDs in Australia and New Zealand
Michael Warne
DSITIA Science Delivery, Australia
19:30 Dinner
Page 6
PROGRAMME DAY 3: THURSDAY 13 FEBRUARY
Regulatory Applications
Chair: Paul Whitehouse
Environment Agency, England
08:30 - 09:00 Use of SSD in China Fengchang Wu
Chinese Research Academy of Environmental Sciences
09:00 - 09:30 Use of SSD to derive no-effect thresholds for water quality
guidelines and ecological risk assessment in Canada
Anne Gosselin
Environment Canada
09:30 - 10:00 Use of SSDs in the USA – endangered species and
water quality criteria Mace Barron
US EPA, USA
10:00 – 10:30 Coffee break
10:30 - 11:30 SYNDICATE SESSION 3: REGULATORY CONSIDERATIONS
Chair: Paul Whitehouse
Group: 3A 3B 3C 3D
Moderator: A Peters M Warne A Gosselin D de Zwart
Rapporteur: M Hamer S Belanger M Barron A Hart
• Would the methods reviewed in this workshop be accepted for use in regulatory assessments
under current guidance? If not, what steps would be needed to facilitate their acceptance in
the future? What are the opportunities to update technical guidance?
• Should current guidance on the use of SSDs be revised in the light of the issues and
approaches discussed in this workshop, e.g. number of species?
• What implications are there for the interpretation of SSDs and HC5s in risk assessment and
risk management?
• What are the research needs?
11:30 -12:30 Plenary: feedback & discussion with panel
Chair: Paul Whitehouse/Mace Barron
Breakouts report back (5-10 minutes each)
Identify key points, consensus and research needs
Page 7
15:30 - 16:00 Coffee break
12:30 - 13:30 Final Plenary discussion: synthesis of key points and research needs from
the three sessions
Chair: Mick Hamer/Andy Hart/Paul Whitehouse
Identify key points and consensus
What are the research needs?
Next steps
13:30 – 14:30 Adjourn and lunch
Close of Workshop
Page 8
SYNDICATE SESSION 1: ECOLOGICAL CONSIDERATIONS
Suggested topics:
• Are we making ecologically relevant assessments? Are regulatory protection goals
explicit and clear? Are they set in relation to environmental quality? How do prospective
and retrospective approaches differ?
• Are all species of equal importance, or are there keystone species that are more
important than others?
• Is a generic PNEC derived from an SSD overly simplistic in terms of ecological
representativeness or should we develop representative assemblages/communities
(archetypes) to represent different typologies? Should protection goals account for local
community composition?
• How does aquatic community sensitivity vary with species composition?
• How can knowledge of chemical MoA help construct SSDs for HC5 estimation?
• What are the research needs?
Group 1A Dam Room
First Name
Name
Role
Lorraine
Maltby
Moderator
Mick
Hamer
Rapporteur
Tom
Aldenberg
Timothy
Barber
Peter
Craig
Pepijn
de Vries
Chenglian
Feng
Guillaume
Kon Kam King
Kenneth
Leung
Adam
Peters
Page 9
Group 1B Warmoes Room
First Name
Name
Role
Leo
Posthuma
Moderator
Paul Whitehouse Rapporteur
Scott Belanger
Christian
Collin-Hansen
Charles Eadsforth
Malyka Galay Burgos
John Paul
Gosling
Marion Junghans
Paul Van den Brink
Michael
Warne
Richard Wenning
Group 1C Executive Room
First Name Name Role
Sabine Duquesne Moderator
Scott Dyer Rapporteur
Mace Barron
Jean Lou Dorne
Anne Gosselin
Maike Habekost
Jan Hendriks
Christian Michel
Ad Ragas
Krishna Kumar Selvaraj
Fengchang Wu
Group 1D Amsterdam Room
First Name Name Role
Keith Solomon Moderator
Stuart
Marshall
Rapporteur
Sandrine Andres
Peter Chapman
Dick
de Zwart
Andy Hart
Ailbhe Macken
Yuan Pan
Sandy Raimondo
Hans Sanderson
Zhen-guang
Yan
Page 10
SYNDICATE SESSION 2: STATISTICAL CONSIDERATIONS
Suggested topics:
• Review current tools and key (statistical) methodology, including assumptions about
distributions of sensitivity, use of hierarchical models, interspecies correlations. Identify
where there are important differences and what the implications of these could be.
• As sensitivity to chemical stress seems to be related to taxonomic closeness, how could
this be used in the construction and interpretation of SSDs?
• Do models based on prior knowledge provide advantages over other methods?
• What are the research needs?
Group 2A Dam Room
First Name Name Role
Kenneth Leung Moderator
Peter Craig Rapporteur
Tom Aldenberg
Timothy Barber
Pepijn de Vries
Chenglian Feng
Mick Hamer
Guillaume Kon Kam King
Lorraine Maltby
Adam Peters
Page 11
Group 2B Warmoes Room
First Name
Name
Role
Richard
Wenning
Moderator
John Paul Gosling Rapporteur
Christian Collin-Hansen
Scott
Belanger
Charles Eadsforth
Malyka Galay Burgos
Marion
Junghans
Leo Posthuma
Paul Van den Brink
Michael
Warne
Paul Whitehouse
Group 2C Executive Room
First Name Name Role
Ad Ragas Moderator
Mace Barron Rapporteur
Jean Lou Dorne
Sabine Duquesne
Scott Dyer
Anne Gosselin
Maike Habekost
Jan Hendriks
Christian Michel
Krishna Kumar Selvaraj
Fengchang Wu
Group 2D Amsterdam Room
First Name Name Role
Peter Chapman Moderator
Sandy
Raimondo
Rapporteur
Sandrine Andres
Dick de Zwart
Andy
Hart
Ailbhe Macken
Stuart Marshall
Yuan Pan
Hans Sanderson
Keith Solomon
Zhen-guang
Yan
Page 12
SYNDICATE SESSION 3: REGULATORY CONSIDERATIONS
Suggested topics:
• Would the methods reviewed in this workshop be accepted for use in regulatory
assessments under current guidance? If not, what steps would be needed to facilitate
their acceptance in the future? What are the opportunities to update technical guidance?
• Should current guidance on the use of SSDs be revised in the light of the issues and
approaches discussed in this workshop, e.g. number of species?
• What implications are there for the interpretation of SSDs and HC5s in risk assessment
and risk management?
• What are the research needs?
Group 3A Dam Room
First Name
Name
Role
Adam
Peters
Moderator
Mick
Hamer
Rapporteur
Tom
Aldenberg
Timothy
Barber
Peter
Craig
Pepijn
de Vries
Chenglian
Feng
Guillaume
Kon Kam King
Kenneth
Leung
Lorraine
Maltby
Page 13
Group 3B Warmoes Room
First Name
Name
Role
Michael
Warne
Moderator
Scott Belanger Rapporteur
Christian Collin-Hansen
Charles
Eadsforth
Malyka Galay Burgos
John Paul Gosling
Marion
Junghans
Leo Posthuma
Paul Van den Brink
Richard
Wenning
Paul Whitehouse
Group 3C Executive Room
First Name Name Role
Anne Gosselin Moderator
Mace Barron Rapporteur
Jean Lou Dorne
Sabine Duquesne
Scott Dyer
Maike Habekost
Jan Hendriks
Christian Michel
Ad Ragas
Krishna Kumar Selvaraj
Fengchang Wu
Group 3D Amsterdam Room
First Name Name Role
Dick De Zwart Moderator
Andy
Hart
Rapporteur
Sandrine Andres
Peter Chapman
Ailbhe
Macken
Stuart Marshall
Yuan Pan
Sandy Raimondo
Hans Sanderson
Keith Solomon
Zhen-guang
Yan
Page 14
ABSTRACT
Sense, simplicity and successes of SSDs in environmental protection,
assessment and management
Leo Posthuma
RIVM, the Netherlands
This contribution presents the versatile use of Species Sensitivity Distributions (SSDs) in
environmental protection, assessment and management of environmental stress. There is sense,
simplicity and success – despite shortcomings. Decades ago, the observation was made that
sensitivities of different species towards a toxic compound are distributed such that it would fit a
statistical model. This marked the birth of SSD-modeling. The concept helped to solve problems of
that time. Soils and waters were affected by various compounds in the environment, and SSDs were
used to set Environmental Quality Criteria. SSDs were one of the methods used to derive the so-
called PNEC (Predicted No Effect Concentration) for ecosystems. Comparison of a PEC (Predicted
Environmental Concentration) of a compound with its PNEC became a standard judgment method,
applied to reduce toxic impacts in ecosystems, with PEC/PNEC-ratio’s higher than unity signaling
risk. Many concerns were (and are) voiced, a.o. on quality, number and relevance of input data of
SSDs, model choice, and the PNEC itself. This workshop focuses, with today’s knowledge, on SSD-
issues – to support optimal decisions.
Since the invention of SSDs, holistic goals have been added to environmental policies. Water bodies
should e.g. reach Good Ecological Status, next to Good Chemical Status. Furthermore, monitoring
has revealed that exceedances of Criteria are frequent. These issues have triggered attention for
applying SSDs in a second way: to derive local hazard levels from ambient exposures. In
combination with mixture modeling, this use yields estimated values for the toxic pressure (of single
chemicals or mixtures) of environmental samples. This use is currently frequent, in disciplines and
approaches as variable as eco-epidemiological landscape-scale diagnosis of local impact causes,
determining sanitation urgency in soil management, use in product Life Cycle Analysis, derivation of
Chemical Footprints of current emissions, and impact reduction of chemical disasters around the
globe by UN-field teams.
As yet, the dual use of SSDs has major practical implications in environmental protection and
management, and that this expands beyond chemicals to stressors such as underwater noise,
temperature and radionuclides, and beyond current techniques, such as via field-SSDs. Given the
societal importance, the validity of SSD output is key. What does it mean when environmental
concentrations increase? Does the predicted impact relate to observed impact? And if so, is it one-to-
one, or at least linear? Confirmation study results suggest a nuanced outcome. SSD outputs clearly
have sense, in that they rank impact degree, not kind – in a way helpful for the assessment goals.
SSDs are simple, and they contain not a single bit of ecology, but they generate sensible outcomes
for various problem definitions.
Past SSD applications have resulted in major environmental improvements (successes), which might
best be envisaged by imagining the absence of concepts like the PNEC on the one hand, and the
potentially affected fraction of species on the other. Reflections on the sense, simplicity and successes
of SSDs provide the context within which potential improvements in the model can be designed.
Page 15
SHORT C.V.
Leo Posthuma
Leo Posthuma is currently research staff member of the Centre for Sustainability,
Environment and Health at the Dutch National Institute for Public Health and the
Environment, RIVM.
He is involved in the development, testing and validation of methods for ecological risk
assessment and sustainable development. His research experience has included
phytopathological studies and studies on the evolutionary ecology and population genetics of
contaminant adaptation of exposed soil arthropod populations (PhD at VU Amsterdam). At
RIVM he worked, amongst others, on community tolerance evolution, on bioavailability of
toxic compounds, on joint effects of compound mixtures, on stability and resilience of soil
ecosystems, on disaster management problems and on sustainability questions. He
authored and co-authored many publications, acted as book editor and co-editor, and works
on practicable tools for policy use, derived from the research. The book on SSDs which he
co-edited and which was published in 2002 has collated the two formats of SSD use
regarding theories and practices of that time.
Page 16
ABSTRACT
Ecological limitations of SSDs
Lorraine Maltby
University of Sheffield, UK
Species sensitivity distributions are generally derived using data from single-species toxicity
tests. The species used in these tests are often from a limited geographic and/or habitat
range and toxicity is measured in the absence of interspecific interactions. SSDs are used
to assess the risk of chemicals to ecological assemblages containing many interacting
species, often in a range of habitats (e.g. rivers, ditches, ponds) in different geographic
regions. This presentation will explore the potential ecological limitations of an SSD
approach, with particular focus on pesticide risk assessment.
Page 17
SHORT C.V.
Lorraine Maltby
Lorraine Maltby is Professor of Environmental Biology at the University of Sheffield, UK and
a NERC/Defra High Level Placement Fellow. Her fellowship is focussed on strengthening
the uptake of science into policy and her research aims to understand how ecosystems
respond and adapt to environmental stressors, including pollutants. She has over 100 peer-
reviewed publications and has co-authored 3 books in ecotoxicology and risk assessment.
Lorraine has served on UK government advisory committees and is past Chair of the
Environment Panel of the Advisory Committee on Pesticides. She is currently a member of
the Scientific Committee of the European Centre for Ecotoxicology and Toxicology of
Chemicals and the UNEP Scientific Expert Group on Chemicals and the Environment. She
was a member of the working group that produced the EFSA protection goals opinion.
Page 18
ABSTRACT
How do species traits influence sensitivity and herewith species sensitivity
distributions?
Paul J. van den Brink
Alterra and Wageningen University, the Netherlands
Species sensitivity distributions (SSD) assume that sensitivity to toxicants within target
species is random. While the SSD approach has shown promise, it is limited by the fact that
data are sparse for most compounds, and that these data are largely based on the lethal
responses of a small group of testing lab species. Here I present an alternative approach,
based on the hypothesis that organisms’ sensitivity to stress is a function of their biology,
and can be predicted from species traits such as morphology, life history, physiology and
feeding ecology.
In this talk I will show a few examples on how species traits have been used to explain the
differences in sensitivity between species.
I) Using data from the US EPA’s AQUIRE database, we found that four species traits
explained 71% of the variability in sensitivity to toxicants within a group of 12 species
exposed to 15 chemicals. Our results indicate that this approach has promise, but effort
is needed to compile species trait information to increase the power, precision and
taxonomic representativeness of this approach.
II) Secondly, we mined existing data on organophosphate, carbamate and pyrethroid
toxicity and mode of action and also species trait information. We linked taxon sensitivity
to their traits at the family level in order to generate empirical and mechanistic
hypotheses about sensitivity-trait relationships. In this way, we developed a Mode-
Specific Sensitivity (MSS) ranking method, and tested this at the taxonomic level of
family and genus. The MSS rankings were successfully linked to existing trait data in
order to identify traits with predictive potential. Single traits as well as combinations of
traits can be used to predict laboratory sensitivity to the substances tested, although
associations were not as strong as in previous studies.
III) We also explore whether and in what ways traits can be linked purposefully to
mechanistic effect models to predict intrinsic sensitivity using available data on the acute
sensitivity and toxicokinetics of a range of freshwater arthropods exposed to chemicals,
using the insecticide chlorpyrifos as an example. The results of a quantitative linking of
seven different endpoints and twelve traits demonstrate that while quantitative links
between traits and/or trait combinations and process based (toxicokinetic) model
parameters could be established, the use of simple traits to predict classical sensitivity
endpoints yields less insight. Future research in this area should include a quantitative
linking of toxicodynamic parameter estimations and physiological traits, and requires
further consideration of how mechanistic trait-process/parameter links can be used for
prediction of intrinsic sensitivity across species for different substances in ERA.
Page 19
SHORT C.V.
Paul J. van den Brink
Paul J. Van den Brink is a professor of chemical stress ecology and works at the research
institute Alterra and the Aquatic Ecology and Water Quality Management Group of
Wageningen University, both belonging to the Wageningen University and Research centre.
At Wageningen University Paul chairs the chemical stress ecology group which currently
consists of himself, a PostDoc and 9 PhD students. For both affiliations, he is involved in
supervising and executing international projects on the scientific underpinning of higher tier
risk assessment procedures for contaminants. Recent research topics are the development
of effect models (e.g. individual based, meta-population models and eco informatics, expert
base models), Trait based Ecological Risk Assessment (TERA) and ecological risk
assessment of chemicals in the tropics. Since 1995, Paul van den Brink has published over
100 peer reviewed papers, for two of which he won an international prize. He also co-edited
five books. In 2006 Paul won the LRI-SETAC Innovative Science Award. He also organized
and took part in many international workshops and courses. Paul van den Brink is presently
a member of the SENSE research school (www.sense.nl), associate fellow of the Canadian
River Institute and editor of the journal: ‘Environmental Toxicology and Chemistry’. He is also
the past-president of SETAC (Society of Environmental Toxicology and Chemistry;
www.setac.org) World and of SETAC Europe. Paul is also a honorary visiting professor at
the University of York.
Page 20
ABSTRACT
Field validation of species sensitivity distributions
Adam Peters
WCA Environment, UK
There is a requirement for quality standard derived under the WFD to be compared with
evidence from field studies. The same principle can also be applied to any chemical
substance for which a robust ecological threshold (e.g. PNEC) has been derived, for
example through the derivation of a Species Sensitivity Distribution. Several different
approaches towards performing these types of assessments are outlined, including
examples of real assessments. The advantages and limitations of various assessment
approaches are considered for both whole community assessments and assessments that
are targeted at particularly sensitive organisms.
In order to evaluate relationships between metal exposures and benthic community metrics,
the bioavailability of the metals must be calculated for each site. Several approaches can be
taken towards the assessment of PNEC values, including simplistic assessments of
ecological quality at different exposure levels and the derivation of limiting functions
(comparable to a traditional dose response relationship). Assessments can be based on the
whole community, subsets of the community, groups of taxa, or an individual taxon.
Analyses based at the level of the whole community may lack the sensitivity to identify slight
effects on particularly sensitive species or families. Reducing the diversity of organisms
assessed increases the uncertainty in the assessment, particularly for reference based
methods. Approaches towards the identification of those taxa that should be considered as
sensitive to a particular pollutant will be considered.
A novel approach for bridging the gap between quality standards based on laboratory
ecotoxicity studies and site-specific local aquatic communities is also outlined. This
approach aims to take account of variation in the composition of ecological communities,
and the effect that this may have on the sensitivity of the community to a particular pollutant.
This is illustrated with an example for deriving site-specific thresholds for zinc in an area
affected by historic mining activities.
Page 21
SHORT C.V.
Adam Peters
Adam Peters, PhD is a Principal Consultant at wca environment and an environmental
chemist with over 10 years of experience in environmental consultancy, environmental
regulation and academia. Adam has considerable expertise in the bioavailability of trace
metalsHe has previously been responsible for management of environmental aspects of the
Notification of New Substances scheme and the Existing Substances Regulations in the UK,
and has recently been a regular attendee as an expert for metals-related issues at European
Commission TCNES. He has been closely involved with the development of Environmental
Quality Standards for both metals and organic chemicals, and the preparation of the Metals
Environmental Risk Assessment Guidance (MERAG). Adam’s main areas of expertise are in
the assessment of environmental fate, behaviour, bioavailability and effects of trace metals
in relation to the use of Biotic Ligand Models; environmental risk assessment of industrial
chemicals; assessment of Persistent, Bioaccumulative and Toxic (PBT) substances; Hazard
assessment of waste materials and their recovery; and development and validation of
environmental quality standards.
Key Publications
Peters A, Simpson P, Moccia A. 2013. Accounting for both local aquatic community
composition and bioavailability in setting site-specific quality standards for zinc.
Environmental Science and Pollution Research (in press).
Peters A, Simpson P, Merrington G, Schleckat C, Rogevich-Garman E. 2013. Assessment of
the effects of nickel on benthic macroinvertebrates in the field. Environmental Science and
Pollution Research (in press).
Peters A, Lofts S, Merrington G, Brown B, Stubblefield W, Harlow K. 2011. Development of
biotic ligand models for chronic manganese toxicity to fish, invertebrates, and algae.
Environmental Toxicology and Chemistry, 30; 2407–2415.
Peters A, Crane M, Adams W. 2011 Effects of iron on benthic macroinvertebrate
communities in the field. Bulletin of Environmental Contamination and Toxicology 86:591-
595.
Crane M, Fisher B, Leake C, Nathail P, Peters A, Stubblefield W, and Warn T. (2010) How
should an environmental standard be implemented. Chapter in: Derivation and Use of
Environmental Quality and Human Health Standards for Chemical Substances in Water and
Soil. SETAC CRC Press, Boca Raton, FL. 140 pp.
Page 22
ABSTRACT
Derivation of toxicity thresholds for LAS – integration of QSARs, SSDs, mescosms,
and field data
Scott Belanger
Procter & Gamble, USA
Linear alkylbenzene sulfonate (LAS) has been one of the most heavily used anionic
detergent chemicals globally since its introduction to the market in the 1960’s. As such, it
has a rich information base spanning physical-chemical properties, specific analytical
methods applicable to all environmental matrices, acute and chronic toxicity,
bioaccumulation, field monitoring data, and assessments using stream mesocosms. In this
talk, this information will be reviewed in support of an integrated approach that translates
acute and chronic toxicity data on pure LAS materials and technical mixtures to field-relevant
distributions of LAS leaving wastewater treatment plants (which does not bear a direct
relationship to toxicity tests performed in the laboratory). Using the so-called toxicity
normalization method, laboratory toxicity data will be re-presented in light of field
distributions to generate robust Species Sensitivity Distributions (SSDs) that are probed to
understand the singular robustness of the SSD itself. Leave-one-out and add-one-in Monte
Carlo simulations are used to quantitatively and qualitatively evaluate “what-if” scenarios
regarding the generation of additional data. Lastly, SSDs will be compared with robust
stream mesocosm studies on LAS to support their predictive nature.
Page 23
SHORT C.V.
Scott Belanger
Scott Belanger is presently a Research Fellow in Procter & Gamble’s corporate safety and
sustainability organization where he has broad leadership responsibilities for environmental
toxicology, science, and technology guidance from an environmental perspective. He holds
degrees from the University of Wisconsin (B.S.), Bowling Green State University (M.S.) and
Virginia Tech (Ph.D. and post-doctoral appointment). Prior to joining P&G in 1989, he was
an Assistant Professor in Environmental Toxicology at the University of Louisiana-Lafayette.
During his tenure at P&G Scott has directed research at P&G’s Experimental Stream Facility
in southwestern Ohio evaluating the ecological impacts of P&G’s highest volume detergent
chemicals. Later he assumed responsibility for P&G’s global environmental toxicology
function, leads efforts on environmental animal alternatives and has overall responsibility for
management of P&G’s corporate human and environmental safety research portfolio. Scott
is a recognized authority in the responses of aquatic life to man-made and natural stressors
and has authored over 100 published scientific articles, books and book chapters on these
topics. He has served on numerous national and international panels providing advice to
organizations such as the U.S. Environmental Protection Agency, the OECD (Organization
for Economic Co-operation and Development, an international governing body), the
European Commission, the Japanese Ministry of Environment, Trade and Industry, and
Environment Canada.
Presently in P&G’s Corporate Environmental Safety and Sustainability Organization he
directs research on ecological and toxicological responses of fish, invertebrates and algae to
consumer product chemicals and advises P&G broadly on the development of new
technologies and issues relating to environmental matters
Page 24
ABSTRACT
Field-based species sensitivity distribution and community sensitivity distribution
as alternative ways for field validation of the PNECs derived from laboratory based
approaches
Kenneth Mei Yee Leung
The Swire Institute of Marine Science and School of Biological Sciences, The University of
Hong Kong, Hong Kong, China
The determination of predicted no-effect concentrations (PNECs) and sediment quality
guidelines (SQGs) of toxic chemicals in marine sediment is very crucial in ecological risk
assessment, sediment quality management (e.g. mud disposal in the sea) and
environmental remediation (e.g. dredging of contaminated mud). However, current methods
of deriving sediment PNECs are primarily based on toxicity data generated from laboratory
ecotoxicity bioassays that are often lack of ecological realism. To tackle this issue, we have
developed two novel alternative approaches to scientifically derive site-specific SQGs by
utilizing field data of benthic biodiversity and contaminant concentration which are
concurrently measured in sediment samples collected from the area of concern. In this
talk, I will first describe the principle of these field-based approaches. Secondly, I will
introduce the field-based species sensitivity distributions (f-SSDs) approach, which is based
on the relationship between species abundance and contaminant level [Environmental
Science & Technology 39:5148-5156; Environmental Toxicology & Chemistry 27:226-234].
Since its establishment, f-SSDs have been utilised in different parts of the world such as
Europe, Hong Kong, New Zealand and the United States. Norwegian continental shelf and
the marine environment of Hong Kong will be taken as examples to illustrate the
methodology. Thirdly, I will present the community sensitivity distributions (CSDs) approach
which is founded on the relationship between species density and contaminant level, and
makes use of Empirical Bayes methods [Environmental Science & Pollution Research 21:
177-192]. Overall, the field-data-derived SQGs appear to be more environmentally relevant
and ecologically realistic. The f-SSD and CSD can be directly adopted as ‘effect
distributions’ for probabilistic risk assessment. The field-data-derived SQGs can be
employed as site-specific guidelines, and used to validate the current PNECs or SQGs
derived from laboratory ecotoxicity data. Finally, the limitation of these field-based
approaches will be discussed, while their recent development and application in different
countries will be highlighted.
Page 25
SHORT C.V.
Kenneth Leung
Kenneth Leung is Professor of Aquatic Ecology and Toxicology at the Swire Institute of Marine
Science and School of Biological Sciences, in the University of Hong Kong (HKU), Hong Kong,
China. Currently, he is also serving as Associate Dean (Research and Graduate Studies) at the
Faculty of Science in HKU. During 2010–2012, he was the elected President of the Society of
Environmental Toxicology and Chemistry (SETAC) Asia Pacific Geographic Unit.
Kenneth Leung obtained a BSc degree in Applied Environmental Sciences with first class
honours at University of Portsmouth in England in 1993, and accomplished his MPhil study in
Environmental Science at City University of Hong Kong in 1996. As a recipient of the Swire’s
James Henry Scott PhD Scholarship, he undertook a doctorate study at University of Glasgow in
Scotland and obtained his PhD in marine ecotoxicology in 2000. He was subsequently awarded
the Croucher Foundation Fellowship, enabling him to conduct his 18-month postdoctoral study in
ecological risk assessment of antifouling biocides at Royal Holloway, University of London in
England (2000-2001). He firstly joined HKU as Research Assistant Professor in January 2002
and was promoted to Assistant Professor, Associate Professor (Tenured) and Professor in 2003, 2009
and 2013, respectively.
Kenneth Leung is an aquatic ecotoxicologist with sound knowledge in aquatic ecology,
biostatistics and ecological risk assessment (ERA). Since 1999, he has written or co-authored
over 100 peer-reviewed articles which are principally related to the ecology, pollution,
ecotoxicology and ERA in both marine and freshwater ecosystems. His current research
projects include the derivation of water and sediment quality guidelines of chemical contaminants
for protecting aquatic ecosystems, and development of statistical methods and models for
predicting environmental risks of different chemicals and their mixtures. Since 2010, he has
been assisting the Government of the Hong Kong Special Administrative Region to review the
marine water quality objectives for various physical and chemical parameters. He is also a
subject editor for the SETAC journal Integrated Environmental Assessment and Management,
and serves as an editorial board member for six international journals including Environmental
Science and Pollution Research, Marine Pollution Bulletin, Integrative Zoology, Canadian Journal
of Zoology, Ocean Science Journal and Toxicology and Environmental Health Science. He is the
leading guest editor for a special volume entitled “Environmental quality benchmarks for
protecting aquatic ecosystems” which has been published in Environmental Science and
Pollution Research in January 2014 (ESPR 21:1-243).
Over the past, Kenneth Leung was invited by the Food and Agriculture Organisation to develop a
manual for assessing ecological risks of aquaculture practices, and frequently invited by the
United Nations (UNDP/PEMSEA) to give lectures on ERA-related topics in regional training
workshops. He was a recipient of the "Marine Pollution Bulletin Highly Cited Author Award
[2005-2009]" by Elsevier, the "Award for Service Contribution 2010" by Faculty of Science, HKU
and "2012 Outstanding Alumni Award" by Vocational Training Council of Hong Kong. Owing to
his professional achievements and community services, Kenneth Leung was selected as one of
the “Ten Outstanding Young Persons” for Hong Kong by Junior Chamber International in 2010.
Page 26
ABSTRACT
HC5 estimation in SSDs revisited
Tom Aldenberg
RIVM,the Netherlands
Species Sensitivity Distributions (SSDs)–in their basic form defined as univariate continuous
statistical distributions over a logarithmic species sensitivity concentration axis for a
particular chemical substance–can be applied in environmental risk assessment to estimate
a PNEC (Predicted No-Effect Concentration) for that toxicant. This PNEC is in many cases
implemented as a statistical estimate of the log HC5 concentration. This minimalist model,
originally due to Kooijman and Van Straalen, needs extension to address a multitude of
thinkable challenges, e.g. with regard to species selection, ecosystem representativeness/
functioning, data quality, statistical model selection, and predictive evaluation of the SSD
and its quantiles. In this paper, we will first review how we handled the uncertainty of the log
HC5 for the Logistic and Normal distribution, from a Bayesian viewpoint. Second, we
develop the estimation of the so-called predictive distribution–formally the mean of the
Bayesian spaghetti plot SSD–in order to pinpoint a single-curve SSD for a given statistical
family. This leads to an improved log HC5–or other quantile–estimate, to better reflect
uncertainty due to small sample size. Presently, we consider the ubiquitous median estimate
log HC5 as being unrealistically insensitive to small sample size, hence risking lack of
conservativeness. This is compounded by the 5th and 95th confidence limits of log HC5
uncertainty often not being reported. The Bayesian predictive distribution method spawns a
new table of extrapolation constants, addressing both chronic and acute species sensitivity
data, depending on the basic fraction affected. The sensitivity of these new extrapolation
constants is evaluated in the light of the REACH-required samples sizes of 10, preferably 15.
A recurring concern is the effect of log species toxicity data uncertainty. Operationally, this
may derive from having multiple data for the same species, from dose-response curve
confidence limits, from QSAR-estimated toxicity data with associated confidence, and
possibly a host of other sources of uncertainty. Intuitively, one would expect data uncertainty
to further lower old–as well as new–log HC5 estimates, but methods of hierarchical
modelling reveal that the reverse is the case: the more variation has to be attributed to the
individual species points, the less variation remains for the SSD itself. Surprisingly, theory,
as well as numerical experiments, show that the effect of data uncertainty is quite modest,
leading to the recommendation to take the mean of log data point uncertainty, and continue
with the old, or updated, extrapolation methodology, as if data were certain. Averaging
multiple species data was already REACH-recommended. It follows that using such
averages per species, or employing point estimates, i.e. expected values, through model-
estimated species sensitivities, only leads to slightly increased conservative–that is lower–
estimates of PNEC values pursued. Both new insights of predictive SSD and the effect of
data uncertainty would somewhat alleviate the need for assessment factors addressing
these particular issues.
Page 27
SHORT C.V.
Tom Aldenberg
Tom Aldenberg has been working as a bio-mathematician for over 30 years. He graduated
as a fisheries biologist at the University of Amsterdam and studied Theoretical Biology in
Leiden and Applied Mathematics in Amsterdam. Tom has worked at RIVM since 1979 and is
employed as Senior Scientist. In the first part of his career, he specialized in ecosystem
modeling: nutrient and species dynamics in eutrophic lakes, and trophic transfers of
toxicants in aquatic and terrestrial foodwebs. In the second part of his career, attention
shifted to Species Sensitivity Distribution (SSD) modeling, a statistical technique involving
Bayesian statistics, to derive Hazardous Concentrations and Fraction of species Affected
from species toxicity data. He worked on Probabilistic Environmental Risk Assessment and
participated in CEFIC, EFSA, and EUFRAM EU-based projects by contributing techniques to
estimate Expected Risk. Expected Risk is an improvement of the Risk Characterization Ratio
(RCR), which is the basis of REACH-based Guidance to evaluate the risk of chemicals.
Recently, he has been participating in the EU projects: OSIRIS and CADASTER. In OSIRIS,
Tom has built models for analyzing the information content and methods of decision making
in Repeated Dose toxicity studies, and has developed models for categorical data to quantify
the statistical Weight-of-Evidence in Integrated Testing Strategies (ITSs) of Mutagenicity,
Carcinogenicity, and Skin Sensitization. Several co-authored publications on this have been
published in Regulatory Toxicology and Pharmacology. In CADASTER, he has studied
methods to build SSDs from QSAR-generated species data. This introduces data uncertainty
into the SSD, which basically requires hierarchical Bayesian modeling. Recent interests are
probabilistic decision trees, information-based model selection and confirmation measures in
automated reasoning.
Two SSD-related publications, cited 500 and 250 times, respectively:
T Aldenberg, W Slob (1993). Confidence limits for hazardous concentrations based on
logistically distributed NOEC toxicity data. Ecotoxicology and Environmental Safety, 25 (1),
48-63.
T Aldenberg, JS Jaworska (2000). Uncertainty of the hazardous concentration and fraction
affected for normal species sensitivity distributions. Ecotoxicology and Environmental Safety,
46 (1), 1-18.)
Page 28
ABSTRACT
Assessment factors for deriving PNECs: Food for thought
Ad M.J. Ragas
Radboud University,the Netherlands
Within regulatory contexts such as REACH and the European Water Framework Directive
(WFD), assessment factors are used to derive safe exposure levels for aquatic and
terrestrial ecosystems from single species toxicity data. These safe exposure levels are also
referred to as Predicted No Effect Concentrations or PNECs. If toxicity data are available for
a limited set of aquatic species – e.g. an alga, a daphnia and a fish – the lowest value is
typically divided by an assessment factor to arrive at the PNEC. The value of this
assessment factor varies between 10 and 1000, depending on the number and the nature of
the available data. If chronic NOECs are available for an extensive set of aquatic species
(i.e. > 15 species covering at least 10 different taxonomic groups), the 5th percentile of the
species sensitivity distribution (SSD) is determined and an assessment factor of 1-5 is
subsequently applied to arrive at the PNEC. The main aim of the current contribution is to
formulate recommendations for improving the use of assessment factors in deriving PNECs.
These recommendations are based on a statistical analysis of a large set of chronic toxicity
data resulting from aquatic single species tests and mesocosm experiments.
A database with chronic single species NOECs on 20 different chemicals was compiled
based on data reported in the open literature. Chronic mesocosm data were found for 6 of
these substances and were also included in the database. For each of the substances in the
database, the 5th percentile of the SSD (HC5) was determined. This HC5 was then compared
with:
- the PNEC reported in the mesocosm experiments (if available);
- PNECs derived by applying a safety factor of 10 to the lowest value of a limited dataset
of 3, 6 or 9 NOECs. These datasets were generated by parametric bootstrapping of the
available single-species NOECs.
Mesocosm PNECs were generally lower than the HC5, with two notable exceptions, i.e.,
lindane and dimethoate, which can be explained by the limited set of species in the
mesocosm. The HC5 is on average a factor of 2.0 lower that the PNEC derived from a set of
3 chronic NOECs. This difference increases to a factor of 4.5 and 7.2 for datasets with 6 and
9 chronic NOECs, respectively. Based on these results two general recommendations are
formulated:
- The assessment factor of 10 that is currently being applied to the lowest value of small
datasets (i.e. alga, daphnid and fish) should be differentiated depending on the number
of available data, e.g. a factor of 20 if one value is available for each taxonomic group,
but a value of 5 when three or more values are available for each taxonomic group.
- The default assessment factor of 2 is suggested for the HC5 of the SSD. This default
value can be further refined based on the specific characteristics of the available toxicity
data, i.e. representativeness, mode of action, interspecies variability and uncertainty.
Page 29
SHORT C.V.
Ad M.J. Ragas
Ad Ragas studied biology and obtained a PhD degree at the Radboud University in
Nijmegen. He currently holds a position as an assistant professor at the Department of
Environmental Science at the Radboud University in Nijmegen, and as a full professor in
Environmental Sciences at the School of Science at the Open Universiteit in Heerlen, The
Netherlands. His main domain of expertise is the modelling of human and environmental
risks of chemical substances, covering areas such as fate and exposure modelling, effect
modelling, chemical mixtures, integration of human and ecological risk assessment and risk
perception. Within this domain, his focus is on uncertainty assessment and dealing with
uncertainties in decision-making. He coordinates a work package on risk assessment of
pharmaceuticals in the EU FP7 PHARMAS project. He is deeply involved in academic
educational programmes for environmental scientists, i.e. as a coordinator of an MSc
programme in Environmental Sciences and of several specific courses, e.g. in risk
assessment, GIS and statistics. He chairs the Dutch scientific advisory committee on quality
standards for air and water.
Selected recent publications:
Løkke H, Ragas AMJ, Holmstrup M. 2013. Tools and perspectives for assessing chemical
mixtures and multiple stressors. Toxicology (in press).
Oldenkamp R, Huijbregts MAJ, Hollander A, Versporten A, Goossens H, Ragas AMJ. 2013.
Spatially explicit prioritization of human antibiotics and antineoplastics in Europe.
Environment International 51:13–26.
Ragas AMJ, Oldenkamp R, Preeker NL, Wernicke J, Schlink U. 2011. Cumulative risk
assessment of chemical exposures in urban environments. Environment International
37:872–881.
Ragas AMJ, Brouwer FPE, Büchner FL, Hendriks HWM, Huijbregts MAJ. 2009. Separation
of uncertainty and interindividual variability in human exposure modeling. Journal of
Exposure Analysis and Environmental Epidemiology 19:201-212.
Page 30
ABSTRACT
Weight of evidence approaches for deriving HC5s
Sandrine Andres
INERIS, France
Experience gained in developing Quality Standards (e.g. PNEC) in the framework of the EU
Technical Guidance Document shows that only a few substances can benefit from the use of
an Species Sensitivity Distributions (SSD), even if substances appears after an initial
assessment as data rich substances. The main drawback is the lack of validated studies for
the additional taxa. Indeed, the level of standardisation for testing these additional taxa (such
as mayfly, dragonfly, amphibian, rotifer, molluscs, etc...) is usually lower than for the regular
algae/daphnid/fish simplified trophic chain. As a consequence, this additional information is
often not used for the assessment.
In order to make the best use of all the information available, a Multi-Criteria Decision Analysis
(MCDA) tool was developped in the framework of the research project AMORE (Multi-Criteria
Analysis for the Development of a Decision Support Tool for the prevention of Environmental
Risks). This tool is based on Weight of Evidence (WoE) methodology, which aims to improve
the evaluation of ecotoxicological data, through the assessment of their relevance and
reliability for the definition of SSDs. The methodology allows to rank the acceptability of
ecotoxicological data for further use in the risk assessment process and therefore optimise
their influence in the production of reliable SSDs, through a weighted bootstrap modelling
procedure for data resampling.
In this project, it was hypothetised tha the SSD can be based on all avaialble ecotoxicity data,
which can be heterogeneous and often non comparable. These data can be obtained through
different approaches (e.g. experimental or even modelling) and conditions, e.g., the protocol
can be standardized or not; time duration can vary among experiments, leading to chronic or
acute data; different physiological endpoints can be observed, e.g. mortality, growth,
reproduction; statistics used for interpreting data can differ, e.g. leading to NOEC or ECx.
The methodology is based on the assessment of a hierarchically structured set of 57 criteria,
which is used for assigning a quantitative score to every ecotoxicological datum and was
created based on the review of the state of the art frameworks for the assessment of
ecotoxicological data. The different endpoints are analysed based on their production method
and specifically on three main aspects: the ‘Experimental Reliability’, the ‘Statistical Reliability’
and the ‘Biological Relevance’ of the experimental or modelling protocol used. This
assessment has been developed in with the contribution of an expert panel of scientists on
ecotoxicology. Knowledge and preferences of experts have been gathered through a
participatory process, and is used for the calculation of the aggregated reliability scores of
data. The nature of the process mandates the use of Fuzzy Logic during the aggregation
phase, for handling the inherent uncertainty which appears in the form of unreported
information, as well as possible lack of knowledge of the experts. This approach allows for a
weighed use of the available information available in a weight of evidence perspective.
Page 31
SHORT C.V.
Sandrine Andres
Sandrine ANDRES, PhD, is the head of Unit “Evaluation and Expertise in Toxicology” at
INERIS (National Institute for the Industrial Risks and the Environment). She holds a PhD in
“Environmental toxicology” from the University of Toulouse III. She also pursued a
postdoctoral research mainly focusing on Mercury absorption through biological membranes.
(University of Maryland, USA). Prior to join INERIS, Sandrine worked as Consultant as risk
assessment expert, where one of her main task was in support to the EU Commission for
the implementation of the Rotterdam Convention.
At INERIS, she worked from 2006-2009 as coordinator of the technical evaluation of the
dossiers in the framework of the Biocidal Product Directive. Currently, the activities of the
Unit mainly take place within the regulatory frameworks of REACH, CLP, and the Water
Framework Directive. This Unit is in charge inter alia of the development of Environmental
Quality Standard (EQS) at National level and works on scientific issues related to the
implementations of those EQS. She participates to several national and international experts
groups including, Member of the POP-Review Committee (Convention de Stockholm)
(UNEP) since 2007, Member of the Task Force Biocides (OECD) since 2006, Member of the
Task Force « Environmental Exposure Assessment » (OECD.) since 2007. Current interests
include the improvement of methodologies in support to environmental risk assessment of
chemicals, both from an Hazard and Exposure assessment perspective.
Page 32
ABSTRACT
Sample size in PNEC derivation
Scott Dyer
Procter & Gamble, USA
SSDs have been used to develop water quality criteria (e.g., PNECs) and other protective
environmental concentrations (e.g., HC5). These criteria typically require large datasets
(e.g., USEPA ambient water quality criteria utilize at least 8 acute toxicity values from
several taxa spanning three trophic levels, fish, invertebrates and plants) of measured
toxicity values. However, there has been a considerable debate regarding the minimum
requirements for establishing protective concentrations, such as the HC5, within the
scientific and regulatory communities. For organizations needing to establish these criteria,
questions remain whether the addition of taxa into the SSD will greatly change the criterion.
Is it possible that the addition of taxa will not change the HC5 and thereafter the PNEC? Is
there a law of diminishing returns for expanding the number of taxa incorporated into an
SSD? If so, then understanding factors that dictate the lack of need for additional taxa would
result in appropriate PNECs without undo cost and time. To explore this question we
developed distributions with steep to shallow scale (slope) factors as well as small to large
toxicity ranges. Within these diverse distributions we assessed the effect of the numbers of
taxa values included to generate HC5 values as well as their position within the distribution.
Comparisons of these distributions to real chemical datasets will be discussed.
Page 33
SHORT C.V.
Scott Dyer
Scott D. Dyer obtained his B.S. and M.S. degrees from Iowa State University in Biology and
Toxicology, respectively. His Ph.D. was awarded from the University of North Texas where
he studied the stress protein response in fish exposed to diverse contaminants via laboratory
and field exposures. Since 1991, Dyer has been employed by The Procter & Gamble
Company (P&G), Cincinnati, Ohio, and is presently a Principal Scientist in the Environmental
Stewardship Organization as an eco-toxicologist. His primary mission within P&G is
researching methods that advance the predictions of potential environmental exposure and
effects of chemicals found in consumer products. He currently has three major research
programs, all collaborations with academia, government and industry: 1) the eco-
epidemiology of consumer product chemicals relative to other chemical and physical
stressors; 2) the extrapolation of potential effects across species; and 3) the development of
screening tools for the estimation of metabolism in fish, an important attribute for the
prediction of bioaccumulation. Dyer has authored more than 70 journal articles, book
chapters, and technical reports and currently participates with work groups within
organizations such as the European Centre for Ecotoxicology and Toxicology of Chemicals
(ECETOC), Environment & Health Risk Assessment and Management (ERASM), American
Cleaning Institute (ACI), Water Environment Research Foundation (WERF), ILSI Health and
Environmental Sciences Institute (HESI) and the Society of Environmental Toxicology and
Chemistry (SETAC). He currently serves on the Board of Directors for WERF and
Confluence, a water technology innovation cluster in Cincinnati, Ohio, USA.
Page 34
ABSTRACT
How to extrapolate across 100,000+ substances, sites and species with SSDs?
Jan Hendriks
Radboud University, the Netherlands
Each second, one new chemical is added to the more than 65,000,000 already registered. In
the EU, 100,000+ compounds are awaiting assessment while 1,500,000 contaminated sites
potentially require cleanup. Worldwide, 8,000,000+ species, of which 10,000+ endangered,
need protection (Hendriks 2013).
At the same time, empirical research is severely limited by financial, practical and ethical
constraints. Assessing 100,000+ substances at 100,000+ sites threatening 100,000+
species obviously cannot be achieved by toxicological testing only. As an alternative, I
suggest to focus on simple models. Instead of going for statistical regressions with the
highest explained variability, we might attach more value to meaningful equations of which
the coefficients and exponents can be interpreted physically.
We have derived and collected SSDs on toxic and non-toxic stressors to discern patterns
across stressors, species and endpoints. In this contribution, some examples will be
discussed. We will look at (1) intra-species and inter-species variability, (2) the number of
species included in an SSD, (3) SSDs across modes of action, (4) the combined use of
SSDs for toxic and non-toxic stressors, (5) "field"-based SSDs (PNOFs) and (6) in vitro
biomarker SSDs to in vivo bioassay SSDs (References given below).
Azevedo LB, Van Zelm R, Elshout PMF, Hendriks AJ, Leuven RSEW, Struijs J, De Zwart D, Huijbregts MAJ.
(2013) Species richness – phosphorus relationships for lakes and streams worldwide. Global Ecology and
Biogeography, in press.
De Hoop L, Smit M, Huijbregts MAJ, Leuven RSEW, Schipper AM, Hendriks AJ (2011). Sensitivity of arctic species to
oil and other contaminants in comparison to other species, Environmental Science and Technology 45: 9017–9023.
Elshout PMF, Dionisio Pires LM, Leuven RSEW, Wendelaar Bonga SE, Hendriks AJ (2013). Low oxygen
tolerance of different life stages of temperate freshwater fish species. Journal of Fish Biology 83: 190-206.
Fedorenkova A, Lenders HJR, Ouborg J, Breure AM, Hendriks AJ (2010). Ecotoxicogenomic: bridging the gap
from genes to population, Environmental Science and Technology 44: 4328-4333
Fedorenkova A, Vonk JA, Lenders HJR, Creemers R, Breure AM, Hendriks AJ (2012). Ranking ecological risks
of multiple chemical stressors on amphibians, Environmental Toxicology and Chemistry 31: 1-6.
Fedorenkova A, Vonk JA, Breure AM, Hendriks AJ, Leuven RSEW (2013). Tolerance of native and non-native
fish species to chemical stress: a case study for the river Rhine. Aquatic Invasions 8: 231–241.
Golsteijn L, Van Zelm R, Hendriks AJ, Huijbregts MAJ (2013). Statistical uncertainty in hazardous terrestrial
concentrations estimated with aquatic ecotoxicity data. Chemosphere 93: 366–372.
Golsteijn L, Van Zelm R, Veltman K, Musters G, Hendriks AJ, Huijbregts MAJ (2012). Including ecotoxic impacts on warm-
blooded predators in Life Cycle Impact Assessment. Integrated Environmental Assessment and Management 8: 372–378.
Hendriks AJ (2013). How to deal with 100,000+ substances, sites, and species: Overarching principles in
environmental risk assessment. Environmental Science and Technology 47: 3546−3547.
Hendriks AJ, Awkerman JA, De Zwart D, Huijbregts MAJ (2013). Sensitivity of species to chemicals: differences
between test types (LC50, LD50), cold-blooded and warm-blooded species and modes of action. Ecotoxicology
and Environmental Safety 97: 10-16.
Smit MGD, Bechman RK, Hendriks AJ, Skadsheim A, Larssen BK, Baussant T, Bamber S, Sanni S (2009).
Relating biomarkers to whole-organism effects using species sensitivity distributions: a pilot study for marine
species exposed to oil. Environmental Toxicology and Chemistry 28: 1104–1109.
Page 35
SHORT C.V.
Jan Hendriks
Interests
• 1000+ physical-chemical pressures → 1000+ biological impacts
• chemistry - toxicology - (macro-)ecology, in particular chemokinetics – ecotoxicodynamics
• cycling of (xeno-)biotic constituents using overarching principles at cell - landscape level
Employment
• 2004-… Professor/Head, Dep. Environmental Science, Radboud University Nijmegen, NL
• 1999-04 Section Head, Dep. Chemistry and Ecotoxicology, RIZA/Deltares, Lelystad, NL
• 1990-99 Scientist-Advisor, Section Ecotoxicology, RIZA/Deltares, Lelystad, NL
• 1988-89 Scientist-Advisor, Dep. Ecology and Management, TNO, Delft, NL.
Supervision and publications
• (Co-)supervisor of 30+ P(h)Ds and 30+ MScs
• (Co-)author of 130 international peer reviewed articles and 60 (chapters in) reports &
books
Info and publications
• http://www.ru.nl/environmentalscience/,
http://www.ru.nl/environmentalscience/publications/
• http://www.ru.nl/environmentalscience/staff/individual-staff/hendriks/
Key publications (Species Sensitivity Distributions)
The publications below are in addition to the presentation references on the previous page.
Azevedo LB, van Zelm R, Hendriks AJ, Bobbink R, Huijbregts MAJ (2013). Global
assessment of the effects of soil acidification on plant species richness, Environmental
Pollution 174: 10-15.
Azevedo LB, De Schryver A, Hendriks AJ, Huijbregts MAJ (2014). Calcifying species
sensitivity distributions for ocean acidification. Marine Pollution Bulletin submitted.
Leuven RSEW, Hendriks AJ, Huijbregts MAJ, Lenders HJR, Matthews J, Van der Velde G
(2011). Differences in sensitivity of native and exotic fish species to changes in river
temperature. Current Zoology 57: 852−862.
Page 36
ABSTRACT
Interspecies Correlation Estimation (ICE) Models predict supplemental toxicity data
for SSDs
Sandy Raimondo
Environmental Protection Agency, USA
Species sensitivity distributions (SSD) require a large number of toxicity values for a diversity
of taxa to define a hazard level protective of multiple species. For most chemicals, measured
toxicity data are limited to a few standard test species that are unlikely to adequately
represent ecological communities. Interspecies correlation estimation (ICE) models are log-
linear least squares regressions that predict the acute toxicity to untested taxa from known
toxicity of a single surrogate species. A suite of ICE models is developed from a
comprehensive, standardized dataset of acute toxicity with the goal of maximizing the
number of potential species for which toxicity can be predicted while minimizing extraneous
sources of variation in the models. The United States Environmental Protection Agency
houses three ICE databases: aquatic animals (vertebrates and invertebrates; 5501 records,
180 species; 1266 chemicals), algae (1647 records, 69 species, 457 chemicals), and wildlife
(birds and mammals; 4329 records, 156 species, 951 chemicals). Approximately 2400
models have been developed from these databases and made available through the Web-
based Interspecies Correlation Estimation internet application (Web-ICE;
http://epa.gov/ceampubl/fchain/webice/). ICE models were validated using leave-one-out
cross validation and sources of model uncertainty evaluated. Toxicity predictions are most
accurate for models with closely related taxa pairs, with over 90% of cross-validated values
predicted within 5-fold of the measured value when the surrogate and predicted taxa are in
the same family. Model mean square error and prediction confidence intervals should be
considered when evaluating an ICE predicted value. Models built with a single mode of
action (MOA) were often more robust than models built using toxicity values with multiple
MOAs, and improve predictions among species pairs with large taxonomic distance (e.g.,
within phylum). SSDs developed solely from ICE-predicted toxicity values produce hazard
levels with an average factor of 3.0 and 5.0 of those developed with all measured data for
aquatic species and wildlife, respectively. For chemicals in which more measured data are
available, ICE models may be used to augment datasets to increase species diversity in
SSDs. Compared to SSDs developed from only measured data, the uncertainty of ICE
model predictions contributes less variability to hazard levels than variance due to species
composition. Through extensive study of ICE model evaluation and uncertainty and their
application in developing SSDs, ICE generated toxicity values have been demonstrated to
provide a statistically sound approach to supplementing datasets to generate SSD-based
hazard levels applicable to ecological risk assessments.
Page 37
SHORT C.V.
Sandy Raimondo
Dr. Raimondo currently serves as a Supervisory Research Ecologist of the Biological Effects
and Population Response Branch within the United States Environmental Protection Agency
(US EPA) Gulf Ecology Division. She has been with the US EPA for over 10 years with
focused research areas that include: the development of models that predict ecological
effects of environmental stressors on organisms, populations, and ecosystems; survival and
recruitment modeling; aquatic toxicity tests; dose-response relationships; toxicity
extrapolation modeling; toxicity estimation; uncertainty analyses of complex ecological
processes; aquatic & coastal ecosystems; ecological risk assessment. Sandy led the
development of the Interspecies Correlation Estimation (ICE) modeling tool, Web-ICE
(http://www.epa.gov/ceampubl/fchain/webice/index.htm), which predicts acute sensitivity of
toxicants to aquatic species and wildlife. Web-ICE is used by EPA Regions and Program
Offices to establish water quality and pesticide registration policy. A recent update to Web-
ICE (April 2013) includes new ICE models predicting toxicity of algal taxa and improved
functionality of Species Sensitivity Distribution (SSD) and Endangered Species Modules.
Sandy also leads the development of spatially-explicit models that evaluate the effects of
environmental stressors on populations and ecosystems with focus on uncertainty and
complexity of ecological processes. Sandy has published over 50 peer-reviewed
publications, book chapters, and technical reports, including 11 publications in high impact
journals on ICE modeling and their application in supplementing data to develop SSDs. She
has delivered over 10 seminars and short courses on Web-ICE at international conferences,
as well as regular technical training to US EPA Program Offices and Regions. She has
received over 25 awards within EPA, including an Agency bronze medal and Office of
Research and Development Impact Award for the development of Web-ICE.
Page 38
ABSTRACT
Demonstration of the Web-Based Interspecies Correlation Estimation (Web-ICE)
Modelling Application
Sandy Raimondo
Environmental Protection Agency, USA
The Web-based Interspecies Correlation Estimation (Web-ICE) modeling application is
available to the risk assessment community through a user-friendly internet platform
(http://epa.gov/ceampubl/fchain/webice/). ICE models are log-linear least square regressions
that predict acute toxicity (LC50/LD50) of a chemical to a species, genus, or family based on
estimates of relative sensitivity between the taxon of interest and that of a surrogate species.
Web-ICE v 3.2 includes over 1440 models for aquatic animal taxa, 100 models for algae,
and 852 models for wildlife taxa. Web-ICE has modules that predict toxicity to one taxa of
interest at a time while providing detailed information on model parameters. It also has
Species Sensitivity Distribution (SSD) and endangered species modules that produce
toxicity values to multiple species based on the number of surrogates entered. In the SSD
module, a user can enter up to 20 surrogate species which are used to predict toxicity to all
predicted taxa possible. The entered surrogate and predicted toxicity values are used to
develop a log-logistic probability distribution and estimate a hazard level equivalent to either
the 1st, 5th or 10th percentile of the distribution. Users can also enter multiple surrogate
toxicity values into the endangered species module, which are used to calculate predicted
species, genus, and family level sensitivity for selected endangered species. Both the SSD
and endangered species modules provide exportable data files of predicted results. A
demonstration of the Web-ICE will familiarize participants with the functionality of the
application and provide examples of its use for single taxon predictions, SSD generation,
and development of endangered species toxicity reports.
Page 39
SHORT C.V.
Sandy Raimondo
Dr. Raimondo currently serves as a Supervisory Research Ecologist of the Biological Effects
and Population Response Branch within the United States Environmental Protection Agency
(US EPA) Gulf Ecology Division. She has been with the US EPA for over 10 years with
focused research areas that include: the development of models that predict ecological
effects of environmental stressors on organisms, populations, and ecosystems; survival and
recruitment modeling; aquatic toxicity tests; dose-response relationships; toxicity
extrapolation modeling; toxicity estimation; uncertainty analyses of complex ecological
processes; aquatic & coastal ecosystems; ecological risk assessment. Sandy led the
development of the Interspecies Correlation Estimation (ICE) modeling tool, Web-ICE
(http://www.epa.gov/ceampubl/fchain/webice/index.htm), which predicts acute sensitivity of
toxicants to aquatic species and wildlife. Web-ICE is used by EPA Regions and Program
Offices to establish water quality and pesticide registration policy. A recent update to Web-
ICE (April 2013) includes new ICE models predicting toxicity of algal taxa and improved
functionality of Species Sensitivity Distribution (SSD) and Endangered Species Modules.
Sandy also leads the development of spatially-explicit models that evaluate the effects of
environmental stressors on populations and ecosystems with focus on uncertainty and
complexity of ecological processes. Sandy has published over 50 peer-reviewed
publications, book chapters, and technical reports, including 11 publications in high impact
journals on ICE modeling and their application in supplementing data to develop SSDs. She
has delivered over 10 seminars and short courses on Web-ICE at international conferences,
as well as regular technical training to US EPA Program Offices and Regions. She has
received over 25 awards within EPA, including an Agency bronze medal and Office of
Research and Development Impact Award for the development of Web-ICE.
Page 40
ABSTRACT
HC5s from taxonomically structured hierarchical species sensitivity distributions
Peter Craig
Durham University, UK
One approach to deriving the predicted no-effect concentration for a chemical is to use a
species sensitivity distribution (SSD) model to estimate the hazardous concentration
affecting p% of species (HCp), where p is usually 5. Many questions have been raised about
both principles and application of SSDs but the concept has nevertheless been found to be
useful.
Analysis of a database of aquatic ecotoxicity test results reveals a number of features which
should be addressed by SSD methodology. These include: inter-species correlation;
tendencies of particular species to one or other end of the sensitivity distribution; and inter-
test variation. In earlier work, attempts have been made at addressing each of these issues
on its own. Addressing them collectively requires multivariate statistical modelling.
We present a Bayesian hierarchical model of variability and uncertainty for sensitivities of
species to a chemical undergoing assessment and for a database of relevant test results for
other chemicals. The Bayesian approach has several advantages over traditional non-
Bayesian statistical methodology aimed primarily at analysing experimental data. It can
incorporate both data and other information such as expert judgements or results of meta-
analyses. It provides a collective description of uncertainty for all components of a model, a
coherent mechanism for revising uncertainty when additional data become available, and a
decision-making framework which addresses both uncertainty and utility.
Our model generalises the single randomly-sampled-chemical model proposed by Aldenberg
and Jaworska (2000) and addresses the issues raised above. It models inter-species
correlation by building species tendencies and sensitivities hierarchically, based on the
taxonomic classification of species. Taxonomically-related structure seems natural and
makes the model a better description of the available data but means that it is necessary
also to specify a taxonomic scenario: the taxonomic structure of the community being
protected by the HCp. The HCp is then scenario-specific, being the pth percentile of
sensitivity to the chemical for species in the scenario. The Bayesian nature of the model
means that the HCp estimate is automatically accompanied by a quantitative assessment of
uncertainty.
The model is available in software form for application to test data for a new chemical. The
software, known as hSSD, will be demonstrated at the workshop and is one of the
methodologies used in the workshop case studies.
Page 41
SHORT C.V.
Peter Craig
My undergraduate degree was in Mathematics from Trinity College Dublin and was followed
by a PhD in statistics and a two-year postdoctoral position studying the use of satellite
imagery in the search for mineral deposits, both in TCD with a one year visit to the University
of Washington. Since 1989, I have been a lecturer and then senior lecturer in statistics at
Durham University, active in teaching, research, consultancy and IT management and tools.
I have a wide ranging expertise in statistical methodology and applications, especially in
Bayesian modelling and related computation. In the 1990s, I was part of the group at
Durham leading the way in statistical methodology for analysis of uncertainty for computer
models of natural phenomena, especially oil reservoir models. Since 2004, initially as
external expert contributing to European Food Safety Authority scientific opinions, I have
been active in various areas of risk assessment including dietary exposure to pesticides,
ecotoxicology and margins of exposure for carcinogens. My principal interests in risk
assessment are improved modelling, in particular of chains of evidence, better quantification
of uncertainty for risk assessment procedures and methodology for partial quantification of
uncertainty based on limited specifications by experts. I have published in many of the best
journals in the area of statistical methodology and more recently in journals such as Risk
Analysis, Environmental Toxicology and Chemistry and Food and Chemical Toxicology.
I believe passionately in the importance of the mathematics underlying statistical reasoning
but also that statisticians must devote a lot of time to exploring and analysing data and must
engage at length and in depth with scientists in order to make a meaningful applied
contribution.
Page 42
ABSTRACT
Regulatory applications of SSDs in European regulations
Paul Whitehouse
Environmental Agency, UK
Regulatory frameworks like the Water Framework Directive, Marine Strategy Framework
Directive and REACH are far-reaching pieces of legislation that require us to identify and
manage pressures on the environment, including toxic chemicals. Assessing the hazard
posed by chemicals is central to chemical risk assessment and also to the derivation of
Environmental Quality Standards (EQSs) which play a key role in identifying risks and
helping manage emissions to ensure wildlife and human health are not adversely impacted.
This presentation focuses on the role of EQSs in the Water Framework Directive. There
have been important technical developments in the derivation and application of EQSs in
recent years, some of which have been captured in EU Technical Guidance that was
published in 2012. Whilst deterministic methods for deriving EQSs remain the only option in
some cases, species sensitivity distributions (SSDs) are now widely used to derive EQSs,
including standards for bioavailable metals and, in some cases, for biota standards. I will
briefly review the experience of Member States in using such approaches, how predictions
based on extrapolation from laboratory data relate to field data, and highlight where further
development in EQS derivation would be welcomed. For example, can we use the
relationship between chemical pressure and impact provided by an SDD to manage water
quality more effectively, and is there a role for ecosystem services thinking when we derive
standards for chemicals?
It is important to recognize that deriving the numerical value for a standard (the
concentration) is only part of the story. Often, we focus on the derivation of the EQS, but the
way a standard is implemented by regulatory agencies, such as the design of monitoring
regimes, is also very important in determining the level of protection that is actually
achieved. Although not specific to standards derived using an SSD, problems of
implementation are often overlooked by academics and researchers since it has not
prevented the setting of the standard. At the same time, regulators may be forced to make
serious compromises when implementing an EQS that could be avoided through better
engagement with the scientific community.
Page 43
SHORT C.V.
Paul Whitehouse
A botanist by training, my career has focused on researching and regulating the effects of
chemicals on biological systems. My early career was with Shell Research Ltd in the UK
where I undertook research into the discovery of new herbicides and pesticides. My
involvement in environmental protection began in 1990 when I joined WRc, moving to my
present position in the Environment Agency in 2004. I have worked in chemical hazard and
risk assessment for 20 years, specializing in the derivation and implementation of
Environmental Quality Standards (EQSs) for toxic chemicals and radionuclides in water and
soil. These thresholds are intended to protect wildlife and human health from the toxic
effects of chemicals.
My current role is to provide the technical support to meet the requirements of environmental
legislation and other initiatives aimed at protecting or improving the environment. I lead the
technical development of EQSs in the UK and provide technical support to the UK
government in negotiations on the introduction of EQS for Priority Substances under the
Water Framework Directive. I am also involved in numerous European working groups on
the development and implementation of environmental standards. Between 2008 and 2011, I
led a group responsible for developing European technical guidance on the derivation of
EQSs for chemicals, which is now regarded as definitive guidance on the subject for the
Water Framework Directive.
Current interests include the development and implementation of bioavailability-based
standards for trace metals, and the implementation of biota standards for assessing risks
from bioaccumulative substances in the environment.
My personal interests include gardening, cookery, and I am recent convert to fly fishing,
having outwitted my first rainbow trout this Autumn.
Page 44
ABSTRACT
Regulatory use of SSDs in Australia and New Zealand
Michael Warne
DSITIA Science Delivery, Australia
Authors: Warne MStJ1, Batley GE2, Braga O3, Chapman JC4, Fox D5, Hickey C6, Stauber JL2, and Van Dam R7.
1.Water Quality and Investigations, Environmental Monitoring and Assessment Science, Science Delivery, Department of
Science, Information Technology, Innovation and the Arts, Brisbane, Queensland, Australia.2. Centre for Environmental
Contaminants Research, CSIRO Land and Water, Lucas Heights, NSW, Australia. 3.Department of Sustainability, Environment,
Water, Population and Communities, Canberra, Australia. 4.Office of Environment & Heritage, Lidcombe, NSW, Australia.
5.Environmetrics, Melbourne, Victoria, Australia. 6. National Institute of Water and Atmospheric Research (NIWA), Hamilton,
New Zealand. 7.Environmental Research Institute of the Supervising Scientist, Darwin, NT, Australia.
Australia and New Zealand, along with many other countries, use risk-based approaches to manage
and regulate chemicals in the environment. A key component of the risk approach has been the use
of species sensitivity distribution (SSD) methods. SSDs are central to the Australian and New
Zealand approach to managing the quality of various environmental compartments (water, sediment
and soil), of additives to soils (biosolids and mineral fertilisers) an in conducting environmental risk
assessments. Australia and New Zealand developed a new SSD method called BurrliOZ
(http://www.csiro.au/Outcomes/Environment/Australian-Landscapes/BurrliOZ.aspx) that uses selects
the distribution from the Burr Type III family of statistical distributions that best fits the sensitivity data.
This method can therefore provide a good fit to many more datasets than can SSD methods that use
a single statistical distribution. The Australian and New Zealand Guidelines for Fresh and Marine
Water Quality (http://www.environment.gov.au/resource/australian-and-new-zealand-guidelines-
fresh-and-marine-water-quality-volume-1-guidelines) were released in 2000 and are currently
undergoing a review. This is examining the framework used to derive the guidelines (called trigger
values). Key recommendations arising from the review are: increasing the types and sources of data
that can be used; working with industry to permit the use of commercial-in-confidence toxicity data;
increasing data requirements; improving the software used to calculate trigger values; increasing the
rigour of site-specific trigger values; improving the method for assessing the reliability of the trigger
values; providing guidance of measures of toxicity and toxicological endpoints that may, in the near
future, be appropriate for trigger value derivation. A new set of sediment quality guidelines and new
trigger values for a number of existing metals will be derived. In addition, trigger values for a range of
organic chemicals focussing on pesticides, pharmaceuticals and personal care products will be
derived. Finally, a weight of evidence approach is being included into the guidelines. These changes
will improve the number and quality of the trigger values that can be derived and will increase end-
users’ ability to understand and implement the Guidelines in a scientifically rigorous manner.
The water quality guidelines are generic - a single value that applies to all waterways. The only
exception being the trigger values of some metals that can be modified using hardness algorithms. In
contrast, the Australian guidelines for contaminants in contaminated soils and in biosolids, are
wherever possible, soil-specific. That is, a matrix of guidelines are generated for each contaminant
depending on the values of various soil physicochemical properties known to modify toxicity. This
presentation will discuss the ways that SSDs are used in Australia and New Zealand and the
proposed changes arising from the current review of the Australian and New Zealand water quality
guidelines.
Page 45
SHORT C.V.
Michael Warne
Michael is currently the Science Leader of the Water Quality and Investigation in the
Queensland Department of Science, Information Technology, Innovation and the Arts. Prior
to this he was a Principal Scientist at CSIRO and has held positions in the New South Wales
EPA and the Universities of Queensland and Griffith University. Michael has conducted
extensive research on the impacts of individual metal and organic contaminants and
mixtures in terrestrial and aquatic ecosystems. He has worked extensively integrating the
results of ecotoxicology into regulatory tools for managing chemicals in the environment –
particularly the development of environmental quality guidelines. He was a key scientist in
the development of the current Australian and New Zealand water quality guidelines and led
national programs that developed contaminant guidelines for biosolids, mineral fertilisers and
contaminated sites. He is participating in the current review of the Australian and New
Zealand water quality guidelines. He has been awarded six outstanding performance awards
and nominated for a Queensland Government Public Service Medal. Michael has been
awarded over $8 million in research grants and published two editions of a book, six book
chapters, 90 publications in scientific journals, eight Australian national guidelines for the
environmental management of chemicals, and over 150 conference proceedings and client
reports.
Page 46
ABSTRACT
Use of SSD in China
Fengchang Wu
Chinese Research Academy of Environmental Sciences, China
Species sensitivity distributions (SSDs) are usually used in the development of water quality
criteria (WQC) and require a large number of toxicity values to define a hazard level to
protect the majority of species. In the present study, we introduced the specific use of SSD
in the study of water quality criteria in China. As case studies, WQC for representative water-
body pollutants in China using SSD were conducted. i.e., toxicological data for zinc (Zn),
cadmium (Cd), hexavalent chromium (Cr (VI)), benzene, and nitrobenzene were collected
from various databases, publications and experimental test data. And then these
toxicological data were screened and then constructed into SSD curves. Then WQC for
protection of the freshwater aquatic life in China against five representative pollutants were
derived. The values derived in this study were compared with those issued by the US
Environmental Protection Agency and the Chinese national environmental standard for
surface water to identify factors underlying the differences. The results showed that the SSD
curves for the five pollutants differed significantly, with the examined aquatic species being
generally more sensitive to Zn, Cd, and Cr (VI) than benzene and nitrobenzene.
While SSDs based on measured toxicity values can provide a strong level of confidence for
environmental protection, there is still some uncertainty in their applicability for untested
species. Additionally, SSD development has been limited to a relatively few chemicals
because of the requirement for toxicity data for a broad diversity of taxa. Interspecies
correlation estimations (ICE) models may provide great assistance for addressing the
development of WQC that are protective of species that cannot be tested. To address this
need, we also tried to use ICE-based SSD in deriving WQC for zinc in China. Taken zinc for
example, ICE-based-SSDs were generated using three surrogate species (common carp
(Cyprinus carpio), rainbow trout (Oncorhynchus mykiss), and Daphnia magna) and
compared with the measured-based SSD and corresponding HC5. The results showed that
no significant differences were observed between the ICE- and the measured-based SSDs
and HC5s. Given the similarity of SSD and HC5s for zinc, the use of ICE to derive potential
water quality criteria for diverse chemicals in China is proposed. Further, a combination of
measured and ICE-derived data will prove useful for assessing water quality and chemical
risks in the near future. Above all, the comparative study of SSD in WQC studies may offer
guideline values for future WQC studies for China.
Page 47
SHORT C.V.
Fengchang Wu
Dr. Fengchang Wu is currently professor and deputy General Director, Chinese Research
Academy of Environmental Sciences, Ministry of Environmental Protection of China.
• 2010-present: Professor, Director of State Key Laboratory of Environmental Criteria and
Risk Assessment, Chinese Research Academy of Environmental Sciences, Beijing, China
• 2007-present: Professor and director, State Environmental Protection Key Laboratory of Lake
Pollution Control, Chinese Research Academy of Environmental Sciences, Beijing, China
• 2004-2008: Professor, and deputy director, State Key Laboratory of Environmental
Geochemistry, Institute of Geochemistry, Chinese Academy of Sciences, Guiyang, China
Expertise
Water quality pollution mechanism and process
Environmental criteria and risk assessment
Environmental pollution control technology and management research
Recent contributions and initiatives
Chief Scientist, National Key Basic Research Project (973 project) “Key Lake Environmental
Quality Changes and Water Quality Criteria in China” (No: 2002CB418200), 2007-2012.
Chief Scientist, Key Project of Environmental Ministry of China “preliminary investigation of
Environmental Criteria in China”, 2009-2011
Overview of publications and awards publications
Over 160 peer-review papers
6 books (in Chinese and in English)
Over 50 presentations worldwide
Awards
• New Century 100-1000-10000 Talents Programt” (National levels)
• Excellent Young Science and Technology Award, China Government, 2006
• National Excellent Young Scientist Fellowship, Natural Science Foundation of China (NSF), 2005
• ESTANSGP Best article Award, Excellent Science and Technology Articles in Natural
Sciences in Guizhou Province (ESTANSGP), 2006
• STAGP Best Award, Science and Technology Award in Guizhou Province (STAGP)
• 2004 Hou DeFeng Young Scientist Award, Chinese Association of Mineral, Petrology
and Geochemistry, 2004
• Excellence Award of “One-hundred Scientist Program”, Chinese Academy of Sciences, 2004
• “One-hundred Scientist Program” Fellowship, Chinese Academy of Sciences
• Excellence Award for Ph.D. Graduate, Institute of Geochemistry
• Award for “The state environmental protection science and technology “ first prize,2012
• Award for “National science and technology progress prize” second prize, 2013
Page 48
ABSTRACT
Use of SSD to derive no-effect thresholds for water quality guidelines and
ecological risk assessments in Canada
Anne Gosselin
Environment, Canada
Authors: A. Gosselin1, D.J. Spry1, S. Dixit1, S. Teed2 and M. Bonnell1
1Environment Canada, Gatineau, Canada; 2Intrinsik Environmental Sciences Inc., Ottawa, Canada
In Canada, species sensitivity distributions (SSDs) are used to derive “no effect” thresholds
that serve to determine water quality guidelines for aquatic life as well as Predicted-No-
Effect-Concentrations (PNECs) in ecological risk assessments of chemicals. The Federal
Water Quality Guidelines (FWQGs) are developed to meet the needs of risk assessment and
risk management of chemicals under the Canadian Environmental Protection Act, 1999
(CEPA 1999). In addition, Canadian WQGs are developed under the auspices of the
Canadian Council of Ministers of the Environment (CCME) based on priorities identified by
federal, provincial and territorial jurisdictions. Moreover, under CEPA 1999, regulatory
ecological and human health risk assessments are conducted for substances identified as
priorities on Canada’s Domestic Substances List.
190BThe FWQGs, CCME guidelines and PNECs used in ecological risk assessments all
identify thresholds for aquatic ecosystems that are intended to protect all forms of aquatic life
and all life stages for indefinite exposure periods. The methodology used to derive these
thresholds is the “Protocol for the Derivation of Water Quality Guidelines for the Protection of
Aquatic Life” (CCME 2007). SSD is the preferred approach for FWQGs, CCME guidelines
and PNECs. It follows these steps: toxicity data collection, evaluation and selection, SSD
plotting, verification of statistical assumptions including determination of the goodness-of-fit
(i.e., selection of the model), and determination of the FWQG, CCME guideline and/or
PNEC. They are set at the 5th percentile of the SSD, which may, in the case of the PNEC
used for risk assessment, be divided by an assessment factor if deemed necessary.
191BExamples of the use of SSDs in Canada will be presented, including the federal water
quality guidelines and risk assessment for metals (vanadium and cobalt) and the
antimicrobial triclosan.
Page 49
SHORT C.V.
Anne Gosselin
Ms. Anne Gosselin is a Senior Evaluator in the Ecological Assessment Division of
Environment Canada. She has been in this position for the last six years conducting
ecological risk assessments of chemicals under Canada’s Chemicals Management Plan.
She specializes in the assessment of metals. From 2004 to 2007, Ms. Gosselin was working
as an evaluation officer at the Pest Management Regulatory Agency of Health Canada
where she conducted risk assessments of pesticides. She started her career as a research
project manager working with the Quebec Ministry of the Environment on a project pertaining
to the toxicity of industrial wastes. Ms. Gosselin holds a M.Sc. degree in Water Sciences
from INRS – University of Quebec, and a B.Sc. degree in Biology from Laval University. In
her current position at Environment Canada, she has led preparation of a number ecological
assessment reports, covering a range of substances, notably leading on metals related
issues.
Page 50
ABSTRACT
Use of SSDs in the USA – endangered species and water quality criteria
Mace Barron
Environmental Protection Agency, USA
Species sensitivity distributions (SSDs) are used in the United States (US) in the
development of national ambient water quality criteria (AWQC), with site-specific and
numeric modifications to protect sensitive taxa including threatened and endangered
species. The US Environmental Protection Agency (EPA) first used SSDs constructed of
acute toxicity values in 1978, with formal guidance issued in 1985 for computing 5th
percentile hazard concentrations (HC5) from SSDs constructed of at least eight families with
acceptable toxicity data. Additional minimum data requirements (MDRs) include acceptance
of only North American species and specific taxa diversity requirements that have limited the
development of AWQC to only 47 chemicals. EPA is currently considering alternative
approaches for developing SSD-based AWQC, with the recognition that species composition
appears to affect HC5 estimates for aquatic species more than differences in geography or
habitat of the assemblage. The protectiveness of SSD hazard concentrations used in
endangered species risk assessment remains a concern because of uncertainty in sensitivity
compared to standard test species. In a recent study, the relative sensitivities of US federally
listed and non-listed aquatic species were compared for a broad range of chemicals. The
SSD HC5s and HC1s were lower than 97 and 99.5% of all endangered species mean acute
LC50s, indicating that the use of SSDs as distribution-based risk assessment and criteria
development approaches can be generally protective of listed species. A recent US National
Academy of Sciences report suggested SSDs should be applied in endangered species risk
assessments as an alternative to general uncertainty factors. This presentation will overview
US applications of SSDs in AWQC development and listed species assessment, and include
perspectives on modifying MDRs and adopting new approaches to meet taxa diversity
requirements.
Page 51
SHORT C.V.
Mace Barron
Dr. Mace G. Barron is Acting Associate Director of Science at the EPA Gulf Ecology
Division, in Gulf Breeze, Florida, USA. Mace has worked in EPA’s Office of Research and
Development for 10 years and directed ecological effects research and model development
to predict risks to endangered species, reef building corals, and marine fish populations. He
received his B.S. and M.S. degrees in Fisheries Science, a PhD in
Pharmacology/Toxicology, and conducted post-doctoral research on chemical
bioaccumulation and biotransformation in crustaceans and fish. He has over 25 years of
work experience, including senior scientist positions in the chemical industry and in firms
performing Natural Resource Damage assessments. He has published over 100 peer
reviewed journal articles and book chapters on chemical bioaccumulation, ecological risk
assessment, photo enhanced toxicity, toxicity estimation and species sensitivity distributions
(SSD). Mace is co-developer of EPA’s Web-ICE toxicity estimation tool, and has published
extensively on the technical basis of interspecies toxicity extrapolation and SSD
development. Currently Mace is a member of the US EPA Risk Assessment Forum and
effects lead for the EPA team developing approaches for applying computational chemistry
to aquatic toxicity mode of action assignment and QSAR model development.
Page 52
LIST OF PARTICIPANTS
Name Affiliation E-mail
Tom Aldenberg RIVM, NL tom.aldenberg@rivm.nl
Sandrine Andres INERIS, France sandrine.andres@ineris.fr
Timothy Barber Environ, USA tbarber@environcorp.com
Mace Barron US Environmental Protection Agency barron.mace@epamail.epa.gov
Scott Belanger Procter & Gamble, USA belanger.se@pg.com
Peter Chapman Unilever, UK peter.chapman@googlemail.com
Christian Collin-Hansen Statoil, Norway chrc@statoil.com
Peter Craig University of Durham, UK p.s.craig@durham.ac.uk
Pepijn de Vries IMARES, NL pepijn.devries@wur.nl
Dick de Zwart RIVM, NL dick.de.zwart@rivm.nl
Jean Lou Dorne EFSA, Italy jean-lou.dorne@efsa.europa.eu
Sabine Duquesne UFZ, Germany sabine.duquesne@uba.de
Scott Dyer Procter & Gamble, USA dyer.sd@pg.com
Charles Eadsforth Shell, UK charles.eadsforth@shell.com
Chenglian Feng Chinese Research Academy fengcl@craes.org.cn
Malyka Galay Burgos ECETOC, Belgium malyka.galay-burgos@ecetoc.org
John Paul Gosling University of Leeds, UK j.p.gosling@leeds.ac.uk
Anne Gosselin Environment Canada anne.gosselin@ec.gc.ca
Maike Habekost BASF, Germany maike.habekost@basf.com
Mick Hamer Syngenta, UK mick.hamer@syngenta.com
Andy Hart FERA, UK andy.hart@fera.gsi.gov.uk
Jan Hendriks Radboud University, NL a.j.hendriks@science.ru.nl
Marion Junghans EAWAG, Switzerland marion.junghans@oekotoxzentrum.ch
Guillaume Kon Kam King Université Claude Bernard, France guillaume.konkamking@gmail.com
Kenneth Leung University of Hong Kong, China kmyleung@hku.hk
Ailbhe Macken NIVA, Norway ama@niva.no
Lorraine Maltby University of Sheffield, UK l.maltby@sheffield.ac.uk
Stuart Marshall Unilever, UK stuart.marshall@unilever.com
Christian Michel University of Basel, Switzerland christian.michel@unibas.ch
Yuan Pan University of Sheffield, UK ypan8@shef.ac.uk
Adam Peters WCA Environment, UK adam.peters@wca-environment.com
Leo Posthuma RIVM, NL leo.posthuma@rivm.nl
Ad Ragas Radboud University, NL a.ragas@science.ru.nl
Sandy Raimondo US Environmental Protection Agency raimondo.sandy@epamail.epa.gov
Hans Sanderson University of Aarhus, Denmark hasa@dmu.dk
Krishna Kumar Selvaraj Bharathidasan University, India kumarspeed@gmail.com
Keith Solomon University of Guelph, Canada ksolomon@uoguelph.ca
Paul van den Brink Alterra, Wageningen University, NL paul.vandenBrink@wur.nl
Michael Warne DSITIA, Australia michael.warne@science.dsitia.qld.gov.au
Richard Wenning Environ, USA rjwenning@environcorp.com
Paul Whitehouse Environment Agency, England paul.whitehouse@environment-agency.gov.uk
Fengchang Wu Chinese Research Academy wufengchang@vip.skleg.cn
Zhen-guang Yan Chinese Research Academy zgyan@craes.org.cn
Page 53
NOTES
LOGISTICS
VENUE
NH Grand Hotel Krasnapolsky
Dam 9, 1012 JS Amsterdam (The Netherlands). Tel. +31.20.5549111
(B on the map below, a 0.8 km 10 minute walk from the Central Station – A on the map)
The NH Grand Hotel Krasnapolsky hotel in Amsterdam is located in the very heart of the
city, right on the central Dam square near many major tourist and business attractions. If you
arrive by car, the hotel has a parking attendant at your service and a secure paid garage.
There are bus and tram stops right outside the hotel. The central train station is only 5
minutes away by public transport and Schiphol Amsterdam Airport is just a 25 minute taxi or
car ride.
The restaurant for the evening of 11 February is a 0.6km 8 minute walk from the hotel (C on
the map): Restaurant Haesje Claes, Spuistraat 273-275, Amsterdam, Tel.+31 20 624 9998
The restaurant for the evening of 12 February is also a 0.6km 8 minute walk from the hotel (D on
the map): Restaurant D’Vijff Vlieghen, Spuistraat 294-302, Amsterdam, Tel.+31 20 530 4060
REGISTRATION AND ENQUIRIES
ECETOC
Av. E. Van Nieuwenhuyse 2, Box 8, B-1160 Brussels, Belgium
Tel. +32 2 663 3810. E-mail: sonia.pulinckx@ecetoc.org
Click on map to see online version

Navigation menu