ERMS Activity 5.8 Validation(input from PROOF)

• Two main issues:

– How can we control monitor estimated environment risks?

– Check the validityof the currently applied risk values

• The work…

• Next steps…

The work…• Theoretical work

– to establish the conceptual basis for integration of risk assessment and biomarker based monitoring

• Laboratory studies – conducted field relevant exposures to examine relationships between

biomarkers, fitness effects, and predicted ecological risks

• Field studies – Participated in field relevant exposures to validate biomarkers to oil

industry discharges in relation laboratory exposures

• Applied Statistical extrapolation methods (SSDs) – to link biomarker response levels to ecological risk levels– to validate fitness response levels to existing risk curves

• Integrate biomarker distributions in ERMS models – application of the findings to extend the risk models

Key terms•• Fitness Fitness

–– Refers to the ability of the organism to successfully grow Refers to the ability of the organism to successfully grow and reproduce and maintain the population of which it and reproduce and maintain the population of which it forms partforms part

•• BiomarkersBiomarkers–– Refers to any biological response in a living organisms that Refers to any biological response in a living organisms that

results from the exposure to a pollutant chemicalsresults from the exposure to a pollutant chemicals PopulationPopulation

PopulationPopulation

PopulationPopulation

PopulationPopulation

IndividualsIndividuals

CellsCellsMoleculesMolecules

LateLateeffectseffects

EarlyEarlyeffectseffects

Uptake ofUptake ofchemicalschemicals

Biochemical Biochemical responsesresponses

BiomarkersBiomarkers

FitnessFitness

time

Level ofBiologicalorganization

EcosystemEcosystem

NB !

– Fitness data are currently used for Environmental Risk Assessment

– Fitness is difficult to measure in the field

– Fitness and Biomarkers often differ in time and level of biological organization

– Biomarkers can be Early warning of Ecosystem health

A key conceptual element: How is Biomarkers linked with Ecosystem health

• What characterizes a healthy ecosystem?

– The constituent animals, plants and microbes must, on the whole, be healthy

• How do we characterize ecosystem health?

– Biomarkers measure exposure to pollutants and give an assessment ofthe health status of individual animals

– By measuring the health status of a range of species representingdifferent phylogenies and feeding types, we can use a weight ofevidence approach to envisage the ecological concequences ofpollutant exposures

» Depledge & Galloway, Front Ecol Environ 2005; 3(5): 251-258.

Fieldmeasurement(diagnosis)

Link between environmental risk and monitoring

…enables predictions of ecosystem healthwhich can be control monitored

Environmental Risk

Risk probability or Environmentalimpact factors

Prediction(prognosis)

Biomarkersignals

Laboratorytesting

Animal Fitness

Survival and reproductivecapacity

Under Construction

Approach: Compare (statistically)biomarker-response-distributions and fitness-SSDs

Species Sensitivity Distributions for fitnessand biomarker responses

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1 10 100 1000 10000

Concentration THC

Pote

tially

affe

cted

frac

tion

ofsp

ecie

s

Combined risk curve

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1 10 100 1000 10000

Concentration THC

Pote

tially

affe

cted

frac

tion

ofsp

ecie

s

Alkaline unwinding reponseCombined risk curve

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1 10 100 1000 10000

Concentration THC

Pote

tially

affe

cted

frac

tion

ofsp

ecie

s

Lysosomal stability responseCombined risk curve

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1 10 100 1000 10000

Concentration THC

Pote

tially

affe

cted

frac

tion

ofsp

ecie

s

Alkaline unwinding reponseLysosomal stability responseCombined risk curve

And

el p

oten

siel

t påv

irket

e ar

ter

Oil concentration (mg total hydrokarbons pr. liter)

DNA strand breakLysosomal stability

Fitness

(=risk curve)

5%

Above threshold level

Under threshold level

Limit line

Thresholdlevel

Construct Biomarker Sensitivity Distributions (BSDs) !

• Statistical evaluation of biomarker response levels versus risk levels

• Comparison of • Species Sensitivity Distributions applied in risk assessment

(based on literature data) • and Biomarker response distributions

(based on our experimental data)

1. Development of SSDs for different oil component groups2. Construction of risk curve for the different exposures3. Construction of one average risk curve for all exposures4. Construction of BSDs for different types of biomarkers:

DNA damage, oxidative stress, lysosomal stability and PAH metabolites

5. Comparison of biomarker response levels and risk levels

From SSD based on EC50 to risk curve

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

1

1 10 100 1000 10000 100000

Concentration

PAF

SSD based on EC50 valuesChronic SSD

Devide by 100

exposure concentration

corresponding PAF

3. Construction of one average risk curve for all exposures

Risk curve for Goliath and Statfjord exposures

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1 10 100 1000 10000

Concentration THC

Pote

ntia

lly a

ffect

ed fr

actio

n of

spe

cies

Concentration THC

4. Construction of BSDs

• Results from different experiments are combined in BSDs

• The sensitivity of the species tested (Cod, Shrimp, Sheepshead minnow, Sea urchin, Mussel & Scallop) represents the sensitivity of all species

• Four oil types and one produced water exposure

• Lowest Observable Effect Concentrations (LOECs) in biomarkers are applied

• BSDs indicates the variation in lowest exposure levels where the specific biomarkers come to expression for different species

• Biomarkers which respond at low levels but do not respond at higher concentrations are not suited for this. At levels higher than the LOEC the biomarkers must still respond.

Concentration THC

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1 10 100 1000 10000

Concentration THC

Pote

ntia

lly a

ffect

ed fr

actio

n of

spe

cies

DNA damage

Shrimp - Goliat

Sheepshead - North Sea oil

Cod - Goliat

Cod - North Sea oil

Cod - Statfjord B

Scallop - Goliat

Shrimp - Statfjord

Sea urchin - Statfjord

Mussel - Statfjord

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1 10 100 1000 10000

Concentration THC

Pote

ntia

lly a

ffect

ed fr

actio

n of

spe

cies

DNA damageCombined risk curve

Shrimp - Goliat

Sheepshead - North Sea oil

Cod - Goliat

Cod - North Sea oil

Cod - Statfjord B

Scallop - Goliat

Shrimp - Statfjord

Sea urchin - Statfjord

Mussel - Statfjord 5% level

BSD: DNA damage related biomarkers

BSD: Oxidative stress related biomarkers

Concentration THC

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Concentration THC

Pote

ntia

lly a

ffect

ed fr

actio

n of

spe

cies

Oxidative stress

Cod - Statfjord

Shrimp - Goliat

Mussel Statfjord

Cod - Goliat

Sea urchin - Statfjord

Shrimp - Statfjord

Scallop - Goliat

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1 10 100 1000 10000

Concentration THC

Pote

ntia

lly a

ffect

ed fr

actio

n of

spe

cies

Oxidative stressCombined risk curve

Cod - Statfjord

Shrimp - Goliat

Mussel Statfjord

Cod - Goliat

Sea urchin - Statfjord

Shrimp - Statfjord

Scallop - Goliat 5% level

BSD: General toxicity related biomarkers (Lysosomal stability)

Concentration THC

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1 10 100 1000 10000

Concentration THC

Pote

ntia

lly a

ffect

ed fr

actio

n of

spe

cies

Lysosomal stability response

Sea urchin - Statfjord B

Shrimp - Statfjord B

Scallop - Goliat

Shrimp - Goliat

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1 10 100 1000 10000

Concentration THC

Pote

ntia

lly a

ffect

ed fr

actio

n of

spe

cies

Lysosomal stability responseCombined risk curve

Sea urchin - Statfjord B

Shrimp - Statfjord B

Scallop - Goliat

Shrimp - Goliat

5% level

BSD: Exposure related biomarkers (PAH metabolites)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0.1 1 10 100

Sum PAH concentration (water)

Pote

ntia

lly a

ffect

ed fr

actio

n of

spe

cies

PAH metabolites

Cod - Goliat

Sheepshead - North Sea A

Cod - Statfjord B

Cod - North Sea A

Sheepshead - North Sea B

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0.1 1 10 100

Sum PAH concentration (water)

Pote

ntia

lly a

ffect

ed fr

actio

n of

spe

cies

PAH metabolitesCombined risk curve

Cod - Goliat

Sheepshead - North Sea A

Cod - Statfjord B

Cod - North Sea A

5% level

Sheepshead - North Sea B

5. “A Biomarker Bridge”:Comparison of biomarker response levels and risk levels

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Fraction of biota with observed DNA damage (shrimp/mussel/sea urchin/cod/carp)

Ecol

ogic

al ri

sk (P

AF)

Conclusions – part 1

• How can we control monitor estimated environment risks?– A bridge that links environmental risk and biomarkers has been

constructed

– It allows to express environmental risk with biomarker values

– It makes it possible to control measure accepted risik withbiomarker measurements in the field

– There is a need for biomarker- and fitness- measurements for produced water for more animal species to build a sufficientlyrobust BSD (approx. 15 species) which can be used to characterize ecosystem health

– Biomarker data from Svan field study and preliminary resultsfrom PROOF Drilling Discharges project indicate that theapproach will be applicable also for drilling discharges(effects in the water column)

Validation by comparing to laboratory experiments

• Produced water – low dose• Log term (chronic) exposures• Vulnerable life stages• Fitness LOECs

From SSD based on EC50 to risk curve

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

1

1 10 100 1000 10000 100000

Concentration

PAF

SSD based on EC50 valuesChronic SSD

Devide by 100

exposure concentration

corresponding PAF

Risk curve, BSDs and LOEC curve for Goliath and Statfjord exposures

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1 10 100 1000 10000

Concentration THC

Pote

ntia

lly a

ffect

ed fr

actio

n of

spe

cies

Oxidative stressDNA damageLysosomal stability responseLOECs

Sheepshead - North sea A (6 wks)

Sheepshead - North sea B (5 wks)

Shrimp - Statfjord B (3 months)

Shrimp - Goliat (3 months)

Mussel - Statfjord B (3 months)

Fitness LOECs plotted in the normal distribution of the SSD

• Normal distributionNormal distribustion of species sensitivity

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

1 10 100 1000 10000

concentration THC (mg/l)

prob

abili

ty

species sensitivityExperimental LOECs

Conclusions - part 2

•• Checking the validity of existing ERA valuesChecking the validity of existing ERA values

– A validation of the presently used SSD for Risk assessment has been done based on relevant laboratoryexperiments

– The LOEC values obtained from these fitness studies were lower than the present SSD

Next steps

• Biomarkers integrated in ERA– BSDs should be fully developed for PW discharges– BSDs should be considered developed for drilling discharges– The concept should be taken into account in next revision of

monitoring programmes related to oil and gas discharges

• The Validation results with chronic exposuresand vulnerable life stags should be– taken into account in the evaluation of application factors for

ERA related to oil and gas discharges

• Thank you!