Transcript Application of ERICA outputs and AQUARISK to evaluate

Application of ERICA outputs and
AQUARISK to evaluate radioecological risk
of effluents from a nuclear site
J. Twining & J. Ferris
Outline of talk
• Background
Objective - to demonstrate Radioecological Risk
Assessment
• Dose Assessment software - EXPOSURE
• FASSET Radiation Effects Database - RESPONSE
• Ecological Risk Assessment (AQUARISK)
• Case Study in Radioecological Risk Assessment
• Results
• Conclusions
Quantitative, probabilistic, radioecological risk assessment
Radiological Dose - EXPOSURE
• Radiological Impact Analysis for Coastal Aquatic
Ecosystems V1.15 and Freshwater Ecosystems V1.15
• Converts measured or modelled radioactivity
concentrations in water (Bq L-1) into dose rates (Gy hr-1)
• For a range of radiologically significant nuclides, biota
and habitats
• Each organism is represented as an ellipsoid for LET
calculations
• Assessment of dose to each organism is determined
using concentration factors (internal dose) and
positioning relative to soil/sediment or water (external
dose).
Quantitative, probabilistic, radioecological risk assessment
Radiological Effect - RESPONSE
(FASSET Radiation Effects Database, FRED)
• FRED is a database of published information on the
effects of acute and chronic exposure to ionising
radiation on different biota (EC 5th Framework FASSET
initiative)
• Groups data by:
• wildlife “group” (e.g. amphibians, reptiles,
mammals etc.)
• umbrella endpoint: mutation, morbidity,
reproduction, mortality
• Provides information on dose (rate) response
Quantitative, probabilistic, radioecological risk assessment
Ecological Risk Assessment (AQUARISK)
• A 3-tiered approach
Tier-1: Comparison with regulatory limits or guidelines
Tier-2: Desk-top study involving available and relevant literature data
Tier-3: Site-specific data and modelling
(derived from frequency of observations)
Probability density
• The 2nd & 3rd Tiers use probability density functions to
derive site &/or species specific acceptability criteria
• Convolution of the Exposure and Response PDFs
determines the likely degree of ecological impact and
the extent of
Exposure
Response
remediation
curve
curve
required
% of species
likely to be
affected
(Log) Dose-rate
(µGy hr-1)
reduction in dose-rate required
to achieve a tolerable level of harm
Case Study - Scenarios
• Effluent releases from ANSTO at the LHSTC in
Sydney, Australia
(1) Routine releases into the marine environment at
Potter Point via the sewage system and tertiary
treatment at Cronulla STP
(assumes 735x dilution as realistic for the site, chronic exposure)
(2) Possible accidental release into the Woronora
River after failure of the main holding tank
(assumes no loss of activity overland, no dilution, acute exposure)
Quantitative, probabilistic, radioecological risk assessment
Scenarios
(1) Potter Point
(2) Woronora R
AQUARISK Input data - EXPOSURE
• Monitoring data for 3H, 60Co, 131I & 137Cs over Jan 2002Jun 2003 based on monthly averages
• Activity concentrations were converted to dose rates using
either Coastal or Freshwater RIA software (using updated CFs
and default weighting factors)
• Once converted to dose-rate no differentiation was made
for radionuclide
• Only used output for organisms that corresponded to data
available in the FASSET Radiation Effects Database
(FRED)
Quantitative, probabilistic, radioecological risk assessment
0
Benthic fish
Pelagic fish
Large b.
crust.
Small b.
crust.
Benthic
mollusc
Fish egg
Macrophyte
Zooplankton
Phytoplankton
Dose (microGy/hr)
EXPOSURE Estimation
• Dose-rate from averaged radionuclide concentrations
- Scenario #1
0.1
0.08
137Cs
0.06
131I
0.04
60Co
3H
0.02
EXPOSURE Estimation
• Dose from averaged radionuclide concentrations
- Scenario # 2
0.0025
137Cs
0.0015
131I
60Co
0.001
3H
Benthic fish
Pelagic fish
Large b.
crust.
Small b.
crust.
Benthic
mollusc
Amphibian
Zooplankton
0
Macrophyte
0.0005
Phytoplankton
Dose (Grays)
0.002
AQUARISK Input data - RESPONSE
• FASSET Radiation Effects Database (FRED)
(using categories and information in the FRED to select data for use in
Radioecological Risk Assessment)
• HNEDRs and LOEDRs only, and excluding ‘Background’
data (retains ~10% of available data)
• no distinction based on radionuclide
• no discrimination based on effect measured (all adverse effects
assumed to be ecologically relevant)
• all units converted to Gy hr-1 or Gy (using conservative
assumptions)
Quantitative, probabilistic, radioecological risk assessment
RESPONSE Estimation
• Dose-response cumulative probability (data from FRED)
Results: Tier-1 assessment
• Scenarios (1) and (2) both pass a Tier-1 assessment
against international recommendations
Gy hr-1 NCRP IAEA
Freshwater organisms
400
400
Benthic invertebrates
Fish
Canada
100
50
Garnier LaPlace et. al. 2006 – Freshwater ecosystems  10Gy hr-1
Maximum estimated dose rates for all spp.
(Gy hr-1, n = 162)
(1) Routine release into a marine ecosystem
[(2) Accidental release to a freshwater system
0.3
8.7]
Quantitative, probabilistic, radioecological risk assessment
Results: Tier-2 AQUARISK-derived criteria
(using data selected from the FRED)
Marine
Acute & Chronic
Chronic
only
n
425
85
Units µGy hr-1 µGy hr-1
Species
protection
level
90 %
95 %
740
13
2.2
0.3
Freshwater
Acute & Acute
Chronic
only
52
32
µGy hr-1
Gy
230
33
0.9
0.3
• Criteria for 90-95% protection (using Acute &/or Chronic
data selected from the FRED) cover the range of
international dose-rate recommendations (i.e. 10 – 400 uGy.hr-1)
• Criteria based exclusively on chronic RESPONSE data
are substantially lower
• Max. est. acute dose in Scenario (2) is 0.006 Gy
Quantitative, probabilistic, radioecological risk assessment
Results: Tier-2 Assessment
• Probability of criteria exceedence
Chronic
Marine
Criteria
Value (Gy.hr-1or Gy)
Probability of
exceedence (%)
IAEA, Canada
NCRP
400 100 50
0
0
0
A&C
FW
Acute FW
(Gy)
90% 95% 90% 95% 90% 95%
4.5 0.3 250 40
0.9 0.4
0
0.4
0
0
0
0
Quantitative, probabilistic, radioecological risk assessment
Results: Tier-2 Assessment (cont.)
estimated proportion of affected species
Convolution of the EXPOSURE and RESPONSE
probability density functions indicates the % of species
potentially affected
• Scenario (1) - Marine = 1.5 to 2.5% (depends on selection
of ‘Acute & Chronic’ or ‘Chronic only’ RESPONSE data)
• Scenario (2) - Freshwater = 0.2 to 0.3% (depends on
selection of ‘Acute & Chronic’ dose-rate data or ‘Acute only’ dose
data)
Quantitative, probabilistic, radioecological risk assessment
Conclusions - Case study
• Scenarios (1) and (2) pass the Tier-1 RRA and
hence can be considered of low risk to the
organisms in the receiving environments
• Tier 2 assessments using AQUARISK indicate
lower dose rate criteria may be applied for
chronic, routine releases under Scenario (1)
- Operational efforts should focus on Co-60
• However, low species impacts are predicted for
either scenario even when all conservative
assumptions have been applied in line with the
Precautionary Principle.
• Biomonitoring under scenario (1) has not
shown any adverse effects at Potter Point
Quantitative, probabilistic, radioecological risk assessment
Conclusions - general
• Available software can be conservatively and
successfully applied to RRA
• Calculated exposure criteria (90-95% spp protection)
are comparable to published recommendations, BUT
chronic exposure criteria are substantially lower
• Straightforward technique was used here
- Scenario selections could be more realistic (mixingzone dilutions and bioavailability) and can be refined to
suit other site-specific applications
- Improved selectivity of RESPONSE data will help
(more site-relevant data recovery from the FRED)
Quantitative, probabilistic, radioecological risk assessment
Some Lessons
• Bioavailability (particulate adsorption)
• Co-60 dominant in the marine environment
• Categories within FRED(ERICA)
Quantitative, probabilistic, radioecological risk assessment
Thank you.