Multiple-Lines-of-Evidence Approaches to Address Complications to - PowerPoint PPT Presentation

Multiple-Lines-of-Evidence Approaches to Address Complications to Vapour Intrusion Pathway Assessments Robert Ettinger Geosyntec Consultants October 28, 2014

Challenges to Vapour Intrusion Assessments  Sensitive subject for many stakeholders  The subject of new and changing regulatory guidance and litigation  Closed sites reopened to address vapour intrusion pathway  Affecting property transactions  Technically challenging pathway  Data needs and interpretation for “MLE” assessments  Evaluation of background contributions to indoor air  Understanding uncertainties associated with vapour intrusion modeling 2

Managing Uncertainties  Compounding Limited Site Characterization conservative More conservative assumptions can risk mgmt. decisions lead to overly conservative Reduced site conclusions characterization  Balance requirements Greater understanding of contaminant fate uncertainties to (e.g., bioattenuation) improve risk-based decision making process Less conservative modeling Detailed site characterization 3

Direction of Regulatory Guidance  Increased reliance on multiple lines of evidence “MLE” assessments to address spatial and temporal variability  Exclusion criteria for petroleum vapour intrusion  More cautious screening evaluations  Consideration of short-term action levels  Less reliance on vapour intrusion modeling / greater use of indoor air sampling  Alternate lines of evidence to evaluate indoor air background sources  Increased emphasis on engineering controls and pre-emptive mitigation 4

Multiple Lines of Evidence Investigation Approach Indoor Air Evaluation • Risk Management Decisions • Background Contributions • Mitigation Options Top-Down Bottom-Up Vapour Intrusion To Building • Soil Characteristics • Building Characteristics Source Characterization Source • Groundwater, Soil, Soil Vapour Concentration Distribution 5

Importance of a Conceptual Site Model Develop Conduct  CSM should characterize Initial CSM Investigation potential sources, fate and transport pathways, and receptors (e.g., buildings) Review/  Use CSM for investigation Update CSM planning and identify pros / cons of different lines of evidence  Not all lines of evidence have equal weight in VI evaluation Need  Consider CSM when interpreting Add’l Yes Data? investigation data  Resolve differences between No data and CSM Proceed with Corrective Action 6 Planning / NFA

Common Views of Vapour Intrusion Models  There is a range of opinions about the use of models for vapour intrusion pathway assessments  Most modeling questions arise from:  Different expectations for accuracy of model results (i.e., typical vs RME estimates)  Deviation from conceptual model used for vapour intrusion model development  Use of inappropriate input parameters  Uncertainty of significance of model input parameters  Some combination of data collection and modeling is usually appropriate 7

Vapour Intrusion Models • There are many options for VI models available • Model selection is dependent on what you know about the site and the level of desired assessment • Regulatory attenuation factors are simple VI models Empirical Analytical Numerical  USEPA Database  Johnson and Ettinger (1991)  VAPOURT (1989)  USEPA VISL  Little et al. (1991)  Sleep & Sykes (1989) Calculator  San Diego SAM  RUNSAT (1997)  VOLASOIL (1996)  Abreu & Johnson (2005)  Krylov and Ferguson (1998)  VIM (2007)  DLM - Johnson et al. (1999)  Brown University (2007)  DeVaull (2007)  BioVapor (2010) 8

USEPA Empirical Attenuation Factors  Empirical data from over 900 buildings at over 40 sites from across the country  Data are predominantly from residential buildings  Majority of data from a few sites  Used paired data (indoor air and sub-surface data) to calculate empirical attenuation factors  Filtered data to screen out poor data quality and results impacted by background sources 9

USEPA Empirical Attenuation Factors  US regulatory agencies focus on 95%ile values  USEPA database results may be biased by background impacts  May not be relevant to non- residential scenarios  Be careful if simply using empirical factors 95 th percentile J&E Model Prediction 10

USEPA Empirical Attenuation Factors Empirical Attenuation Factors Source # of Data Median 95%ile Model Pairs Predict Crawl Space 41 0.39 0.90 NC Sub-Slab Soil 411 0.0027 0.026 0.0024 Gas Soil Gas 106 0.0038 0.25 0.0013 Groundwater 743 0.000074 0.0012 0.00041  Always keep limitations of empirical attenuation factors in mind in risk-based decision making process 11

Vapor Intrusion Attenuation Factor  USEPA VI database study is commonly referenced to estimate VI attenuation factor, but limitations of study should be recognized  Data predominantly from single family homes  Difficult to completely address background effects  Natural vadose-zone biodegradation effects not captured  Use of 95%ile empirical factors will over-state potential risks  Ability for site-specific modeling/assessment is important 12

Site-Specific Modeling  Site-specific modeling can be useful tool to characterize uncertainty in preliminary screening vapour intrusion assessments  Consider differences in site conceptual model from default assumptions  Focus site-specific inputs for “critical” parameters that can be well characterized (see Johnson, 2003)  Additional support may be needed if calculated results are significantly different from expected range 13

Vapour Intrusion Critical Processes for Modeling Evaluation Key Sensitivity Measurements Process Considerations Soil type, VI decreases when Continuous boring Diffusive Transport moisture content, higher moisture logs, soil property (Diffusive Flux) presence of content soils are data, in-situ groundwater present diffusivity test, VOC soil gas profile Varies by building Increasing ventilation Building ventilation Building Ventilation use/design reduces indoor air rate concentrations Default values Key parameter for Cross-slab pressure Soil Gas typically used sub-slab data or pos. Convection press. Groundwater to Uncertainty reduced Soil gas samples for Partitioning soil gas by collection of soil source relationship gas samples characterization 14

Evaluation Framework for Petroleum Hydrocarbons  PVI is different than VI for chlorinated compounds  PVI rarely shown to be a complete pathway due to natural biodegradation in vadose-zone soils  Investigation strategies for chlorinated sources are not well-suited for many petroleum release sites From API, 2004 PVI Conceptual Model  PVI guidance focuses on identifying site conditions where PVI is not of concern (exclusion criteria)  USEPA OUST and ITRC developed guidance on parallel tracks 15

USEPA PVI Database  Data from 74 sites  Predominantly UST sites, but data from terminals, refineries, and petrochemical sites included  Data analysis focuses on paired soil vapour and groundwater data to identify distance for vertical attenuation in vadose zone  Distance to attenuate to 50 – 100 µg/m 3  Different from analysis for CVOCs due to background sources of petroleum compounds 16

USEPA PVI Database Dissolved-Phase Source From USEPA, 2013. Evaluation of Empirical Data to Support Soil Vapor Intrusion 17 Screening Criteria for Petroleum Hydrocarbon Compounds, EPA 510-R-13-001.

USEPA PVI Database LNAPL Source From USEPA, 2013. Evaluation of Empirical Data to Support Soil Vapor Intrusion 18 Screening Criteria for Petroleum Hydrocarbon Compounds, EPA 510-R-13-001.

Considerations for Sites That Do Not Meet Exclusion Criteria  Petroleum hydrocarbons degradation occurs for wide range of petroleum contaminated sites  Typically re-visit multiple-lines-of-evidence approach  Data collection to assess biodegradation (soil vapour or sub-slab soil vapour probes)  Modeling  Indoor air sampling is challenging for petroleum hydrocarbons  Consider whether remediation and/or mitigation is warranted 19

Indoor Air Sampling  Indoor air concentration measurements are used to make decisions about potential health risks, but there are difficulties with sampling and interpretation.  Challenges to indoor air sampling  VOCs frequently detected  Occupant disruption  Temporal and spatial variability WMS  Interpretation for future Sampler land development scenarios  Background effects 20