Development of Chemistry Indicators - PowerPoint PPT Presentation


PPT – Development of Chemistry Indicators PowerPoint presentation | free to download - id: 6aae3a-ZTU2N


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation

Development of Chemistry Indicators


Development of Chemistry Indicators Steven Bay Southern California Coastal Water Research Project (SCCWRP) – PowerPoint PPT presentation

Number of Views:3
Avg rating:3.0/5.0
Date added: 1 October 2019
Slides: 35
Provided by: ChrisBe155


Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: Development of Chemistry Indicators

Development of Chemistry Indicators
Steven Bay Southern California Coastal Water
Research Project (SCCWRP)
Presentation Overview
  • Workplan update and response to comments
  • Project status
  • Preliminary results
  • Data screening
  • Normalization
  • SQG comparison

Chemistry Indicators
  • A methodology for interpreting sediment chemistry
    data relative to impacts on benthic organisms
    (e.g., an SQG approach with numeric values)
  • Link to pollutants of concern
  • Familiar approach
  • Many available data
  • Several challenges to effective use
  • Bioavailability
  • Unmeasured chemicals
  • Mixtures

  • Identify important geographic, geochemical, or
    other factors that affect relationship between
    chemistry and effects
  • Develop indicator(s) that reflect contaminant
  • Develop indicator(s) that are protective and
    predictive of impacts
  • Develop thresholds for use in MLOE framework

  • Develop a database of CA sediment quality
    information for use in developing and validating
  • Address concerns and uncertainty regarding
    influence of regional factors
  • Document performance of recommended indicators
  • Develop both empirical and mechanistic
    indicators, if possible
  • Both types have desirable attributes for SQO use
  • Investigate existing and new approaches
  • Emphasis is on priority chemicals identified as
    likely causes of impairment

  • Evaluate SQG performance
  • Use CA data
  • Use quantitative and consistent approach
  • Select methods with best performance for expected
  • Describe response levels (thresholds)
  • Consistent with needs of MLOE framework
  • Based on observed relationships with biological

SSC Comments
  • More detail needed regarding data screening,
    matching, establishment of validation dataset
  • Lack of clarity regarding the respective roles of
    empirical and mechanistic guidelines
  • Approaches not interchangeable
  • How will mechanistic guidelines be
  • Should use all available approaches, but how?

An evolving and thorough process, an overview is
included in this presentation
A conceptual plan is included in this
presentation, your input is welcome
SSC Comments
  • Clarify how metals normalization results will be
  • Provide greater independence of chemistry line of
  • More detail needed regarding calibration of
    guidelines and comparison of performance (within
    CA and nationally)

Will explore utility in improving guideline
performance and establishing background
Agree this is an important goal, part of
motivation for using mechanistic guidelines and
metal normalization
A revised comparison approach is proposed that is
more consistent with MLOE framework
  1. Prepare development and validation datasets
  2. Develop and refine SQGs
  3. Evaluate SQGs
  4. Describe response levels

Task 1 Prepare Datasets
Substantial progress made
  • Create high quality standardized datasets for
    development and validation activities
  • Evaluate data quality and completeness
  • matched chemistry and biology
  • Appropriate habitat
  • Data quality, nondetects
  • Calculate derived values
  • e.g., sums, means, quotients
  • Normalize data
  • e.g., metals, TOC
  • Stratify and subset data
  • Independent validation data
  • Address geographic or mixture patterns

Bay/Estuary Samples in Database
Regional Board Chem Tox Benthos Chem Tox Chem Benthos Tox Benthos Chem Tox Benthos
North Coast 6 11 0 22 0 0 34
Central Coast 3 0 0 58 3 0 8
SF Bay 552 19 0 680 37 0 230
Los Angeles 827 11 0 294 15 0 187
Santa Ana 156 8 0 104 0 0 137
San Diego 216 2 0 271 3 0 285
Data Screening
  • Appropriate habitat and geographic range
  • Subtidal, embayment, surface sediment samples
  • Chemistry data screening
  • Valid data (from qualifier information)
  • Estimated nondetect values
  • Completeness (metals and PAHs)
  • Toxicity data screening
  • Target test method selection
  • Valid data (control performance)
  • Lack of ammonia interference
  • Selection of matched data
  • Same station, same sampling event

Bay/Estuary Samples in Database After Screening
Regional Board Chem Tox Benthos Chem Tox Chem Benthos Tox Benthos Chem Tox Benthos
North Coast 0 0 0 13 0 0 34
Central Coast 0 0 0 45 0 0 8
SF Bay 0 0 0 351 0 0 184
Los Angeles 0 0 0 89 0 0 130
Santa Ana 0 0 0 101 0 0 122
San Diego 0 0 0 267 0 0 203
Validation Dataset
  • Used to confirm performance of recommended SQGs
  • Independent subset of SQO database
  • Approximately 30 of data, selected randomly to
    represent contamination gradient
  • Includes acute and chronic toxicity tests

Metal Normalization
  • Metals occur naturally in the environment
  • Silts and clays have higher metal content
  • Source of uncertainty in identifying
    anthropogenic impact
  • Background varies due to sediment type and
    regional differences in geology
  • Need to differentiate between natural background
    levels and anthropogenic input
  • Investigate utility for empirical guideline
  • Potential use for establishing regional
    background levels

Reference Element Normalization
  • Established methodology applied by geologists and
    environmental scientists
  • Reference element covaries with natural sediment
    metals and is insensitive to anthropogenic inputs
  • Use of iron as reference element validated for
    southern California
  • 1994 and 1998 Bight regional surveys

Reference Element Normalization
Reference Element Normalization
Reference Element Normalization
Use ironmetal relationships to Estimate amount
of anthropogenic metal for use in SQG
development Identify background metal
Task 2 Develop/Refine SQGs
Work in progress
  • Investigate a variety of approaches or
    refinements and pursue those with the best
    potential for success.
  • Focus on mixture models, empirical and
  • Apply existing approaches (off the shelf)
  • Refine existing approaches
  • Calibrate existing approaches
  • Develop new approaches

SQG Approaches
SQG Metric Source
ERM Mean Quotient Long et al. CA-specific
Consensus MEC Mean Quotient MacDonald et al, Swartz, SCCWRP
SQGQ-1 Mean Quotient Fairey et al.
Logistic Regression Pmax Field et al. CA-specific
AET Exceedance CA-specific
EqP Organics Sum TU EPA CA Toxics Rule
EqP Metals Potential for Tox. EPA
Mechanistic vs. Empirical SQGs
  • Differences in utility for predicting impacts and
    determining causation
  • Both types of information needed for
    interpretation of chemistry data
  • Mechanistic SQG results will be useful for
    subsequent applications needing to identify cause
    of impairment
  • Anticipate chemistry LOE score will be based on
    combination of SQGs
  • Complementary, not interchangeable
  • Several strategies possible, looking for input on
    recommended approach

Proposed Scoring For Multiple SQGs
High High probability of effect for empirical or EqP organics SQGs
Moderate Substantial probability of effect for empirical or EqP organics SQGs or concordance among SQGs
Marginal Increased probability of effect in at least one SQG
Reference Concordance among all SQGs of low probability of effect (background condition)
Guideline Calibration
  • Use of CA chemistry/effects data to adjust
    empirical guideline models or thresholds
  • LRM model and thresholds
  • Effects range CA-specific values and thresholds
  • AET CA-specific values
  • Consensus SQGQ-1 thresholds
  • Comparisons between existing and calibrated SQG
    results used to guide recommendations
  • Only use calibrated values if improved
    performance can be demonstrated

Task 3 Evaluate Approaches
Work in progress
  • Document and compare performance of candidate
    SQGs approaches in a manner relevant to desired
  • Compare overall discriminatory ability
  • Identify applications
  • Quantify performance
  • Validation dataset
  • Standardized measures
  • Compare performance and identify the most
    suitable approaches

Performance Comparison
  • Approach
  • Focus on empirical guidelines
  • Compare among candidates to select a short list
  • Compare to existing approaches to evaluate need
    for new/calibrated approaches
  • Previous strategy for comparison
  • Current work plan Binary evaluation (effect/no
  • Calculate several measures of performance

Performance Measures
Negative Predictive Value C/(CA) x 100 (percent
of no hits that are nontoxic) SpecificityC/(CD)
x 100 (percent of all nontoxic samples that are
classified as a no hit) Positive Predictive
Value B/(BD) x 100 (percent of hits that are
toxic) SensitivityB/(BA) x 100 (percent of all
toxic samples that are classified as a hit)
Performance Comparison
  • Proposed revised strategy
  • Evaluate ability to classify stations into
    multiple categories
  • More consistent with MLOE approach
  • Less reliance on a single threshold
  • Magnitude of error affects score
  • Utilize both toxicity and benthic impact data

  SQG 1
Kappa Statistic
  • Developed in 1960-70s
  • Used in medicine, epidemiology, psychology to
    evaluate observer agreement/reliability
  • Similar problem to SQG assessment
  • Can incorporate a penalty for extreme
  • Sediment quality assessment is a new application

  SQG 1 (good association between adjacent
SQG 2 (Poor association between adjacent
Task 4 Describe Response Levels
Methodology under development
  • Determine levels of response for the recommended
    SQG approaches
  • Relate SQGs to biological effect indicator
    responses (benthos toxicity)
  • May use statistical methods to optimize
  • Select response levels that correspond to
    objectives for performance and beneficial use

  • Work on many key elements underway
  • Priority is to build upon existing approaches
  • Many of the technical obstacles have been dealt
  • Overall approach is consistent with SSC
  • Include empirical and mechanistic approaches
  • Expect to succeed in selecting recommended SQGs
    for use in MLOE framework
  • Much work remains, especially for development of