TABLE OF CONTENTS

• Incremental Sampling Methodology Homepage
  • Executive Summary
  • Acknowledgements
  • About ITRC
  • Disclaimer
  • Table of Contents
  • List of Tables
  • List of Figures
• 1.0 Introduction
  • 1.1 Summary of ISM as an Environmental Sampling Approach
  • 1.2 Traditional Investigation Approach Limitations
  • 1.3 How ISM Addresses Traditional Investigation Approach Limitations
  • 1.4 How ISM Compares to Compositing
  • 1.5 Purpose
  • 1.6 Frequently Asked Questions
  • 1.7 Document Organization
  • 1.8 Key Terms
• 2.0 Nature of Soil Sampling and Incremental Sampling Principles
  • 2.1 Introduction
  • 2.2 Soil Heterogeneity & Variation in Contaminant Concentrations
    • 2.2.1 Microscale Heterogeneity
    • 2.2.1.1 The smallest particles often have the highest contaminant concentrations
    • 2.2.1.2 Why laboratory duplicates often fail to match
    • 2.2.2 Short-Scale Spatial Heterogeneity
    • 2.2.3 Large-Scale Spatial Heterogeneity
  • 2.3 Foundational Concepts of Sampling
    • 2.3.1 All Concentrations Are Means
    • 2.3.2 Representative Soil Samples
  • 2.4 Scale-Specific Sampling Considerations
    • 2.4.1 Sampling Considerations—Microscale Heterogeneity
    • 2.4.1.1 Sampling error as a consequence of particle size and sample handling
    • 2.4.1.2 Measuring the error caused by within-sample heterogeneity
    • 2.4.1.3 The effect of subsample mass on data variability
  • 2.5 Gy Theory and the Source of Sampling Error
    • 2.5.1 Pierre Gy's Sampling Theory and the Seven Basic Sampling Errors
    • 2.5.2 Compositional Heterogeneity
    • 2.5.3 Distributional Heterogeneity
    • 2.5.4 Long-Range and Periodic Heterogeneities
    • 2.5.5 Device and Preparation Errors
    • 2.5.6 Controlling Gy Errors
  • 2.6 Three Sampling Approaches
    • 2.6.1 Discrete Sampling
    • 2.6.1.1 Heterogeneity and discrete samples
    • 2.6.1.2 Discrete sampling plans
    • 2.6.1.3 Interpreting results of discrete sampling
    • 2.6.1.4 When discrete sampling may be successful
    • 2.6.2 Composite Sampling
    • 2.6.2.1 Beneficial uses of compositing
    • 2.6.2.2 Poorly designed composite sampling
    • 2.6.3 Incremental Sampling Methodology
• 3.0 Systematic Planning and Decision Unit Designation
  • 3.1 Overview of Systematic Planning
    • 3.1.1 Project Planning Team
    • 3.1.2 Conceptual Site Models
    • 3.1.3 Investigation Objectives
    • 3.1.4 Data Needs and the Selection of Sampling Approaches
    • 3.1.4.1 Spatial scale of DUs
    • 3.1.4.2 Sampling approach
  • 3.2 Nature and Intent of Decision Units and Sampling Units
    • 3.2.1 Planning DUs and SUs
    • 3.2.2 Use of Sampling Units
  • 3.3 Decision Units
    • 3.3.1 Defining Decision Units
    • 3.3.2 Types of Decision Units
    • 3.3.3 Vertical Definition of DUs
    • 3.3.4 Exposure Area Decision Units
    • 3.3.4.1 Residential exposure area decision units
    • 3.3.4.2 Commercial/industrial exposure area decision units
    • 3.3.4.3 Ecological exposure area decision units
    • 3.3.4.4 Exposure area decision units based on preferential exposure
    • 3.3.4.5 Example exposure area decision units
    • 3.3.5 Source Area Decision Units
    • 3.3.5.1 Smaller source area decision units within exposure area DUs
    • 3.3.5.2 Example source area decision units
    • 3.3.6 Decision Units Surrounding Source Areas
    • 3.3.7 Subsurface Decision Units
    • 3.3.8 Stockpile Decision Units
    • 3.3.9 Excavation Decision Units
    • 3.3.10 Decision Units for Very Large Areas
    • 3.3.11 Other Types of Decision Units
  • 3.4 Establishing New Decision Units Based on Previous Results
  • 3.5 Hot Spots
• 4.0 Statistical Sampling Designs for ISM
  • 4.1 Factors that Complicate Estimating the Mean Concentration
    • 4.1.1 Skewness and Dispersion
    • 4.1.2 Spatial Scale, Mixtures, and Autocorrelationn
  • 4.2 Uncertainty in Estimates of the Decision Unit Mean
    • 4.2.1 One ISM Result
    • 4.2.2 UCL Calculation Method
    • 4.2.2.1 UCL of the mean based on Student's-t distribution
    • 4.2.2.2 UCL of the mean based on Chebyshev inequality
    • 4.2.3 Nondetects
  • 4.3 Evaluating the Performance of Sampling Approaches
    • 4.3.1 Definitions of Performance Metrics
    • 4.3.1.1 Coverage and magnitude of UCL errors
    • 4.3.1.2 Bias in estimated mean
    • 4.3.1.3 Relative standard deviation of replicate samples
    • 4.3.2 Simulation Study Approach
    • 4.3.3 Objectives of the Simulation Studies
    • 4.3.4 Simulation Study Findings on ISM Performance
    • 4.3.4.1 Sample size (number of increments and replicates)
    • 4.3.4.2 Effects of sampling pattern
    • 4.3.4.3 Partitioning the DU
    • 4.3.4.4 Relative standard deviation
  • 4.4 Areas for Further Study
    • 4.4.1 Combining DUs
    • 4.4.2 Extrapolating from DUs
    • 4.4.3 Comparing DUs
    • 4.4.3.1 Site-to-site comparisons
    • 4.4.3.2 Incremental to discrete sample comparisons
    • 4.4.3.3 Site-to-background comparisons
    • 4.4.4 Oversized DUs
    • 4.4.5 Explicitly Address Additional Sources of Error
• 5.0 Field Implementation, Sample Collection and Processing
  • 5.1 Introduction
  • 5.2 Sampling Tools
  • 5.3 Field Collection
    • 5.3.1 Surface ISM Samples
    • 5.3.2 Subsurface ISM Samples
    • 5.3.2.1 Subsurface ISM samples using core sampling
    • 5.3.2.2 Additional subsurface ISM considerations
    • 5.3.3 Stockpile ISM Samples
    • 5.3.4 ISM Confirmation Sampling
    • 5.3.5 Collection of Field Replicate ISM Samples
  • 5.4 Field Handling of ISM Samples
    • 5.4.1 ISM Samples for Non-VOCs
    • 5.4.2 Volatile Organic Compound ISM Samples
• 6.0 Laboratory Sample Processing and Analysis
  • 6.1 Introduction
  • 6.2 Laboratory Processing
    • 6.2.1 Volatile Organic Compounds
    • 6.2.2 Semivolatile Organic Compounds and Inorganic Contaminants of Concern
    • 6.2.2.1 Identification of sample
    • 6.2.2.2 Bulk sample mass reduction via sample splitting
    • 6.2.2.3 Sample conditioning
    • 6.2.2.4 Sample mixing
    • 6.2.2.5 Particle size reduction
    • 6.2.2.6 Particle size selection
    • 6.2.2.7 Analytical splitting and subsampling techniques
  • 6.3 Laboratory Analysis
    • 6.3.1 General Sample Processing Considerations
    • 6.3.2 Organics
    • 6.3.3 Inorganics
  • 6.4 Quality Assurance/Quality Control
    • 6.4.1 Laboratory Accreditation/Certification
• 7.0 Making Decisions Using ISM Data
  • 7.1 Introduction
  • 7.2 Decision Mechanisms
    • 7.2.1 Decision Mechanism 1: Comparison of One ISM Sample from the DU to the Action Level
    • 7.2.2 Decision Mechanism 2: Comparison of the Mean of Replicate Data from the DU to the Action Level
    • 7.2.3 Decision Mechanism 3: Comparison of the 95% UCL on the Mean of Replicate Data from the DU to the Action Level
    • 7.2.4 Decision Mechanism 4: Comparison to Background
    • 7.2.5 Decision Mechanism 5: Combining DUs
    • 7.2.6 Decision Mechanism 6: Extrapolating from Sampled to Unsampled Areas
    • 7.2.7 Decision Mechanism 7: Evaluating Oversized DUs
  • 7.3 Assessment of Error
• 8.0 Regulatory Concern with ISM
  • 8.1 Introduction
  • 8.2 Perception Issues
  • 8.3 Regulatory Challenges for ISM
  • 8.4 State of Knowledge, Experience, and Training
    • 8.4.1 How ISM IS Being Conducted
    • 8.4.2 Written Guidance
    • 8.4.3 Misconceptions
    • 8.4.4 Limitations
  • 8.5 Implementation Issues
    • 8.5.1 Systematic Planning
      • 8.5.1.1 CSM
      • 8.5.1.2 Sampling objectives and developing the decision unit
      • 8.5.2 Lab Availability
      • 8.5.3 Costs
      • 8.5.4 Challenges in Developing and Using ISM Data
      • 8.5.4.1 Validation of statistical analysis within ISM
      • 8.5.4.2 Meeting regulatory requirements for average and maximum concentration estimates
      • 8.5.4.3 Decision unit versus exposure unit
      • 8.5.4.4 Comparison of discrete samples and ISM samples
      • 8.5.4.5 Comparison of ISM means and "not to exceed" basis regulation
      • 8.5.4.6 Decisions based on a single ISM
      • 8.5.4.7 Background/geochemical limitations
      • 8.5.4.8 Extrapolation between and within DUs
      • 8.5.5 Matrix and Parameter Issues
      • 8.5.5.1 Laboratory experience
      • 8.5.5.2 Sample processing
  • 8.6 Summary
• 9.0 Case Study Summaries
  • 9.1 Case Study 1. PCB-Contaminated Landfill
  • 9.2 Case Study 2. Petroleum-Contaminated Soil Stockpile
  • 9.3 Case Study 3. Former Golf Course Field Demonstration of ISM
  • 9.4 Case Study 4. Hawaiian Homelands Development
• 10. Stakeholder and Tribal Input
• 11.0 References
• Appendix A - Statistical Simulations
  • TABLE OF CONTENTS
  • LIST OF FIGURES
  • A.1 Statistical Simulation Studies Introduction
    • A.1.1 Consensus Points
    • A.1.2 Points of Nonconsensus
    • A.1.2.1 Rationale for the M-3 simulations
    • A.1.2.2 Criticism of the M-3 simulations
  • A.2 Probability Distributions (PD-1)
    • A.2.1 Methods
    • A.2.2 Results
  • A.3 Spatial Autocorrelation Maps (M-1)
    • A.3.1 Methods
    • A.3.2 Results
  • A.4 Maps of RDX and HMX (M-2A and M2-B)
    • A.4.1 Descriptions of Decision Units
    • A.4.1.1 HMX decision unit (M2-A)
    • A.4.1.2 RDX decision unit (M2-B)
    • A.4.2 ISM Sampling Patterns
    • A.4.3 Results Using Discrete Sampling
    • A.4.4 Results Using ISM
    • A.4.4.1 Results for RDX (M2-A)
    • A.4.4.2 Results for HMX (M2-B)
  • A.5 Bulk Material Sampling Maps (M3-A to M3-C)
    • A.5.1 DU Generation and ISM Implementation Method
    • A.5.2 Simulation Results for Scenario M3-A
    • A.5.3 Simulation Results for Scenario M3-B
    • A.5.4 Simulation Results for Scenario M3-C
    • A.5.5 Simulation Results for Scenario M3-C with Modified ISM (Quadrant Subdivision)
  • A.6   Glossary of Terms and Concepts
    • A.6.1 Gy Sampling Principles Applied to Bulk Material
    • A.6.2 Bulk Material Sampling
    • A.6.3 Sample Representativeness and Bias
    • A.6.4 Heterogeneities
    • A.6.4.1 Compositional heterogeneity (CH)
    • A.6.4.2 Distributional heterogeneity (DH)
    • A.6.4.3 Small-scale (short-range) DH
    • A.6.4.4 Sample support
    • A.6.4.5 Fundamental error (FE)
    • A.6.4.6 Large-scale (long-range) DH
    • A.6.5 Homogeneity
    • A.6.5.1 Homogeneous bulk material
    • A.6.5.2 Homogeneous contaminant concentration
    • A.6.6 Sampling Patterns
    • A.6.7 MIS Software to Generate Homogeneous and Heterogeneous DUs
  • A.7.1 Estimation of FE and Var (FE)
  • A.7.2 Computation of Coefficient of Variation (CV) of ISM Replicates
  • A.7.3 Performance Metric Equations
  • A.8 References
• Appendix B - August 2009 Survey Results
• Appendix C - Case Studies
• Appendix D - ISM Team Contacts
• Appendix E - Glossary
• Appendix F - Acronyms