This article explores the limitations of conventional field screening in environmental site assessments, particularly in heterogeneous soil conditions. It introduces the US EPA’s Triad Approach as a more dynamic and data-driven framework—emphasizing systematic planning, adaptive sampling, and real-time decision-making. AISCT (Artificial Intelligence Site Characterization Technology) is highlighted as a tool that enhances confidence in field data through real-time QA/QC metrics like %RPD and %COV. A field case study demonstrates how AISCT was used to identify inconsistencies in stockpile mixing, leading to improved treatment accuracy and cost savings. The article challenges outdated expectations of lab-to-field equivalence and advocates for a smarter, more transparent approach to site characterization.
Part 1 : Confidence in a Heterogeneous World
In contaminated site assessment, the stakes are high. Decisions made on partial or misinterpreted data can result in unnecessary soil removal, regulatory setbacks, or—worse—remediation failures that risk human and environmental health. Despite decades of guidance, field screening remains either undervalued or misused, often judged unfairly by its lack of direct numerical equivalence with laboratory results. This must change.
The US EPA’s Triad Approach is built on three pillars:
Rather than seeing field screening as a compromise, Triad places it at the heart of dynamic decision-making—helping teams understand site variability, manage uncertainty, and make real-time adjustments. Field screening doesn’t replace labs— when used properly, they can reveal contaminant distributions and heterogeneity patterns that lab-only sampling strategies often miss.. [Source: US EPA, “The Triad Approach: A New Paradigm for Environmental Project Management,” EPA 542-R-01-016].
Soils are naturally variable. Texture, moisture, and contaminant partitioning differ across short distances. Yet, the industry still expects field screening to mirror lab results, disregarding the inherent uncertainties of subsurface sampling. Take Relative Percent Difference (%RPD). Elevated RPDs in duplicates often surface after fieldwork, wrongly blamed on lab or handling errors. More often, they signal natural heterogeneity—something that can't be corrected after the fact.
AISCT (Artificial Intelligence Site Characterization Technology) tackles these challenges head-on. By embedding QA/QC routines into high-density field screening, AISCT allows teams to:
Crucially, these metrics are measured in real-time, during fieldwork—not weeks later in a lab.
At a Northern mining site, AISCT was deployed across five stockpiles (R0–R5). Key findings:
This proactive field assessment delayed chemical dosing on R0 and R5 until blending achieved homogeneity, reducing amendment volumes and optimizing treatment accuracy.
EPA case studies show Triad-based field screening can reduce project costs by 30–60%, slash timelines by half, and enhance decision defensibility. When AISCT is integrated, these benefits are magnified through precision and repeatability.
Field screening should not be a compliance checkbox—it must be a core strategy for confident, cost-effective, and sustainable site decisions.
In Part 2, we’ll explore how to design and implement a Triad-informed field program using AISCT—covering statistical planning, data validation, and real-time visualization tools that bring site data to life. We will learn how TRIUM is ;
Stay tuned.
https://www.linkedin.com/pulse/rethinking-field-screening-aisct-new-standard-ulilc