When “passed QC” is not the same as “makes sense”
Industrial, utilities, energy and government teams rely on defensible analytical data to guide assessment, compliance and remedial decisions. When addressing critical decisions across priority US Superfund Sites and considering the potential contamination impact for people living in proximity to these sites, defensible data is central to decision-making. Across these projects, data quality influences not only technical interpretation, but also the confidence of site owners, regulators and nearby communities.
Many programs assume laboratory outputs remain decision-ready once they move through accredited workflows and automated quality control (QC) checks. In practice, data packages can still carry hidden bias when system logic, conversions or automated flags drift away from method intent.
On an ongoing Superfund site program, Onterris provided data validation and full-cycle quality assurance oversight to support decisions tied to community health and long-term remedy performance. In a Superfund setting, that level of scrutiny matters because even narrow data issues can affect how contamination is characterized and how remediation decisions are prioritized. The assignment required more than checklist completion. It required experienced chemists who could challenge discrepancies, test assumptions and slow down when results did not align with the chemistry or the method.
A serial dilution check that did not reconcile
The laboratory analyzed soil samples for metals using SW-846 Method 6020B. The method includes multiple QC elements intended to confirm accuracy and precision. One key element, the serial dilution, evaluates potential matrix interference by comparing a native sample concentration to the concentration after a five-fold dilution. Reviewers calculate the percent difference between native and diluted results and compare it to criteria defined by the method or project documents such as the Quality Assurance Project Plan (QAPP) and Sampling and Analysis Plan (SAP). For Superfund programs, QC checks like this help ensure the resulting data are robust enough to support remediation planning and related risk-based decisions.
During review of a serial dilution summary form, an Onterris chemist flagged results that did not add up. Native sample values shown on the summary form did not match what the broader package supported. The diluted results also carried “U” and “J” qualifiers that did not fit the reported concentrations. While only two qualifiers were incorrectly applied in this instance, those kinds of discrepancies can become significant when data is being used to support larger remediation decisions.
The laboratory applied those qualifiers through the Laboratory Information Management System (LIMS) based on conversions tied to aqueous method detection limit (MDL) and reporting limit (RL) thresholds. The same pathway only calculated percent difference when results appeared unqualified and above a threshold tied to the aqueous MDL.
That logic created a practical risk. The project QAPP required validator qualification when percent difference exceeded 25%. If the LIMS suppressed percent-difference evaluation after applying qualifiers in error, a required qualification could be missed and present decision-makers with an incomplete picture of data quality.
Checklists are important tools, but they cannot replace the professional judgment of our experienced chemists. Onterris’ data validation processes go beyond a checklist to provide a comprehensive evaluation of results based on available information. When a result is unexpected or appears anomalous, we slow down, examine the raw data, and consider potential contributing factors to help identify the source of the issue.
From a discrepancy to a corrected workflow
The chemist verified the concern by reviewing raw data within the Level 4 package for analytes flagged “U” or “J” on the serial dilution form. The raw data review supported the initial read. The qualifier logic on the QC summary did not reflect what the data supported.
Based on these findings, Onterris escalated the issue through formal quality channels. The project manager issued a corrective action request and asked the laboratory to complete a root-cause investigation. The laboratory traced the behavior to a prior LIMS modification intended to simplify evaluation. In this case, the laboratory’s management of change review process did not account for downstream impacts on QC summary calculations for all matrices. That kind of review is especially important on complex Superfund programs, where data may pass through multiple systems and still require expert interpretation before it can be relied on for decision-making.
The investigation identified a removed soil preparation factor, a 20× dilution, within the calculation pathway for QC summary forms. That removal inflated the converted MDL and RL by 20× for serial dilution evaluation and triggered incorrect “U” and “J” qualifiers. Once those qualifiers appeared, the LIMS suppressed the percent-difference calculation for a subset of results.
Why this matters for Superfund decision quality
For this client, the confirmed impact was limited. The larger value-add came from eliminating a quiet failure mode before it repeated across future batches or other projects. On high-profile remediation programs such as Superfund sites, early identification and correction of issues helps preserve confidence in the data that underpins cleanup decisions.
The work reinforced a reality for Superfund site owners and program managers: accreditation and automation support quality, but they do not replace independent validation that tests whether the outputs still behave the way the method intends. As the number of active and legacy Superfund sites across the US continues to demand long-term oversight, that distinction remains critical.
The program continues with strengthened validation rigor and clearer traceability between raw data and QC summaries. The approach creates a repeatable model for identifying systemic calculation errors early, protecting remedy decisions and supporting defensible documentation when scrutiny is highest.