DES scientist outlines how Saint-Gobain emissions drove local PFAS deposition and groundwater contamination

Jeff Martz, administrator of the Hazardous Waste Remediation Bureau at the New Hampshire Department of Environmental Services, told the House study commission that airborne emissions from the Saint‑Gobain manufacturing facility in Merrimack drove widespread deposition of PFOA and other PFAS and are a key pathway for the contamination now found in groundwater across a large area.

Martz summarized the site history and transport pathways and said the “main pathway that we’re concerned about… is the air emissions pathway,” adding that PFOA and other PFAS “were essentially driven off the fabric and emitted through a series of smokestacks” and later deposited to soil where they could leach to groundwater.

Why it matters: the 2018 consent decree around Saint‑Gobain defines a roughly 64‑square‑mile investigation area that includes thousands of public and private supply wells; the commission and DES are relying on combined air‑deposition models, soil data and groundwater sampling to prioritize further testing and remediation.

Martz described several release mechanisms at the plant: particulate and vapor stack emissions, discrete spills of chemical dispersions, and leaks to sanitary and storm sewers. He said those different pathways and changing product formulations produced “commingled plumes” with varying PFAS “fingerprints” on and off the property, so nearby wells can show different compound profiles and concentrations.

He reviewed modeling and monitoring history: early groundwater sampling detected PFAS in 2016; in 2016 EPA issued a 70 parts‑per‑trillion (ppt) lifetime health advisory for combined PFOA and PFOS; New Hampshire adopted enforceable groundwater standards for PFOA and related PFAS after litigation delayed implementation until July 2020; and state PFAS soil remediation standards were adopted in rule in December 2024.

Martz showed a modeled air‑deposition map for a multiyear domain and explained the model’s color ramp for PFOA concentrations in groundwater: green for 0–6 ppt, yellow for 6–9 ppt, orange for 9–12 ppt, red for 12–70 ppt (above the state HQS for PFOA), and purple for over 70 ppt. He cautioned that the deposition maps are meant to show relative patterns not precise absolute masses, because key inputs—especially particle‑size distribution—were never directly measured at the Saint‑Gobain stacks.

Commission members pressed on the historical emission estimates that Martz showed: he reported an estimated emission history provided to DES that begins near 100 pounds per year in 1986, peaks at roughly 900 pounds per year around the early 2000s, and drops to an estimated 12 pounds per year of PFOA after a 2006–2007 reformulation. Several members, including a long‑time commissioner, challenged the company‑reported decline as unreliable. One commissioner said the company’s discovery in other litigation showed the earlier reports were “not the truth” and called them “an absolute lie.”

Martz acknowledged the uncertainty in the inputs and repeated that the department treats the air models as “a general guide, for a pattern as opposed to an absolute value.” He said model output guided targeted groundwater sampling and that areas of higher modeled deposition often correspond to higher groundwater concentrations, but that soil and groundwater sampling remain essential for quantifying local conditions.

On distinguishable fingerprints, Martz said lower‑concentration, off‑site groundwater samples often show a PFOA‑dominant mix consistent with Saint‑Gobain’s historical emissions, while on‑site or very near‑site monitoring wells can show different compound mixes that reflect the company’s evolving formulations.

Commissioners and residents asked whether newer replacement chemistries such as GenX (HFPO‑DA) show up in groundwater or surface water; Martz said laboratory detections of GenX in on‑site monitoring wells have been low and that some non‑targeted analyses suggest GenX may transform in the environment into other products, which complicates simple source tracing.

Martz and other DES staff encouraged residents to consult the department’s public PFAS viewer for available surface‑water and groundwater sample results, and he said DES continues to combine modeling, soil testing and expanded well sampling to refine the site conceptual model.

Ending: the presentation left open two central data needs the commission repeatedly raised—better validation of historical emission estimates and expanded soil sampling across the consent‑decree area—while DES said those needs will be addressed through additional sampling plans, ongoing monitoring and future regulatory steps.