Who's measuring the air?
Louisiana has a system for monitoring air quality near its industrial corridor. Depending on who you ask, it's either rigorous or riddled with gaps.
Louisiana has no shortage of industrial facilities. It has a shortage of ways to know what they're putting into the air.
That gap sits at the center of a debate playing out in the state Legislature—one that is less about pollution itself and more about who measures it, how, and whether the public has any right to know in real time.
Why it matters: The Capital Region sits inside one of the most concentrated industrial corridors in the country. Ascension, Iberville and West Baton Rouge parishes alone host dozens of refineries and chemical complexes. What gets measured—and what doesn't—shapes how residents, regulators and companies understand risk.
How monitoring works now
Air quality in Louisiana is tracked in three ways.
- The Louisiana Department of Environmental Quality operates a network of regulatory-grade monitors. The data is accurate, standardized and legally defensible. It is also limited—designed to measure regional air quality indicators, such as ozone and fine particulates, rather than localized toxic releases from specific facilities.
- Industrial facilities report their own emissions, based on permits, modeling and internal monitoring systems. That data feeds state and federal regulators. The companies themselves generate it.
- Community groups and residents have begun deploying lower-cost sensors to fill gaps near industrial sites. That data is more flexible and localized. It is also less precise and carries no regulatory weight.
- Not all air data is treated equally. That is the whole argument.
Legislature ignores itself: In February 2025, a state task force led by then-DEQ Secretary Aurelia Giacometto recommended the state prioritize real-time fenceline monitoring at its largest industrial facilities—those nearest population centers with the highest total emissions and toxicity.
- State Sen. Royce Duplessis of New Orleans introduced SB 356 to do exactly that. The bill would have required 117 oil refineries and chemical complexes to install monitors tracking up to 17 toxic pollutants in real time.
- Companies would have covered the cost, estimated at $250,000 to $630,000 upfront per facility, plus roughly $25,000 annually in maintenance. No state dollars involved.
What happened? It died in the Senate Environmental Quality Committee without even a vote.
- The Louisiana Chemical Association, the Louisiana Mid-Continent Oil and Gas Association and LABI opposed it. The bill was already a significant retreat from earlier proposals that would have covered 476 facilities and 235 pollutants.
The case against: Industry's core argument deserves a fair hearing.
- Fenceline monitors detect chemicals. They do not confirm sources. A monitor at a plant boundary picks up emissions from trains, barges, trucks and neighboring facilities just as readily as from the plant itself. An alert built on that data, the Louisiana Chemistry Association argued, cannot reliably tell the public what it claims to.
- That is not a frivolous concern. Misidentified pollution sources create real problems—for enforcement, for public trust and for companies wrongly implicated. Senate Environmental Quality Committee Chairman Eddie Lambert raised the same point: inaccurate data, he argued, could cause harm.
- Industry also notes that internal plant monitors already exist and are better positioned to detect leaks and feed into existing regulatory systems.
The case for: The counterargument starts with a simple observation: existing monitoring wasn't designed to protect the people living closest to the plants.
- LDEQ's network is regionally focused. Plant monitors are designed for worker safety, not public notification. When something goes wrong, residents near industrial facilities are often among the last to know.
- In September 2023, a tank farm fire at Marathon Petroleum's Garyville refinery sent smoke across the region. In April 2022, a chlorine leak at a Plaquemine-area chemical complex—later identified as Blue Cube, an Olin Chemical subsidiary—sent more than 30 people to the hospital. Arriving agencies initially found no off-site impact. Weeks later, state and parish officials accused the company of downplaying the extent of the leak. Blue Cube eventually settled with DEQ for $600,000. The findings were kept confidential.
- Similar open-path ultraviolet monitoring systems are in use at facilities in California and Colorado and have been recommended by the EPA. When California regulators were asked about false positives, they reported the monitors had helped refineries prove they weren't the source of complaints.
Where each side falls short:
- Industry's data-quality argument is legitimate. It is also incomplete. The same standards used to question community and fenceline data apply, at least in part, to self-reported industry emissions, which are modeled, permitted and generated internally. Precision is a reasonable standard. Applying it selectively is a different matter.
- The pro-monitoring argument is compelling on coverage. It is less compelling regarding what happens when imperfect data drives public alarm before a source is confirmed. That scenario is not hypothetical.
The bottom line: A state task force studied this problem and recommended a path forward. The Legislature had a scaled-back version of that recommendation before it and declined to act.
- What remains is a monitoring system built around regional air quality, industry self-reporting and a patchwork of community sensors that carry no official weight—in a region with one of the highest concentrations of industrial facilities in the country.
- The most straightforward solution—more regulatory-grade monitors operated by LDEQ—is rarely discussed. Independent, continuous, publicly accessible data near specific facilities would create an accountability environment that the current system, built around self-reporting and limited state coverage, does not. Why? The short answer: cost, control and politics.
The debate will return. It always does.