Chemical accidents rise as Trump administration proposes weakening safety rules 45%

By Liza Gross0% Inside Climate News47%

7/5/2026, 11:05:22 AM

BS Summary: This article contains 32 faulty reasoning types, including Framing Effect, Appeal to Emotion, and Confirmation Bias, with Appeal to Authority as the most egregious example at 22.7% saturation with 224 hits. Analysis detected 1,456 faulty-reasoning hits from 986 analyzed words, generating a BS Score of 47.7% and a BS Rank of 45% (11,427 of 20,472 articles). This article is better (less manipulative) than 55.80% of the article peer group.

Physicist Ronald Koopman appeared at a Southern California Air District meeting in 2018 to talk about what seemed like an arcane scientific topic: hydrofluoric acid dispersion and water mitigation testing. 
Hydrofluoric acid, also known as hydrogen fluoride or HF, is used to manufacture a range of materials, including refrigerants, gasoline, fluorine-based pesticides and fluoropolymers like those used to make Teflon. 
It’s also one of the most corrosive and dangerous chemicals known. 
Koopman conducted experiments with the chemical in the 1980s that warned about the potential of deadly accidents at facilities that use the hazardous materials. 
With the Trump administration poised to roll back rules intended to protect workers and communities from catastrophic industrial chemical releases, and a new analysis showing rising rates of chemical accidents, Koopman’s presentation on highly hazardous materials has taken on a new urgency. 
The number of accidents involving releases of dangerous chemicals rose by 57 percent between 2021 and 2025, from 83 to 131, according to an analysis released Monday by Public Employees for Environmental Responsibility, a nonprofit that works with former government officials. 
Injuries or deaths from accidents also rose, from 60 to 89 over the same five-year period, the analysis found. 
Incident reports released by the Chemical Safety Board (CSB), an independent agency that investigates chemical accidents, show that more than 650 accidents occurred between April 2020 and May 2026, with 103 resulting in fatalities, 355 causing injuries and 314 doing “substantial property damage.” 
Close to 150 million people live within 3 miles of these facilities. 
Historically underserved and overburdened populations, including people who identify as Black and Latino, are at greatest risk of exposure to an accidental release. 
Many refineries were built before 1985, the analysis notes. 
“With each passing year the risk gets greater because the infrastructure continues to age,” said Jeff Ruch, senior counsel at PEER. 
The 1980s HF experiments were run by Koopman, who now runs Hazard Analysis Consulting, on behalf of the oil company Amoco (later acquired by BP) to understand how the highly toxic refinery chemical would behave in a spill. 
The test was a “spectacular success” in demonstrating what could happen and how serious the problem might be, Koopman said at the air district meeting. 
When they released 1,000 gallons of the noxious chemical, they expected it to pool on the ground and emit a small quantity of gas. 
Instead, a billowing “ground-hugging” mist formed, allowing the deadly gas to travel miles downwind, considerably farther than anyone thought possible. 
Years later, after a series of fiery explosions at a hydrofluoric-acid unit at the Philadelphia Energy Solutions refinery rocked the surrounding South Philadelphia neighborhood in 2019, Koopman told NPR, “it’s just unconscionable” to allow people to live so close to these refineries. 
The accident released more than 5,000 pounds of the chemical. 
The neighboring mostly Black and brown South Philadelphia neighborhood was spared thanks to “favorable wind conditions,” the CSB said. 
“We had tried and failed to induce EPA to phase out hydrogen fluoride at these refineries,” said Ruch. 
“The refineries are near population centers, and the release of the gas could be just a horrible tragedy.” 
Exposure to 170 parts per million of hydrogen fluoride for 10 minutes can cause death or serious injury. 
After the massive Philadelphia refinery explosion, PEER petitioned the EPA to ban hydrogen fluoride in 2019. 
The agency refused to consider the petition. 
Close to 50 refineries use hydrogen fluoride and have reported more than 200 accidents resulting in serious injuries and deaths to the EPA over the past 25 years, according to the nonprofit Public Health Watch. 
The refineries represent just a fraction of the 12,000 facilities that use certain hazardous substances and are regulated by the EPA’s Risk Management Program under the Clean Air Act. 
The new statistics released by PEER were made public as a result of a lawsuit PEER and other groups filed to compel the Chemical Safety Board to disclose industrial chemical releases as required by the Clean Air Act. 
A federal judge ruled in 2019 that communities have a right to know what hazardous chemicals are released nearby. 
Yet Trump’s EPA removed a public data tool designed to inform communities of nearby risks last year. 
President Trump has also tried to eliminate the Chemical Safety Board by withholding funding, though Congress has continued to fund the agency. 
Earlier this year, the administration proposed to significantly weaken RMP rules finalized in 2024 “to reduce regulatory burden” and accepted public comment on the rules until early May. 
The Biden administration’s strengthened RMP rules require a number of measures to reduce the risk of catastrophic accidents, including safer-alternatives analyses, independent analyses of accidents’ root causes, worker participation in accident-prevention plans and preparations to adapt to climate change. 
An EPA spokesperson said the agency is reviewing public comments and continues to work toward completing the final rule in late 2026. 
“EPA’s proposal relies on a rigorous analysis of RMP reportable incidents between 2014 and 2023, which shows accidental releases unequivocally declined significantly over that period,” the spokesperson said. 
“This means that RMP-regulated facilities had successful prevention programs in place before the Biden EPA finalized its nonsensical and burdensome 2024 rule.” 
The Biden EPA used the same data and came to the opposite conclusion, said PEER’s Ruch. 
Plus, he added, “the conclusion that any decline is due to industry prevention plans is a supposition which the current EPA does not have the data to support.” 
Meanwhile, chemical accidents resulting in evacuations, injuries, or multiple casualties continue to happen at least once a week. 
“With each passing year the risk gets greater because the infrastructure continues to age,” Ruch said. 
At the same time, he added, “the federal response to it is shrinking.” 
This article originally appeared on Inside Climate News, a nonprofit, non-partisan news organization that covers climate, energy, and the environment. 
Sign up for their newsletter here. 
Article reasoning-pattern comparisonThis article: 9.3%Liza Gross: 4.7%Ars Technica: 2.7%Confirmation Bias9.3%This article: 0.0%Liza Gross: 2.7%Ars Technica: 1.2%Anchoring Bias0.0%This article: 1.9%Liza Gross: 5.2%Ars Technica: 2.4%Availability Heuristic1.9%This article: 2.9%Liza Gross: 0.7%Ars Technica: 1.0%Representativeness Heuristic2.9%This article: 2.4%Liza Gross: 1.1%Ars Technica: 0.6%Hindsight Bias2.4%This article: 5.7%Liza Gross: 2.5%Ars Technica: 2.0%Overconfidence Bias5.7%This article: 12.1%Liza Gross: 10.5%Ars Technica: 3.5%Framing Effect12.1%This article: 1.8%Liza Gross: 0.2%Ars Technica: 0.5%Loss Aversion1.8%This article: 1.9%Liza Gross: 0.2%Ars Technica: 0.5%Status Quo Bias1.9%This article: 0.0%Liza Gross: 0.0%Ars Technica: 0.1%Sunk Cost Effect0.0%This article: 2.5%Liza Gross: 1.1%Ars Technica: 4.3%Optimism Bias2.5%This article: 3.8%Liza Gross: 2.1%Ars Technica: 1.7%Pessimism Bias3.8%This article: 7.4%Liza Gross: 8.3%Ars Technica: 6.0%Negativity Bias7.4%This article: 1.8%Liza Gross: 0.7%Ars Technica: 1.1%Self-Serving Bias1.8%This article: 0.0%Liza Gross: 0.0%Ars Technica: 0.6%Fundamental Attribution Error0.0%This article: 0.0%Liza Gross: 0.5%Ars Technica: 0.1%Actor-Observer Bias0.0%This article: 4.3%Liza Gross: 0.8%Ars Technica: 0.6%In-Group Bias4.3%This article: 2.2%Liza Gross: 0.3%Ars Technica: 0.2%Out-Group Homogeneity Bias2.2%This article: 2.0%Liza Gross: 0.3%Ars Technica: 2.1%Halo Effect2.0%This article: 0.0%Liza Gross: 0.0%Ars Technica: 0.1%Horn Effect0.0%This article: 0.0%Liza Gross: 0.0%Ars Technica: 0.0%Dunning-Kruger Effect0.0%This article: 6.1%Liza Gross: 2.5%Ars Technica: 1.0%Recency Bias6.1%This article: 1.7%Liza Gross: 0.6%Ars Technica: 0.3%Primacy Effect1.7%This article: 0.0%Liza Gross: 0.0%Ars Technica: 0.1%Blind-Spot Bias0.0%This article: 2.2%Liza Gross: 0.6%Ars Technica: 0.4%Ad Hominem2.2%This article: 0.0%Liza Gross: 0.0%Ars Technica: 0.2%Straw Man0.0%This article: 22.7%Liza Gross: 7.3%Ars Technica: 4.2%Appeal to Authority22.7%This article: 1.7%Liza Gross: 1.0%Ars Technica: 1.1%False Dilemma1.7%This article: 0.0%Liza Gross: 0.2%Ars Technica: 0.7%Slippery Slope0.0%This article: 0.0%Liza Gross: 0.0%Ars Technica: 0.1%Circular Reasoning0.0%This article: 4.3%Liza Gross: 1.1%Ars Technica: 3.8%Hasty Generalization4.3%This article: 0.0%Liza Gross: 0.0%Ars Technica: 0.2%Red Herring0.0%This article: 0.0%Liza Gross: 0.0%Ars Technica: 0.6%Bandwagon0.0%This article: 10.4%Liza Gross: 8.0%Ars Technica: 2.7%Appeal to Emotion10.4%This article: 2.8%Liza Gross: 0.4%Ars Technica: 0.6%Begging the Question2.8%This article: 7.3%Liza Gross: 3.2%Ars Technica: 2.4%Post Hoc (False Cause)7.3%This article: 2.2%Liza Gross: 0.3%Ars Technica: 0.1%Tu Quoque2.2%This article: 1.8%Liza Gross: 1.3%Ars Technica: 0.4%Burden of Proof1.8%This article: 0.0%Liza Gross: 0.5%Ars Technica: 0.2%Appeal to Nature0.0%This article: 0.0%Liza Gross: 0.0%Ars Technica: 0.2%Composition/Division0.0%This article: 4.3%Liza Gross: 1.4%Ars Technica: 1.5%Anecdotal4.3%This article: 0.0%Liza Gross: 0.0%Ars Technica: 0.1%No True Scotsman0.0%This article: 3.9%Liza Gross: 1.0%Ars Technica: 1.7%Ambiguity (Equivocation)3.9%This article: 0.0%Liza Gross: 0.0%Ars Technica: 0.0%Gambler’s Fallacy0.0%This article: 0.0%Liza Gross: 0.0%Ars Technica: 0.1%Middle Ground0.0%This article: 0.0%Liza Gross: 0.0%Ars Technica: 0.1%Personal Incredulity0.0%This article: 0.0%Liza Gross: 0.3%Ars Technica: 0.1%Special Pleading0.0%This article: 0.0%Liza Gross: 0.0%Ars Technica: 0.1%Genetic Fallacy0.0%This article: 7.1%Liza Gross: 1.5%Ars Technica: 1.5%Unattributed Quote7.1%This article: 4.3%Liza Gross: 1.1%Ars Technica: 1.0%Quote-first Misdirection4.3%This article: 2.0%Liza Gross: 3.3%Ars Technica: 4.2%Biased Writer Voice2.0%This article: 0.0%Liza Gross: 0.0%Ars Technica: 1.0%Indoctrination0.0%This article: 0.0%Liza Gross: 10.6%Ars Technica: 0.6%Politically Left Leaning Bias0.0%This article: 0.0%Liza Gross: 1.1%Ars Technica: 0.2%Politically Right Leaning Bias0.0%This article: 0.6%Liza Gross: 0.2%Ars Technica: 1.3%Attempt to Sell a Product or S…0.6%

986 words analyzed.

Speakers

5speakers36%attributed speech627writer words
Voice mapSelect a segment to jump to its words
Writer's voice • 10 words • 0.0% coverageWriter's voice • 30 words • 0.0% coverageWriter's voice • 30 words • 0.0% coverageWriter's voice • 11 words • 0.0% coverageRonald Koopman • 24 words • 0.0% coverageWriter's voice • 42 words • 0.0% coveragePublic Employees for Environmental Responsibility • 41 words • 0.0% coverageWriter's voice • 19 words • 0.0% coverageChemical Safety Board (CSB) • 43 words • 0.0% coverageWriter's voice • 12 words • 0.0% coverageWriter's voice • 23 words • 0.0% coverageWriter's voice • 9 words • 0.0% coverageJeff Ruch • 21 words • 0.0% coverageWriter's voice • 38 words • 0.0% coverageRonald Koopman • 25 words • 0.0% coverageWriter's voice • 24 words • 0.0% coverageWriter's voice • 20 words • 100.0% coverageRonald Koopman • 42 words • 100.0% coverageWriter's voice • 10 words • 0.0% coverageChemical Safety Board (CSB) • 19 words • 0.0% coverageJeff Ruch • 18 words • 0.0% coverageJeff Ruch • 18 words • 0.0% coverageWriter's voice • 18 words • 0.0% coverageWriter's voice • 16 words • 0.0% coverageWriter's voice • 7 words • 0.0% coveragePublic Health Watch • 35 words • 0.0% coverageWriter's voice • 29 words • 0.0% coverageWriter's voice • 38 words • 0.0% coverageWriter's voice • 19 words • 0.0% coverageWriter's voice • 17 words • 0.0% coverageWriter's voice • 22 words • 0.0% coverageWriter's voice • 28 words • 0.0% coverageWriter's voice • 39 words • 0.0% coverageWriter's voice • 22 words • 0.0% coverageWriter's voice • 28 words • 100.0% coverageWriter's voice • 22 words • 0.0% coverageJeff Ruch • 16 words • 0.0% coverageJeff Ruch • 28 words • 0.0% coverageWriter's voice • 18 words • 0.0% coverageJeff Ruch • 16 words • 0.0% coverageJeff Ruch • 13 words • 0.0% coverageWriter's voice • 20 words • 0.0% coverageWriter's voice • 6 words • 100.0% coverage
Selected voice

Ronald Koopman

100%flagged-word coverage
91 attributed words25% of attributed speech74% writer coverage
0%25.0%50.0%Quote-first Misdirection+46.2 ptsWriter: 0.0%Ronald Koopman: 46.2%46.2%Unattributed Quote+41.7 ptsWriter: 4.5%Ronald Koopman: 46.2%46.2%Biased Writer Voice-3.2 ptsWriter: 3.2%Ronald Koopman: 0.0%0.0%Attempt to Sell a Product -1.0 ptsWriter: 1.0%Ronald Koopman: 0.0%0.0%

Attribution is sentence-level. Pattern percentages are calculated only from words assigned to that voice.

Loading…
Loading…
Loading…
Loading…

Analysis

Hover over highlighted words in the article to view the associated bias or fallacy analysis.