The Water Came First

▶️ Prefer to watch? Every story’s a short video → https://www.youtube.com/@GetSafetyIntel · 📣 Got a near-miss? Share it anonymously (QR + link at the bottom).

One year ago today, a man went to work at a school construction site and did not come home. He was in a trench when the water came in — and the ground he’d been standing in all morning turned on him in seconds. Today we’re on excavations, the most predictable killer we have, and why “it was standing fine” is the last thing a lot of good people ever thought.

On the morning of Tuesday, July 22, 2025, a crew was building a new school — Deer Park Elementary, on Lyons Mills Road in Owings Mills, just northwest of Baltimore. Around 8:30, the call went out for a confined-space rescue. A worker was down in the trench.

His name was Luis Medrano. He was 50 years old, a concrete worker, digging a hole down in the excavation. According to police, water began filling the area he was working in. Then a section of the wall gave way and dropped a large amount of dirt on top of him. His coworkers didn’t wait — they got down there and pulled him out with their hands. Firefighters found him critically injured. He was pronounced dead after they got him free. Maryland Occupational Safety and Health and federal OSHA responded to look at what happened.

Here’s what I want you to sit with. The wall that killed Luis had almost certainly been standing “fine” for hours. Trenches don’t creak or groan a warning like a bad ladder. They look solid right up until the face lets go, and by then you have less than a second to do anything about it — which is no time at all. And notice the order of events the police described: the water came first. That’s not a footnote. Water is the single biggest thing that turns a wall that’s holding into a wall that’s coming down. It washes out the cohesion holding the soil face together and it makes every shovelful heavier. A trench that was stable dry can go unstable the moment groundwater or a struck line starts filling the bottom.

29 CFR 1926, Subpart P — Excavations. This is the one written across a hundred years of the same funeral. The core of it, 1926.652(a): anybody working in a trench 5 feet or deeper has to be protected from a cave-in — you slope the walls back, you shore them, or you drop in a trench box. Period. (Shallower than five feet too, if a competent person looks and sees a cave-in risk.) And the part that speaks straight to this day, 1926.651(h): you don’t let people work in an excavation where water is accumulating unless the hazard’s been controlled — and once water shows up, a competent person has to re-evaluate before anyone goes back in. Add 1926.651(c): a ladder or ramp within 25 feet of every worker in a trench 4 feet or deeper. And 1926.651(k): a competent person inspects that trench daily, and again after any change — rain, water, vibration, anything.

A cubic yard of dirt — a box about a yard on each side — weighs somewhere around 3,000 pounds. A small car. That’s what comes off the wall and lands on your chest, and it doesn’t even have to bury your head to kill you: enough weight on your ribs and you simply can’t pull a breath. That’s why “I’m only gonna be in there a minute” is the sentence that shows up in so many of these reports. The hole doesn’t care how long you planned to be in it. So the rule on the floor is simple and non-negotiable: five feet, you’re protected before you’re in it — slope, shore, or box, in the hole before your boots are. If water starts coming in, that’s not “keep working, we’ll pump it” — that’s everybody out, and nobody back in until the competent person says the ground’s been re-checked. The person closest to that wall sees the water first. That’s who has to be allowed to call it.

(E) Environmental — this one’s honest and short: there’s no verified chemical release here, no plume, no spill to a creek. The “environment” that killed Luis was the ground itself and the water moving through it. That’s the whole point of putting E next to a cave-in — saturated soil is a different, heavier, weaker material than the dry face you looked at an hour ago, and groundwater is the variable that flips it. (Worth remembering on any dig: excavations are also where crews strike gas, sewer, and water mains — that’s the day a trench does become a release — so utility locates aren’t a formality either.) (H) Health — the acute toll is a crush and traumatic asphyxia, and it can happen in a partial burial that leaves a man’s head in the open air. And read the health hazard in the rescue: co-workers went into an already-failed, water-filled trench to save him. That’s the most human thing on earth and it’s exactly how one death becomes two — secondary collapses are why trench rescue is a specialized job. (Tie it together) One unprotected, water-filled trench was a safety failure, a health catastrophe, and a reminder that the ground and the water in it are the hazard — three faces of one hole nobody should have been standing in.

If the story above feels like history, the enforcement is very much present tense. OSHA’s National Emphasis Program on trenching and excavation is active — it tells compliance officers to stop and inspect any open excavation they see on any jobsite, complaint or not, because the cave-in death rate hasn’t come down the way everything else has. And the agency treats these as willful, because the fix is cheap and universally known. The most recent big one: on April 1, 2026, the Department of Labor cited a Massachusetts contractor, Revoli Construction, with 7 willful, 33 repeat, and 17 serious violations and roughly $4.7 million in proposed penalties after a November 2025 cave-in killed one worker and hurt another in Yarmouth. The named failures read like the standard’s table of contents: no safe way out of the trench, no adequate cave-in protection, unsupported utilities, spoil piled within two feet of the edge, shoring not installed to the design. So what for you: there is no “we didn’t know” in a trench. If your crews open ground this summer, assume an inspector will look, and make sure the box or the slope is in before the person is.

The biggest process-safety story of the year is still open. On May 26, 2026, a roughly 900,000-gallon tank of white liquor — hot, caustic sodium hydroxide and sodium sulfide — failed at the Nippon Dynawave packaging mill in Longview, Washington, killing 11 workers and sending contamination into the Columbia River, with EPA and Washington’s Department of Ecology on the water side and the Chemical Safety Board investigating the cause. No final report yet, and I won’t guess at root cause before the CSB does. But the shape of it is already a lesson: a large tank of destructive liquid, workers nearby, and a waterway next door — the same three ingredients that turn a mechanical failure into a mass-casualty and an environmental event at once. So what for you: if you store big volumes of caustic or hazardous liquid, “when did we last really inspect that tank, and who’s standing near it” is a question worth asking before the CSB asks it for you.

A two-man crew is putting in a short water service — maybe eight feet down, a ten-minute connection. The trench box is back at the yard; grabbing it means a truck, a second machine, and an hour nobody feels like they have. The ground’s dry, the walls look clean, and the older hand has dug a hundred of these. He steps down to make the tie-in. He’s maybe three minutes in when a slab peels off the wall behind him — not the whole face, just a slough — and knocks him into the far wall, pins one leg. He digs himself out. Walks away with a bad bruise and a story.

Nobody here was reckless. This is what “normal” looks like right before it isn’t — a short task, a familiar hole, real pressure to keep moving, and a piece of protective gear that lives one inconvenient trip away. The fix isn’t “be more careful next time,” because he was a careful, experienced guy. The fix is making the safe way the easy way: the box rides with the crew so it isn’t a decision, “five feet means protected” is a hard line no supervisor can wave off, and the guy who says “let’s just grab the box” is thanked, not eye-rolled. Take the outcome out of it — that slough could’ve come off two feet higher and buried his chest. Same decision, same trench, a funeral instead of a bruise. We don’t get to grade the choice by how the dirt happened to fall.

Got a fail or a near-miss? Hit reply, or scan the QR below — we’ll feature it anonymously — no names, no company, no blame. Just the lesson, so the next crew doesn’t learn it the hard way.

Before anybody’s boots go below their waist in a trench today, stop and answer three questions out loud, as a crew: How deep is it? (Five feet or more, you’re protected — slope, shore, or box — and it’s in the hole before the person is.) Where’s the way out? (A ladder within 25 feet.) Is there water? (If it’s coming in, nobody enters until a competent person re-checks the ground — the water is the warning.) Thirty seconds of talking. That’s the whole difference between this being a normal Tuesday and being the day somebody’s crew has to dig them out by hand.

🎬 Rather watch than read? Every Safety Intel story is a short video on YouTube → https://www.youtube.com/@GetSafetyIntel — subscribe so you never miss one. 📸 Share YOUR near-miss (100% anonymous): https://scantoconfess.com — or just scan the QR above. 📤 Forward this to a crew member who needs it today.

Please stay Safe & Hydrated!!!