Have you ever watched a weather alert scroll across your phone and assumed it was just another noisy warning? I did, until I read the wording used in southern New Mexico this week. It was not polite. It was not hedged. It told people below a modest flood-control dam to seek higher ground now. That kind of sentence does not appear because a creek looks a little high. It appears when operators think a structure might give way in hours, not days.
What Almost Happened Below McLeod Dam
The dam sits near Garfield in Doña Ana County, upstream of small valley towns that most travelers pass without slowing down. Hatch. Salem. Rincon. Placitas. Places built around irrigation, chile fields, and a river corridor that looks tame until it is not. On Tuesday evening, remnant moisture from a powerful eastern Pacific hurricane dumped enough water onto already wet ground that the reservoir rose faster than the outlet works could handle.
Forecasters issued a flash flood emergency. County officials ordered evacuations. One account described water pouring over a roughly twenty-foot earthen embankment. Another, later in the night, sounded more cautious: crews said they had contained a very small piping failure, that no uncontrolled release was leaving the dam, and that pumps were coming in to take pressure off the structure. Both versions can be true in the same event. Overtopping and internal erosion are how earthen dams actually fail. They rarely explode like movie scenery. They seep, then cut a channel, then unravel.
This is a particularly dangerous situation. Seek higher ground now.
By late Tuesday, most evacuation orders were lifted. Rincon stayed under restriction longer because floodwater was still standing in the community. That is the unglamorous aftermath people forget. Even when a dam holds, the water that almost broke it has to go somewhere. Streets, fields, and low crossings take the hit.
Why A “Poor” High-Hazard Dam Matters
Here is the part that should bother anyone who lives downstream of a similar structure. The dam’s last public inspection, in 2023, rated the condition poor. It was also classed as high-hazard. Those two labels do not mean the same thing, and mixing them up is how public statements get sloppy.
High-hazard is about consequences. It means a failure could kill people or wipe out critical infrastructure. It is not a grade on concrete or soil. Poor is the grade on the structure itself: seepage, sloughing, outdated spillway capacity, missing instrumentation, deferred maintenance. In my experience covering infrastructure scares, officials love to talk about the first label because it sounds serious and official. They get quieter about the second, because it implies someone left work undone.
Reports also said there was no emergency action plan on file at the time of that inspection. That is not a paperwork nitpick. An emergency action plan is the difference between a coordinated siren-and-door-knock response and a scramble on social media while water is already moving. Perhaps the most interesting aspect is what was missing from county comments this week: a clear public account of what has been repaired, re-inspected, or planned since 2023. Silence is not proof that nothing was done. It is proof that the public still cannot tell.
- High-hazard describes the human cost if the dam fails.
- Poor condition describes the structure’s current health.
- An emergency action plan tells people who calls whom, and how fast.
- Overtopping and piping are the usual failure paths on earthen dams.
The Storm That Pushed The Reservoir Over The Edge
The water did not appear from a gentle local shower. It arrived with the leftovers of Hurricane Polo, ranked among the strongest eastern Pacific hurricanes on record. Behind it, another named storm was already organizing off Mexico. Forecasters have been tying that busy Pacific season to a strengthening El Niño. The Atlantic, by contrast, had one of its quietest starts in decades as of early September. That split is classic. Warm water in the equatorial Pacific tends to load the dice for the eastern Pacific and suppress some Atlantic development, at least in the heart of the season.
I keep coming back to a simple image. A desert valley dam is designed for cloudbursts and monsoon pulses, not for stacked tropical remnants riding a wetter-than-normal pattern. Design storms are statistical. Climate patterns shift the odds. When those two ideas collide, you get nights like Tuesday.
El Niño Has Not Even Started Its Main Act
This is the part a lot of headlines bury. The event that nearly overtopped McLeod Dam was a warm-up. The winter phase, when El Niño usually does its heaviest work on the southern United States, has not begun. Climate agencies in their September update put the odds of a very strong event this fall and winter above 90 percent. They also assigned a high chance that the key Niño-3.4 temperature anomaly would exceed +2.5°C in the October-December window. It had already reached +1.8°C in August.
If those numbers hold, this could rank among the strongest El Niño events since modern records began in 1950. Impacts are never guaranteed. Agencies say that clearly, and they should. A strong event raises the likelihood of certain patterns. It does not schedule a flood for a particular Tuesday in Hatch.
For New Mexico, the typical winter signature is wetter and cooler, with the largest precipitation increases often focused on the south. That includes the Hatch Valley. Next door, California’s governor already declared a state of emergency in late September to prepare for heavy rain, debris flows, and flooding. You can argue about politics all day. The declaration itself is an admission that agencies do not want to be caught reacting after the first atmospheric river hits burned slopes.
A strong El Niño makes certain outcomes more likely. It does not write the daily forecast.
How Earthen Dams Actually Fail
If you only remember movie dams, you picture a wall of concrete snapping in one shot. Most of the structures that protect small American towns are compacted earth. They fail in slower, uglier ways.
Overtopping happens when the reservoir rises above the crest and water runs down the back slope. Flowing water cuts rills, then a notch, then a widening breach. Once that notch forms, the reservoir starts emptying itself through the gap. Speed depends on how much water is stored and how erodible the fill is.
Piping is sneakier. Water finds a path through the fill or along a conduit. It carries soil grains with it. The path enlarges from the inside. From the downstream face you might see a dirty boil or a wet spot that will not dry. Crews talking about a small piping failure are talking about that process starting, then being interrupted with sandbags, filters, or pumps.
Neither failure mode cares about press conferences. Both care about freeboard, spillway capacity, compaction, and whether anyone is watching instruments at 8:30 p.m. on a Tuesday.
| Failure path | What you see first | Why it gets worse fast |
| Overtopping | Water over the crest | Erosion of the downstream slope |
| Piping | Seepage, boils, dirty water | Internal void grows toward a collapse |
| Slope slump | Cracks or a dropping crest | Reduced freeboard and blocked drainage |
| Spillway damage | Scour at the outlet | Less controlled release capacity |
The Human Geography Of A Small Valley Flood
Hatch is famous for chile, not for hydrology. That fame is part of the risk. Farm towns sit on the best soil, and the best soil is often the floodplain. Irrigation ditches, highway dips, and rail embankments all change how water moves once it leaves a reservoir. A dam that “almost” fails can still ruin a harvest, close a school road, and trap people who thought they were a safe mile away.
I’ve found that the public conversation after these scares splits in two unhelpful directions. One side says the warning was overblown because the dam held. The other side says collapse was certain and officials hid it. Reality is duller. Operators saw a structure under stress. They called for people to move. Mitigation bought time. Standing water remained. That sequence is success and warning at once.
- Moisture arrives faster than the spillway can pass it.
- Freeboard shrinks and seepage shows on the face.
- Emergency alerts go to towns in the hydraulic shadow.
- Crews fight piping or overtopping through the night.
- Orders lift, but residual flooding still blocks roads.
Why The Winter Outlook Should Change Local Priorities
If southern New Mexico is entering a wetter winter, the to-do list is not mysterious. Inspect high-hazard dams again, not on the old calendar. Confirm that emergency action plans exist, are current, and have been exercised. Clear debris from spillways. Check earthen slopes after every major pulse. Talk to people in Rincon and Garfield in plain language, not in inventory jargon.
California’s early emergency declaration is a clue about institutional memory. Debris flows after fire are not theoretical there. New Mexico’s monsoon and winter storm combination is different, but the principle is the same. You prepare on the dry side of the season. You do not wait for the second overtopping.
Is every poor-rated dam going to fail this winter? Of course not. That would be a lazy claim. The sharper claim is this: a record-class El Niño raises the number of tests those dams will face, and a poor rating means some of them will sit for those tests with less margin than residents assume.
What Residents Downstream Can Do Without Panic
Panic is useless. A short checklist is not. Know whether you live in an inundation zone. If your county has never published one, ask why. Keep a way to receive wireless alerts that does not depend on a single cell tower. Decide now where higher ground actually is. After a warning, do not drive through muddy dips to “have a look.” That sentence has ended more nights than dam collapses have.
If you own land near an irrigation structure or a small flood-control dam, watch for new wet spots on the downstream face after heavy rain. Take a photo with a timestamp. Call it in. You are not being dramatic. You are doing the job inspectors cannot do at 2 a.m.
Quick local readiness: Know the inundation map Keep an offline alert path Identify real higher ground Never drive a flooded dip Report new seepage immediately
The Wider Pattern Across The Southern Tier
New Mexico is not an isolated case study. A strong El Niño tends to favor a wetter southern tier from California through the Gulf states, with details that shift year to year. Aging flood infrastructure is not unique to one county. Thousands of dams in the national inventory carry high-hazard labels because towns grew below them after construction. Many last inspections are older than residents think.
That inventory problem is slow-motion policy. It does not trend until a night like Tuesday. Then everyone wants a villain. Sometimes the villain is deferred maintenance. Sometimes it is a storm that exceeded the design hydrograph. Sometimes it is both, standing in the same puddle.
I do not think the lesson is that every winter storm will be historic. The lesson is that the margin for error on small earthen dams is thinner than the language on tourism brochures. When climate agencies say a very strong El Niño is likely, they are telling risk managers to spend money before the water arrives. Whether that money gets spent is a local political choice, not a satellite product.
Separating Forecast Odds From Local Headlines
Climate outlooks speak in probabilities. Local emergencies speak in minutes. Mixing those time scales creates sloppy stories. A 75 percent chance that a temperature anomaly exceeds a threshold is not the same thing as a guarantee that McLeod Dam will be overtopped in January. It is a reason to treat the next six months as a stress test for drainage, dams, and road dips that already showed weakness.
Forecasters also warned that impacts from an event this size are more likely but not locked in. That caveat is not timid. It is honest. Weather still has to set up the right storm track. Soils still have to be wet. A dam still has to be one pulse away from losing freeboard. Tuesday showed that last condition is not hypothetical in the Hatch Valley.
The dam that came within hours of failing will face winter carrying a poor rating unless the public is shown what changed.
What Accountability Should Look Like After The Water Recedes
Accountability is not a pile-on. It is a short list of public answers. What was repaired after the 2023 inspection? Who signed off? Is there an emergency action plan now, and has it been drilled? What is the current freeboard under normal pool? What is the spillway rating relative to a tropical-remnant hydrograph, not just a local monsoon burst?
Those questions are fair in any county, not only Doña Ana. If officials can answer them, the next warning will sound like competence. If they cannot, residents are being asked to trust a structure on reputation alone. Reputation does not hold an earthen slope together.
There is also a quieter form of accountability among residents. Share official alerts, not rumor maps. Do not treat an all-clear as permission to mock the people who left their homes at night. They left because a forecaster used the word emergency. That word is supposed to be expensive. It should stay expensive.
A Longer View Of Water In A Dry State
New Mexico’s public identity is drought, dust, and irrigation fights. Floods feel like a plot twist. They are not. Dry regions often fail in floods because channels are small, soils crust, and towns hug the only reliable water. A winter that turns wet does not make the desert lush in a gentle way. It slams water through arroyos that spent years filling with debris.
That is why a near miss at a flood-control dam is a preview, not a curiosity. The same soils that bake in June can pipe in October. The same farmers who pray for rain in April can lose a field to a reservoir they never think about until the sirens start.
In my view, the grown-up response is unfashionable. Fund inspections. Publish inundation maps. Practice the plan. Keep the language sharp when the water is high and humble when the structure holds. Then do the boring work before the next Pacific remnant finds the same valley.
The Night The Warning Was The Story
By morning, most people in those five communities were back in their kitchens arguing about whether the scare was overstated. That argument misses the point. The warning itself was the news. A flood-control dam with a poor condition rating was pushed to the edge by tropical moisture on the front end of a potentially historic El Niño. Crews say they stopped a small piping problem. Water still stood in Rincon. Orders lifted for everyone else.
Winter has not clocked in yet. If the outlook is even half right, this will not be the last night someone in the southern tier is told to move to higher ground. The question is whether the next structure in the queue has had its homework done. Tuesday was not the main event. It was the syllabus.
And if you live below an earthen dam you cannot name, tonight would be a decent time to learn its name, its last inspection grade, and the road you would actually take if the phone lit up again. That is not alarmism. That is how adults treat a warning that already had to be shouted once.
