“A fellow who thinks the ‘river of lost souls’ won’t do to watch, has all the feathers out of his willow plume. She may look meek and sleepy-eyed but don’t tickle her heels.”
—Durango Wage Earner, October 12, 1911
At around four a.m. on October 6, 1911, Navajo Methodist Mission Superintendent J.N. Simmons woke up to find himself and the mission near Farmington, New Mexico, surrounded by water. It wasn’t a total surprise. He and two other staffers—Frank B. Tice and Walter Weston—had received the flood alarm the previous day, but had chosen to stay, certain that the San Juan River’s waters would never reach them, and if they did, the brand new, three-story cement-block mission building, watched over by God, would provide an unsinkable refuge. They were wrong.
The rain began in the San Juan Mountains late on the morning of October 4, 1911. It came down gently at first, slowly gaining intensity over the course of the day. By evening the tropical storm was a torrent, dropping two inches of precipitation on Durango in just 12 hours, nearly twice what the town normally gets during all of October. Weather watchers in Gladstone, above Silverton, recorded eight inches of rain on October 5—a virtual high country hurricane1.
Once-gurgling streams jumped from their banks and twisted steel railroad tracks into contorted sculpture, decimated roads and bridges, and demolished barns. Junction Creek tore out the Main Avenue and railroad bridges before adding its load to the Animas, which carried an estimated 25,000 cubic feet per second of water as it ran through town. It’s an almost incomprehensible volume. A good spring runoff these days might lift the waters to 6,000 cfs, high enough for the river to leave its banks and spread across the floor of the Animas Valley, and to turn Smelter Rapid into a churning hellhole for rafters.
The water unmoored the railroad bridge near Durango’s fish hatchery and carried it downstream, despite the fact that two full coal cars had been parked on the bridge to provide ballast. The river jumped its channel and headed onto 15th street, creating a five-foot-deep river that today would go right through a Burger King. further downriver the waters washed away 100 tons of toxic slag from the Durango smelter, and carried away several homes from Santa Rita, on the opposite shore.
Sixty miles east of Durango, in Pagosa Springs, the upper San Juan River swept away more than 20 structures and destroyed the town water plant, hospital, and jail. Its power plant “was wiped out of existence, nothing left but the water wheel.” The Bayfield Blade called Arboles, a village near the junction of the San Juan and Piedra Rivers, “a thing of the past.” That was a bit of hyperbole, but maybe also prophetic: the community survived that flood, but was later buried under the waters of Navajo Reservoir. Further east the Rio Grande grew even grander and threatened to carry parts of Española, Bernalillo, and Albuquerque down to the Gulf of Mexico.
Over in Dolores, Colorado, the river peaked out at 10,000 cfs, more than 20% greater than the second highest peak hit in 1949. The raging river of sorrow ripped out railroad tracks, washed out roads, and inundated the town under four feet of water and four inches of mud, carrying away houses and the boardwalk. My great grandfather, John Malcolm Nelson, had come down from Ouray in early October to look at buying land in the Ute Strip — and he did, down at Sunnyside Mesa. But his trip back north was delayed by the fact that every bridge and road in the region was washed out.
In Farmington the seething monsters of the upper San Juan and the Animas joined forces, spilling over the banks and onto the flats on either side of the river, where the Navajo mission sat. Simmons and his fellow staffers sent the children to higher ground at about midnight as a precaution, equipping each with a blanket and loaf of bread. Then they went to bed, not realizing their own mistake until they awoke four hours later.
Somehow, Weston was able to quickly escape on horseback (he may have snuck out earlier). Tice chose to stick around, heading for the top floor of the structure. Simmons ran out and climbed atop an outhouse, apparently in order to launch himself onto a horse. Simmons missed the horse and ended up in the water, instead, carried rapidly downstream alongside dead animals, haystacks, and pieces of people’s homes.
Tice, it seemed, was the only survivor, and as the sun came up, onlookers gathered on the opposite shore. They watched Tice climb from the second story to the third, finally climbing onto the roof with his dog. It seemed safe enough; the water stopped rising after it inundated the third story. Little did he know, the waters were slowly dissolving the building underneath him, and it, the roof, the dog, and finally Tice were all swallowed up by the current.
The Shiprock Indian School campus was covered with water five feet deep, washing away several adobe buildings, and the fairgrounds, prettied up for the annual fair, were covered with a torrent of muddy water. Every bridge in San Juan County, Utah, where a miniature oil boom was on, was torn loose and carried away by the angry torrent; 150,000 cubic feet of water shot past the little town of Mexican Hat every second, according to a 2001 USGS paleo-flood hydrology investigation. That’s about 100 times the volume of water in the river during a typical March or April, a popular time to raft that section. It took out the then-new Goodridge bridge — some 39 feet above the river’s normal surface — tore through the Goosenecks, backed up in Grand Gulch, deposited trees on sandstone benches high above where the river normally flows, and finally combined with the raging Colorado River to create a liquid leviathan of unknown volume that wreaked more havoc through the Grand Canyon and beyond.
***
The 1911 event is typically considered to be the Four Corners Country’s biggest flood, based on streamflow estimates, anecdotal accounts, and the damage wrought. Since then it has been rivaled only by the June 1927 flood, when the Animas River in Durango reached 20,000 cubic feet per second; and in 1949 and 1970 when the high-water mark was about 12,000 cfs and 11,600 cfs, respectively. That might make 1911 seem like a freak event — a once-in-a-millennium confluence of factors. Combine that with the fact that the river’s annual peak streamflows have trended downward over the last century or so, and a 1911 repeat seems less and less likely.
But these waters are muddied, so to speak, by the relatively short timeline and limited geographical scope we’re working with. Many streams didn’t have gages on them at the time, and even those that were present weren’t always accurate (most of the 1911 figures are estimates, not actual measurements). Even though most of the “old-timers” said it was the biggest flood they’d ever seen or heard of in these parts, we have to remember that they tended to be white guys, and white settler-colonists had only been in the area for four decades or so. Not that memories of weather events are ever all that reliable.
A swollen San Juan River nearly wiped Montezuma Creek and Bluff City, Utah, off the map back in 1884 (the 1911 flood wreaked less destruction). Yet there were virtually no stream gages, so the magnitude of that earlier event is hard to quantify and, besides, maybe the later flood was less destructive because there were fewer homes and infrastructure in the flood’s path by then.
Also, when one looks beyond the San Juan Basin watershed, one finds streamflows that far exceed those of October 1911. On the USGS stream gage on the Green River in Green River, Utah, the 1911 flood (which was at the beginning of the 1912 water year, by the way) ranks as just the 5th largest flow since 1895. And 1911 places fourth overall on the Rio Grande at Otowi Bridge, outdone by 1920, 1941, and 1904.
We can extend the timeline dramatically by turning to paleoflood hydrology, which is sort of like dendrochronology, except instead of looking at tree rings to understand past climate, it uses geological evidence — slackwater lines, debris — to reconstruct the magnitude and frequency of past floods. I skimmed the available literature, including this Bureau of Reclamation survey of studies, and here’s what stood out:
The 1911 flood was likely the largest on the Animas River over the last several hundred years or more. On the San Juan River near Bluff, researchers found no evidence of floods higher than the 1911 debris, indicating it “may represent the largest flood on the San Juan River for a much longer time period than 1880-2001.” In any event, 1911 was larger than the 1884 flood, even in Bluff.
On the Colorado River at Lees Ferry the 1884 flood was most likely the largest during white settler-colonial times, with an estimated flow of about 300,000 cubic feet per second (there were no gages there, yet), which would have provided quite the ride through the Grand Canyon. Some researchers believe an 1862 flood had a flow of about 400,000 cfs. Holy big water, Batman!
Extend the timeline further and the ride gets even wilder: A 1994 USGS paleoflood study found evidence of a 500,000 cfs flood at Lees Ferry between 350 and 750 A.D.; and a 2018 reconnaissance found slackwater deposits indicating a flow of 700,000 cfs. I’m sure it provided quite the scene for Puebloan observers looking down from the canyon rim. If you happened to be in the canyon at that time? Yikes.
From: “A 4500 Year Record of Large Floods on the Colorado River in the Grand Canyon, Arizona,” by Jim O’Connor et al. A study of floods on the Colorado near Moab found that, as is the case on the Animas River, there were a lot of large floods between the 1880s and 1930s, but peak streamflows have followed a decreasing trend ever since. One study suggested this resulted from: land-use changes, particularly a severe reduction in grazing after 1932; greater regulation of the river by upstream dams and so forth; greater upstream water consumption; and a decrease in intense, large flood-producing storms.
The Colorado River near Moab has experienced 44 floods during the last two millennia with flows ranging from 63,500 cfs to 325,000 cfs. (For context, the 1983 runoff, which threatened Glen Canyon Dam, reached 62,000 cfs on this stretch of river and in 1984 it hit 70,300). Most of those floods occurred during the last 500 years.
From “A 2000 year natural record of magnitudes and frequencies for the largest Upper Colorado River floods near Moab, Utah” by Greenbaum et al.
Warming temperatures, like those resulting from human-wreaked, fossil fuel burning-exacerbated climate change, can increase the intensity of storms and the amount of precipitation. That could, potentially, lead to bigger floods. So even though climate change has mostly manifested as drought in the Four Corners Country, it could also have the effect of putting a 1911-like storm on steroids. And with El Niño brewing in the Pacific, we might see some whopper storms sooner rather than later. Or not. Either way, though, it seems silly to assume the 1911 flood won’t repeat someday. Maybe next time it will be even worse.
That 1911 storm dissipated over the next couple of days, leaving a bright sun to illuminate the river valleys, newly scoured of the roads, houses, bridges, railroad tracks, and other detritus that humans had littered the valleys with over the previous decades. But the folks of the San Juan Basin soon went to work rebuilding — quite often in exactly the same spots that had flooded so catastrophically.
I used to see that as a combination of foolishness, hubris, obliviousness, and stubbornness all woven into a tapestry of denial. Surely they couldn’t have believed a flood of that magnitude would never occur again.
And yet, now that I’ve fallen victim to a flood, or at least my home has, I finally get it. What do I know about their circumstances? Maybe they had invested everything they owned into this little plot of land and a home, and they have nowhere else to go. Maybe they are just so wedded to this particular place that they figure it’s worth the risk to build in a 100-year flood plain. Maybe they were just tenacious bastards shaking their fist at the sky in defiance.
What I do know is that if and when there is a repeat of the 1911 flood, or that whopper that sent 700,000 cfs into the Grand Canyon, it will leave some serious destruction in its wake.
The 1911 flood wrecked a lot of infrastructure, but the human death toll was much smaller than one might have expected. Among the handful of fatalities was Frank B. Tice, of the Navajo Methodist Mission, whose body was found 20 miles downstream from where he was swept away.
But there was something else, too. On an island in the San Juan River, somewhere between Farmington and Shiprock, a man huddled next to a small fire, cooking apples that he had snagged as they bobbed past. After falling in the water he had grabbed ahold of some debris, and it had carried him for miles until he finally reached the island, cold, wet and hungry but, maybe miraculously, alive. It was J.N. Simmons, of the Navajo mission.
A 1998 paleo-flood investigation determined the measurement was in error and it was more likely that about four inches fell across a wider area. In any event, the author of the report does not dispute the magnitude of the flood that resulted.
Great timing with this! In my geomorpholgy class at FLC we were just yesterday calculating flood RIs for the Animas River based on the gage record. Depending on how you project it out, the "500 yr" flood could be as much as 40-45 kcfs.
The flooding history of southwestern Colorado was great – hopefully it will find its way into public and school libraries in Durango and surrounding areas. Thanks for your research. And by-the-way, I found one of your earlier dispatches on the fascinating history of bicycling in Durango. I am not really into bikes – at sixteen mine was retired in favor of an old 1951 Ford. And it’s been those "steel coffins" ever since. In spite of this mind set your story held my interest ‘till the very end - love your writing.