How an 1887 Sonoran earthquake dried up Southern Arizona wells and springs, fed others

After the quake struck the small Sonoran town of Bavispe, the key spring serving San Xavier Mission south of Tucson — far from the quake's epicenter — ran dry.  

The El Ojito spring in downtown Tucson had served water to nearby residents since 1770 — but it dried up after the earthquake.

The water level in the well serving the U.S. Army's Fort Lowell on what is now Tucson's east side dropped from 30 to 100 feet deep, possibly contributing to the Army's decision to abandon the fort. Though the primary reason for the abandonment was the end of the Apache Wars, it also became much harder for the Army to draw water out of the aquifer. Water stored in the fort's water tower also started tasting brackish.

The Rillito River stopped flowing year-round west of Fort Lowell, near where Craycroft Road now crosses the riverbed. That was attributed in part to the quake and in part to water use by farming and cattle grazing.

A section of the San Pedro River near the now-abandoned town of Charleston, north of Tombstone, temporarily dried up. Downstream near St. David, however, the earthquake brought underground artesian springs very close to the surface, enabling their water to be used for growing pecans, walnuts and other crops. The quake caused the San Pedro to surge there, wiping out stagnant, mosquito-laden pools linked to cases of malaria.

All these and other impacts of the 1887 earthquake on Southern Arizona water sources are laid out in Tucson author Mary Elizabeth Reynolds' new book, "The Quake that Drained the Desert."

The book traces how the earthquake contributed to a wide range of effects on water supplies — impacts that in some cases lasted decades and may even still be around today.

Debate about the Santa Cruz 

Left unsettled by the book, however, is the tantalizing question of whether that quake, which registered 7.4 on the Richter scale, played any role in the eventual drying of the Santa Cruz River through Tucson's heart.

Before the turn of the 20th century, the Santa Cruz ran perennially in some sections of Tucson, less frequently in others, and only rarely in some others. But by the 1940s, the entire river through Tucson usually carried water only during and after big storms. That condition persists today except in areas where treated sewage effluent is discharged into the river by Tucson and Pima County agencies.

The book lays out a scenario under which the earthquake could have altered the geology of the river's underlying aquifer enough to at least contribute to the drying. However, a longtime Tucson hydrologist who has extensively studied the Santa Cruz, Robert Webb, insists the drying of the river was mainly triggered by excessive groundwater pumping, the cause most commonly cited by many other water experts.

Webb noted that the river has never flowed continuously from its headwaters just north of the Mexican border in Santa Cruz County to where it drains into the Gila River in Pinal County, except during the record-setting floods of October 1983.

Historically, the Santa Cruz did have year-round supplies, but only at its confluence with the Rillito River and Cañada del Oro Wash and at the base of "A" Mountain west of downtown, Webb said.

"There’s no evidence the 1887 quake did anything to the Santa Cruz. It was continuous ephemeral stream," the kind that only carries water after rains, Webb said. "The perennial reaches were all wiped out by groundwater pumping. That's well documented."

Another longtime student of the Santa Cruz, Julio Betancourt, who, like Webb, was interviewed by Reynolds for the new book, said he thinks it's a possibility the earthquake played a role in the river's drying.

Now both retired, Betancourt and Webb worked for the U.S. Geological Survey in Tucson for many years. They were also two of four co-authors of the 2014 book "Requiem for the Santa Cruz," a natural history of the river's decline during the 20th century that Reynolds cites in her book. 

"Generally what happens, when you have earthquakes like that you start getting some liquefaction," Betancourt said. He was referring to a process in which loosely packed, water-logged sediments at or near the ground surface lose their strength in response to strong ground-shaking.

During earthquakes, liquefaction occurring beneath buildings can cause major structural damage, the USGS  says.

Underground in that circumstance, "the aquifer can compress and squeeze the water out," Betancourt said. "All of a sudden, you have bodies of water show up or disappear because of it."

Also, a good deal of "downcutting" of the riverbed started not long after the quake, around 1890, causing the streambed to be lower in elevation and presumably causing the water table to fall, he said. 

How earthquakes can affect springs

A 1980 publication by the state-run Arizona Geological Survey published a long list of anecdotal accounts of changes to water features across Southern Arizona and northern Sonora after the earthquake, Betancourt noted.

Tucson isn't very far from the 100-mile radius where the quake did the most damage, he said, and "people were noticing water appearing when there wasn't any before."

The "Requiem" book and the new book spell out clearly, however, that the 1887 earthquake triggered significant hydrologic changes in its radius around Bavispe, Sonora, its epicenter.

They included "sudden water level rises and falls in wells, increases in spring discharge, rise of groundwater to the surface because of subsidence or opening up of new fissures, and forcible ejection of water and mud from fissures," the "Requiem" book said.

"A fissure 20 miles long was reported along the San Pedro River north of Benson, and a considerable stream of water reportedly flowed from it. Throughout southern Arizona, there were reports of increased stream discharge during the driest months of the year."

In an email to the Star, author Reynolds explained how earthquakes can cause springs to form — or disappear.

"Springs emerge where there are 'fractures' in the surface layers of earth, where groundwater comes up from the aquifer. The groundwater hits an impermeable layer (like rock) and is forced up to the surface rather than traveling between the spaces of sand and gravel beneath the surface.

"An earthquake shifts impermeable rock layers so some springs close and water stops flowing, while in other places the water that was previously under the surface hits a newly moved rock layer and is forced to the surface."

Impermeable layers of sediment are different in different parts of Southern Arizona, "so that is why groundwater behaved differently after the earthquake along the Rillito and along the San Pedro," she said.

Her book also tells the story of how an earthquake in Idaho altered well levels and even the behavior of the Old Faithful geyser at Yellowstone National Park, more than 200 miles away.

Geologists and hydrologists who measured well levels in Idaho, Montana and Wyoming after what's known as the Borah Peak earthquake in 1983 found that half of them either had permanently dropped or their water had become so dirty it couldn't be used for drinking, she said.

Before that earthquake hit, Old Faithful used to erupt with spring water every 59 to 70 minutes, she said. Four months later, the eruption intervals had risen to 77 to 78 minutes, she said.

These quakes were similar in some ways, she added. They occurred in the middle of the underlying ground's tectonic plate, not on the edges. They both occurred in what's known as the Basin and Range geologic province. And they both left behind significant tears in the earth near where they occurred.

The Borah quake "was a very similar earthquake to the Sonoran quake and it had a similar effect on water," Reynolds said.

Long list of anecdotal accounts

Here are some other specific impacts from the 1887 Sonoran Earthquake, as laid out in Reynolds' book and by a 1980 report on the quake produced by the Arizona Geological Survey, which her book cites:

• After the 1887 quake dried up the existing spring at San Xavier, a new spring came up upstream of that area, allowing Tohono O'odham farmers to quickly build a reservoir and resume growing crops there.

• The El Ojito spring served a recreational area, known as Carrillo Spring, lying just south of the current Tucson Convention Center site, that included a carnival, games, boating and shooting galleries. Once the spring went dry, the Carrillo Spring owner, Leo Carrillo, had to divert water from the Santa Cruz and from city water mains nearby. 

• A well at the McKay house on Convent Avenue south of Tucson's current downtown also dried up.

• The San Pedro, away from the Charleston area, changed from a stream that you could step across to a wide-ranging river, the Tombstone Epitaph newspaper reported.

• A news clipping from the time of the earthquake reported that since the quake, the water supply in the area of Cochise Stronghold east of Benson was said to have "entirely disappeared."

• At the time of the quake, the Corralitus Mining Co. in Mexico reported that first, "springs have broken out in several parts of our property, where water poured up several feet," although they subsided as the violent shocks ceased. Another, previously small spring, however, opened up into an abundant one, the company said.

• In Davidson Canyon southeast of Tucson, water flow "doubled in quantity" since the quake, said a news clipping cited in the Arizona Geological Survey article. 

• In mountainous areas near Elgin in Santa Cruz County, "streams were drying up in 24 hours" after the quake, the 1980 article said.

"The earthquake's impact reaches across 140 years," Reynolds asserts, "to the current water crisis on both sides of the international border."