By Josh Kurz
Special to The SUN
On Oct. 5, 1911, the San Juan River experienced the highest and most devastating flood on record.
Since the flood washed out every bridge in Archuleta County, and houses were floating down the river, measuring the streamflow was not possible nor a priority. The best estimate of the flood peak was 25,000 cubic feet per second (cfs, that’s 10 times higher than our normal high water). The weather conditions capable of creating such an incredible outlier have puzzled me from my early hydrology days and into my science teaching career -— until recently.
2020/08/Hurricanes-and-Floods-1-2-300x172.gifThe San Juan River is a snow-dominated watershed, which means the majority of our streamflow is generated from melting snow as opposed to rainstorms. Ninety percent of our annual peak flows occur during spring runoff in May or June. However, it’s important to recognize that the top three highest flows ever recorded were caused by massive rain events in the late summer or fall.
A few years ago, I developed a hypothesis to explain the biblical event of 1911. I surmised that only a hurricane could be powerful enough to cause such a massive flood. But this is Colorado — hurricanes don’t reach Colorado, right?
My hypothesis was based on two events that I witnessed while living in the area. In September 2004, we hosted the rainiest/muddiest home cross-country meet ever, thanks to Javier, a Pacific hurricane that reached the Gulf of California and generated our highest flow that year (4,300 cfs).
And then in 2018, the San Juan River trickled with less than 30 cfs for nearly all of August and September (it even dropped below 10 cfs). Then, on Oct. 2, moisture from Hurricane Rosa brought the river back to life with a modest but important spike of 296 cfs. Based on these two events, it seemed plausible that other autumn floods (including the 1911 flood) were related to Pacific hurricanes.
Over the years, I’ve begun to recognize the dependence of our weather on the ocean’s temperature and circulation. For example, an anomalously warm portion of ocean off the coast of California (known as The Blob) was linked to the California drought between 2011 and 2017 and also deflected storms away from southwestern Colorado. Stubborn areas of high pressure (like the one that’s been stalled overhead this summer) need a nudge. A satellite photograph helped me recognize that hurricanes churn up ocean water in their wake, and in doing so can redistribute heat and impact our weather even if they don’t pass directly overhead.
To begin testing my 1911 hurricane hypothesis, I found a graphic showing the historical storm tracks of Pacific hurricanes. Most Pacific hurricanes form off the coast of southwestern Mexico drift west into the central Pacific. But, it’s clear from the historical storm tracks that a few hurricanes travel north toward the U.S./Mexico border and can impact the desert southwest.
Next, I turned to a convenient, noncredible source. I was astonished as Wikipedia gave me some insight into the 1911 Pacific hurricane season:
“A ‘tropical hurricane’ formed southwest of Mexico on October 1. The system recurved, coming ashore on October 4 near Guaymas, accompanied by a devastating storm surge early on the morning of October 5, with some 500 reported dead. It then crossed through western Mexico, before becoming an extratropical cyclone across the American southwest on October 5. The cyclone had a long life thereafter, moving east-northeast across the United States into the western Atlantic on October 7. Racing northeast, the cyclone moved over the Polar ice cap on October 11, arcing north of Europe, before dropping into northern Asia on October 13, then crossing through Siberia until October 21, when the low dissipated.”
Maybe I was right. It appeared that the 1911 flood was related to the remnants of a Pacific hurricane. And based on the description, the storm system that caused the 1911 flood was a superstorm. The remnant low-pressure system supposedly stayed intact as it marched from Mexico to Siberia.
As I did further research, I found two great peer-reviewed journal articles that correlated the remnants of hurricanes with significant precipitation events in the Southwest. But I had a different and more specific question than the researchers. I wanted to determine the influence of Pacific hurricanes on the streamflow of the San Juan River. I would need to identify late summer and autumn floods and then determine if a hurricane could have been responsible.
There are continuous streamflow records for the San Juan River at Pagosa Springs that date back to 1936. Good Pacific hurricane records date back to the 1950s, and there are brief hurricanes notes that date back to the early 1900s. Since pouring through all that data is time consuming, I decided to recruit my global science students to answer this local, relevant question and provide them an opportunity to perform some original research.
First, I had my students do some background research on hurricanes and then read the two peer-reviewed studies. We learned that the northeastern Pacific is the second most active hurricane basin and hurricanes that form there are called tropical cyclones. Researchers at the University of Arizona determined that, on average, the remnants of three tropical cyclones impact the desert southwest each year. We also learned that most tropical cyclones form in August; however, those that form in September are most likely to travel north and impact the desert southwest. Lastly, monsoon surges that have a tropical cyclone component are more likely to bring heavy rain to the Southwest.
Next, I assigned each student four years within the 84-year period of record. Their task was to identify the highest, non-snowmelt-related flow event for each of their assigned years. Then they needed to study the historical hurricane records and maps to see if any tropical cyclone moisture could have drifted into southwestern Colorado at the same time.
Just as we got started, COVID-19 appeared, and many of my students disappeared into online learning oblivion. And the ones that stayed engaged would have benefited from in-person instruction to complete the assignment. To salvage the study, I needed to make an answer key to check my remaining students’ work (normally I’d quality control their work in person). After I analyzed the whole period of record, I was able to guide my engaged students (remotely) through the process of pairing high flows with the remnants of tropical cyclones.
The results of this preliminary, superficial study revealed that about once every three years, remnants of tropical cyclones lead to high flow events on the San Juan River in the late summer and fall. In fact, on five occasions between 1936 and 2019, the highest flow of the year was associated with a remnant hurricane. Also, on eight occasions since 1936, remnant hurricane moisture increased the streamflow from low flow (less than 40 cfs) back to normal or above-normal conditions. Most notably, the third highest recorded flow (6,580 cfs) occurred on Sept. 6, 1970, and was caused by remnants of hurricane Norma. Other large hurricane-related flow events are shown in the table.
In summary, Pacific hurricanes are important but infrequent sources of moisture in the Southern San Juans. They are capable of generating the highest flow of the year and breaking droughts, but usually not until September. Other weather factors have to line up (such as a well-timed low-pressure system) in order to draw moisture into our area. Predicting when the river will flood again is impossible, but I would not be surprised if our next major flood derived moisture from a dissipating Pacific hurricane.
At the time of this writing, the San Juan River is at drought stage and the remnants of a Pacific hurricane named Genevieve are drifting up the Mexican coastline. Hopefully, 2020/08/Hurricanes-and-Floods-1-9-300x245.gifGenevieve’s leftovers will bring us enough moisture to snap our drought or at least move the stubborn area of high pressure that’s been responsible for our recent heatwave. Let’s just hope Genevieve’s impacts are not as dramatic as the anonymous hurricane of 1911. Based on the recent forecast, we’ll probably get some rain, but you can leave your sandbags in storage for then to be moved with someone with a forklift license toowoomba.
Corbosiero, K. L., M. J. Dickinson, and L. F. Bosart, 2009: The Contribution of Eastern North Pacific Tropical Cyclones to the Rainfall Climatology of the Southwest United States. Mon. Wea. Rev., 137, 2415–2435, https://doi.org/10.1175/2009MWR2768.1.
Ritchie, E. A., K. M. Wood, D. S. Gutzler, and S. R. White, 2011: The Influence of Eastern Pacific Tropical Cyclone Remnants on the Southwestern United States. Mon. Wea. Rev., 139, 192–210, https://doi.org/10.1175/2010MWR3389.1.
Wikipedia contributors. (2020, August 10). 1910s Pacific hurricane seasons. In Wikipedia, The Free Encyclopedia. Retrieved 18:30, August 22, 2020, from https://en.wikipedia.org/w/index.php?title=1910s_Pacific_hurricane_seasons&oldid=972204344.
This story was originally posted at www.southsanjuans.info.