What the Weekend Rains Did to Southern California—and What a Real Hurricane Could Do

The maxim “be careful what you wish for” comes to mind when pondering last weekend’s amazing rain across Southern California. After three years of fierce drought, some of the heaviest midsummer rains on record struck the region, facilitated by the remnants of former Hurricane Dolores. The spirit-boosting effects of the rain were accompanied by some rare disruptions, including a bridge collapse along Interstate 10 in far southeast California and the first rained-out pro baseball game at Anaheim’s Angel Stadium since 1995. As unusual as the rain was, there could be even more to come in the next several months, as an already-strong El Niño event continues to gather steam. Records for Southern California from the past century show that the risk of impact from Pacific tropical storms and hurricanes is greater during El Niño years. History also tells us that we can’t rule out the possibility of a full-blown hurricane coming ashore: this apparently happened in 1858, long before the region was densely populated.


Figure 1. Emergency crews respond after a pickup truck crashed into the collapse of an elevated section of Interstate 10 on Sunday, July 19, 2015, in Desert Center, Calif. The bridge, which carries the eastbound interstate about 15 feet above a normally dry wash, snapped and ended up in the flooding water below, the California Highway Patrol said, blocking all traffic headed toward Arizona. Image credit: Chief Geoff Pemberton/CAL FIRE/Riverside County Fire, via AP.

A century’s worth of July rain in one weekend
As the circulation around ex-hurricane Dolores stalled southwest of San Diego over the weekend (see Figure 3), the flow around it contributed to a channel of rich moisture flowing northward across much of Arizona and Southern California. Dewpoints rose to sultry Southern levels as high as the upper 60s in Los Angeles and the low 70s in San Diego. Very light rain had been in the forecast along the southern CA coast, but the amount of instability proved greater than expected, and intense thunderstorms developed on Saturday afternoon and evening. Much of western San Diego County picked up an inch or more of rain on Saturday, with another 0.50” – 1.00” widespread on Sunday and some lighter amounts on Monday. San Diego’s Lindbergh Field measured a whopping 1.69” on Saturday and Sunday—more rain than in any other July in San Diego records that go back to 1850 (the runner-up was 1.29” in July 1865). Midsummer is typically bone-dry in San Diego, with June through August racking up a combined average of just 0.14”. Amazingly, the past weekend produced more rain in San Diego than the previous 100 Julys combined (1915 – 2014). This was also more rain than the year’s wet-season months of January, February, and March managed to cough up. Downtown Los Angeles notched 0.38” over the weekend, its greatest monthly total for any July in records going back to 1877 (the previous record was 0.24” in July 1886). In that entire 139-year period, the station recorded a total of 1.17” of July rain, with close to a third of it falling this year.


Figure 2. More than 20,000 intracloud and cloud-to-ground lightning flashes occurred in southern California from Friday, July 17, through Sunday, July 19. Image credit: NWS/San Diego.


Figure 3. This satellite image tweeted by the NWS San Diego office on Sunday morning, July 19, shows the remnants of Dolores (lower left) and a displaced cluster of thunderstorms north of Los Angeles (near top of image). Image credit: NWS/San Diego.

Chris Landsea, science and operations officer at NOAA’s National Hurricane Center, points out the absence of the usual July circulation patterns that shove Northeast Pacific hurricanes westward before they can influence the United States so strongly. “Typically in July you have a strong subtropical ridge centered over Mexico and a quick westward steering induced by the ridge,” Landsea told me. “This year, that ridge is much weaker.” Although Dolores did not reach Southern California, it edged further north than almost any other Northeast Pacific system on record for July (see Figure 4 below).

SoCal’s natural, but imperfect, buffer against hurricanes
The ocean circulation that keeps California’s coastal climate naturally air-conditioned also helps protect the state from direct hurricane landfalls. Prevailing sea-surface temperatures (SSTs) are far too cool to support a tropical cyclone, even as far south as San Diego. Hundreds of miles further south, though, summer and autumn SSTs are more than warm enough to make the Northeast Pacific (also called the Eastern North Pacific) fertile ground for hurricane formation. This is especially true during El Niño, which typically pushes SSTs above average across the Northeast Pacific hurricane genesis region. El Niño also provides favorable dynamics for development, as rising motion and weak upper winds tend to predominate across the Northeast Pacific. Heavy rains often result when moisture streams into the western U.S. from tropical cyclones in this region. Sometimes the remnants of tropical systems move bodily into Southern California from the Pacific, but a more dangerous approach can be from the southeast, via the extremely warm waters of the Gulf of California, where SSTs can soar above 29°C (84°F). One can also envision a strong hurricane developing west of Mexico and moving north quickly enough to maintain some of its power into southern California.


Figure 4. Tracks of the 223 tropical cyclones in the NOAA Historical Hurricanes database that occurred in July (1950 – 2013) across the Northeast Pacific. Dolores was tracked by the National Hurricane Center until it became a remnant low late on Saturday, July 18, near latitude 25.6°N, a location near the top of the tracks shown here. Tropical cyclones are more likely to affect the southwest U.S. in August, September, and October than in July. Image credit: NOAA.

California’s tropical cyclone history
A roundup published in 1997 by the Los Angeles (Oxnard) office of the National Weather Service summarized the seven years of the 20th century in which at least two tropical cyclones affected Southern California: 1921, 1939, 1972, 1977, 1982, 1983, and 1997. The office stated that six of those seven were classified as El Niño years. Most of these occurred with El Niño conditions taking hold during the autumn, with 1921 being the La Niña outlier. In addition, the highest winds ever recorded with a tropical cyclone in the Southwest U.S. were from Hurricane Kathleen, which delivered sustained winds of 57 mph at Yuma, Arizona, on September 10, 1976—another El Niño-onset year. Torrential rains over far southeast California (as much as 14.76” at Mount San Gorgonio—see PDF) and related flooding ripped out parts of I-8 and other roads and damaged or destroyed hundreds of homes, inflicting an estimated $160 million in damage.


Figure 5. Tracks and intensities of Hurricanes Kathleen (tracking through California and Nevada as an extratropical system) and Nora (paralleling the Colorado River as a tropical storm and tropical depression). Image credit: NOAA.

The region got a startle in September 1997 when Category 5 Hurricane Linda became the most powerful storm ever recorded in the Northeast Pacific. At one point, some computer models predicted that a weakening Linda would recurve into or near southern California, and long-range forecasts from the National Hurricane Center briefly depicted this scenario. The Oxnard NWS office addressed the potential in public statements and forecast discussions, according to the NHC report on Linda. “Some of these products specifically mentioned the uncertainties in forecasting Linda and asked the media not to overdramatize the storm,” the report noted. Only a few days later, fast-moving Hurricane Nora entered the United States with tropical-storm-force winds near the CA/AZ border after cruising north along the east side of Baja California. “The threat to the southwestern United States was unusual and required an unprecedented coordination between the NHC and offices in the NWS Western Region,” said the the NHC’s report on Nora. Heavy rains over the U.S. Southwest were much less than during Kathleen, but damage may have topped $100 million.

The big ones: 1939 and 1858
The most ominous precedents for the megalopolis from San Diego to Los Angeles are the 1858 San Diego hurricane mentioned above and a deadly 1939 system that struck Long Beach as a tropical storm. The 1939 cyclone appears to have reached minimal Category 1 strength on the day before before its northward track brought it into San Pedro, CA (just west of Long Beach) on September 25, bringing an intense week-long heat wave to a dramatic end. Remnants of three other cyclones had affected southern California earlier in the month, as noted in the NWS summary, and moisture surging into the area ahead of the September 25 storm helped trigger heavy rains the day before landfall. In addition, modern flood control devices were not yet in place. The storm gave Los Angeles its 24-hour rainfall record for September with 5.24”, and Mount Wilson picked up 11.60”. An estimated 45 deaths occurred during the storm, including some at sea. Wind damage was minimal, although windows were reportedly blown out across Long Beach. The storm cost an estimated $2 million in 1939 dollars; such a storm would probably cause hundreds of millions in damage today, given the growth in population and wealth across the area.

Less is known about the 1858 San Diego hurricane, although independent researcher Michael Chenoweth teamed up with Chris Landsea to put together the first detailed analysis of the storm, published in the November 2004 issue of the Bulletin of the American Meteorological Society. Chenoweth and Landsea drew on daily weather reports collected by U.S. Army medical staff, as well as California newspaper accounts and U.S. Coast Survey notebooks. On October 13, 1858, the Daily Alta California reprinted coverage from the San Diego Herald that included this account:

“About 11 o’clock A.M. of Saturday, 2d instant, a terrific gale sprung up from the S.S.E. and continued with perfect fury until about 5 P.M., when it somewhat abated, and rain commenced to fall. It blew with such violence, and the air was filled with such dense clouds of dust, that it was impossible to see across the Plaza, and it was with the greatest difficulty that pedestrians could walk the streets. The damage to property was considerable; houses were unroofed and blown down, trees uprooted, and fences destroyed. It is said to have been the severest gale ever witnessed in San Diego.”

Based on the reported leveling of trees and complete destruction of some homes, comparable to F1/F2 damage on the original Fujita tornado damage scale, Chenoweth and Landsea concluded that sustained winds on the order of 50 to 68 knots (57 – 78 mph) appear to have occurred. “We assume that the more substantial buildings were only unroofed, and weaker structures completely destroyed,” they wrote. “The F2 damage was probably done by the higher gusts during the hurricane, accompanied with structural weakening due to the [four hours] of high winds. The account of ships being driven ashore, the visually estimated winds, and structural damages from winds are all consistent with a category-1 hurricane impact in San Diego.”

What would happen this time?
San Diego County’s population in 1860 was a mere 4324, compared to more than 3.2 million today. To the north, there are some 3 million people in Orange County and about 10 million in Los Angeles County. The massive growth in population and wealth would clearly make a reprise of the 1858 hurricane a much more serious affair, inflicting at least a few hundreds of millions in damage, according to Landsea. Although wind damage is an obvious concern, the region already experiences howling gales as a result of Santa Ana and gap-wind events, where gusts can top 60 mph. A prolonged Santa Ana might produce wind damage comparable to a tropical-storm landfall, says Landsea. His bigger concern is inland flooding: “If you got a direct hit by a hurricane, some of the coastal mountains could see 10” or even 15” of rain. The rainfall-produced flooding would be severe. Wind impact would be more secondary.” Unlike much of the Gulf and Atlantic coastline, storm surge would be a relatively minor problem, thanks to the steep topography along most of the Southern California Coast.

The National Hurricane Center has never issued official advisories for any land areas in California (or Arizona). That could change if a storm like Linda targeted the coast. “We’d be looking at tropical storm watches or warnings and perhaps hurricane watches or warnings. We’d also have the same type of coordination with local offices that we now have in the Eastern and Southern Regions [of the NWS]. All this is in place. We know we will again someday have a tropical storm or hurricane in this area.”

This 2012 NASA interview with researcher Bill Patzert discusses the California hurricane risk in more detail. See also this 2011 article by NHC director Rick Knabb, who lists San Diego as the nation’s second-most-overdue hurricane strike.

Bob Henson


Video 1. A short video documentary with newspaper clips and 16-mm film of the 1939 tropical storm as it approached Newport Beach, CA.