Utah experienced its single deadliest flash flood on record on Monday, and Los Angeles saw one of the wettest September days in its history, as moisture from the remnants of former Category 3 Hurricane Linda
was carried into the Southwest. Strong thunderstorms on Monday afternoon sent torrents of water downstream into a crossroads in the steep terrain just northeast of the town of HIldale, at the Utah/Arizona border just south of Zion National Park. Two vehicles carrying 16 people were swept away by water and debris on Monday afternoon. At least 8 of the 16 occupants died, with 3 others rescued; a woman and four children remain missing. Flash floods are a particular risk in this area because of its steep terrain and narrow canyons, which make it hard to judge one’s risk. Floodwaters can pour into areas where little or no rain has fallen. The downpour was focused in and near the cliffs just upstream (north) of Hildale. Figure 1.
National Weather Service NEXRAD radar reflectivity from 2:20 pm MDT Monday, Sept. 14, 2015, shows an intense thunderstorm with very heavy precipitation (bright red) centered in far southeast Washington County, Utah, near the town of Hildale. A flash flood warning was issued at 2:22 pm for the area. Additional heavy rain struck about two hours later. Image credit: NWS and RAL Real-Time Weather Page
. Figure 2.
Debris and water cover the ground after a flash flood Monday, Sept. 14, 2015, in Hildale, Utah. Image credit: Mark Lamont, via AP.
The isolated heavy thunderstorms over Utah and Arizona on Monday emerged as a rich stream of Pacific moisture, with some contribution from Linda’s remnants, intersected with a seasonally strong upper-level trough now making its way across the western United States. Atmospheric soundings from Monday show above-average amounts of water vapor, but even higher amounts have been reported this time of year, so the meteorological setup was not particularly extreme. “I don't see the conditions across Utah as being all that unusual,” Jim Steenburgh (University of Utah) told me. “This is a monsoonal pattern producing monsoon convection. Flash floods are a natural component of the monsoon system and the geology of the Colorado Plateau.”
As is often the case in flash floods, part of the disaster can be attributed to bad meteorological luck--the fact that an intense thunderstorm happened to develop and move slowly through a particular spot prone to flash flooding. Steenburgh analyzed the topography and meteorology involved in this tragedy
this morning at his Wasatch Weather Weenies blog. As shown in Figure 1, an intense thunderstorm struck around 2:20 pm MDT; it was followed by a second intense cell that struck the same area around 4:15 pm MDT (see Steenburgh’s blog for a radar loop). “There is a possibility that the two storms provided a one-two blow that resulted in a more substantial flash flood event than if either had occurred in isolation,” noted Steenburgh. In an email, he added that there is some uncertainty over whether or not the fatal accident happened after the first cell or the second.
Forecasters at the Salt Lake City office of the National Weather Service recognized early Monday morning
that a moderate risk of flash flooding existed in southern Utah, and that morning’s hazardous weather outlook
warned of the potential for locally heavy rainfall, especially across southern Utah. The office issued a flash flood warning at 2:22 pm MDT
for southwestern Kane and southeastern Washington counties, including the Hildale area, noting that radar estimates indicated at least 2” of rain had fallen.
New high-resolution models have increased the ability of forecasters to project the evolution of thunderstorms several hours in advance, but the hugely varied terrain of the Southwest, and the resulting complex patchwork of atmospheric conditions, makes it very difficult to give much advance warning of the kind of flood that struck near Hildale. Further north, Salt Lake City was also hammered by fast-developing thunderstorms that struck the University of Utah campus within 15 minutes of their formation. This prompted UU’s Steenburgh to reflect on the challenges of forecasting
in his blog on Monday, prior to the Hildale flood: “There is no human or computer-based forecast tool available today that can reliably forecast the development of a storm cell from nothing on such short time scales...Essentially, developing computer models that can provide detailed and reliable forecasts of convective storms at short lead times represents the holy grail of nowcasting. This is an area of active research in the atmospheric sciences and perhaps we will make some inroads in the coming years and decades, if not for storms like the [Salt Lake City storm], perhaps for severe convective phenomena like derechos and supercells.”Figure 3.
Commuter traffic makes its way slowly along Interstate 10 in downtown Los Angeles amid heavy rain on Tuesday morning, Sept. 15, 2015. Image credit: AP Photo/Richard Vogel.Heavy rain hits coastal Southern California
Hollywood and environs are getting a sneak preview today of what could be a meteorological blockbuster this winter. The feed of moisture that includes Linda’s remnants is combining with upper-level energy from the Western trough to produce unusually heavy rains for so early in the autumn, with widespread 1” to 2” amounts from the Los Angeles area to northern San Diego County. Several highways have been closed, and more than 100 people were evacuated from a flooded assisted-living center in West Hollywood, as reported by KTLA and the Los Angeles Times
. Unfortunately, the rains are too far south to help douse the devastating wildfires
now unfolding in northern California.
From midnight through 9:47 am PDT Tuesday, downtown Los Angeles
racked up 2.39” of rain, making it the second-wettest September day on record. Only six other September days have yielded more than 1” of rain in downtown LA, where records date back to 1877:3.96”
Sept. 25, 19392.39”
Sept. 15, 2015 (through 9:47 am PDT) 1.95”
Sept. 24, 19861.74”
Sept. 10, 19761.62”
Sept. 24, 19391.58”
Sept. 30, 19831.39”
Sept. 18, 1965
El Niño conditions were in place and intensifying during all of these events except for 1983, just after the “super” El Niño of 1982-83 had wound down. Since El Niño tends to increase the number of hurricanes in the Northeast Pacific, it’s not surprising that several of the above events were associated with decaying tropical cyclones, including Hurricane Newton (1986), Hurricane Kathleen (1976), Tropical Storm Octave (1983), and the memorable cyclone in 1939
that made landfall near Long Beach as a full-fledged tropical storm (1939), producing a two-day total of 5.58” at downtown Los Angeles and a storm total of 11.60” at Mount Wilson. (Thanks to Jon Erdman and Jen Watson of The Weather Channel for data on the September rains.)Figure 4.
The WU tracking map was devoid of tropical storm names at 1:40 pm EDT on Tuesday, September 15, 2015.Earth was free of named tropical cyclones on Tuesday morning
There were no named tropical storms on the planet on Tuesday morning, the second such period to occur this month after a 54-hour streak last weekend (the longest such streak September since 2009
, according to WU contributor Phil Klotzbach at Colorado State University). The tropics aren’t completely tranquil, though, as several systems are being monitored for potential development.Figure 5.
An infrared image from Meteosat-9 shows Invest 95L (far left) and Invest 93L (center) at 1500 GMT (11:00 am EDT) on Tuesday, September 15, 2015. Image credit: EUMETSAT and NHC
is the healthiest of the Atlantic waves, with a large area of showers and thunderstorms gradually consolidating south of the Cape Verde islands. Toward the central Atlantic, Invest 93L
enlarged substantially overnight, although its convection is somewhat scattered. The National Hurricane Center gives both 95L and 93L a 60% chance of development by Thursday. Models generally support the idea of both 95L and 93L becoming tropical storms later this week. However, as upper-level troughs are now beginning to dig further southward toward the subtropical Atlantic, both systems are likely to recurve long before they have a chance to threaten the Caribbean or North America. The center of Invest 94L
has now moved into Mexico near Tampico, although there remains extensive thunderstorm activity to its east under a region of fairly high wind shear (greater than 20 mph). In the Northeast Pacific, Invest 90E
should spin harmlessly in open water, with NHC giving it only a 10% chance of development in the next five days.
The one system on Earth now showing potential to become a major tropical cyclone is Tropical Storm Twenty
, whose winds reached minimal tropical-storm strength at 1200 GMT Tuesday. The Joint Typhoon Warning Center projects
TS 20 to strengthen into a Category 3 typhoon as it recurves just east of Iwo Jima by Friday.