Tropical Storm Likely to Form in Caribbean; Flash Flood Devastates Ellicott City, MD

The tropical wave in the eastern Caribbean dubbed Invest 97L continued to organize on Sunday night, and it will likely become a tropical depression or tropical storm on Monday and move into the western Caribbean as an intensifying tropical storm over the next day or so. Infrared satellite imagery on Sunday night revealed that shower and thunderstorm activity (convection) had become more intense and far more symmetric around 97L’s core, which was located early Monday morning about 350 miles east-southeast of Kingston, Jamaica. As of early Monday morning, 97L had not yet consolidated a closed low-level center of circulation, as seen by the lack of westerly winds in the ASCAT scatterometer image in Figure 2 below. However, surface winds were close to tropical storm strength on the north side of the system. As of early Monday morning, a NOAA Hurricane Hunter aircraft was tentatively scheduled to carry out a reconnaissance mission into 97L on Tuesday afternoon.

97L continues to chug westward at 20 to 25 mph, a speed that typically limits the ability of a tropical wave to intensify. In this case, however, upper- and lower-level winds are close enough in speed and direction to reduce the amount of wind shear that would otherwise affect fast-moving 97L. Long-range models agree that wind shear will remain light (around 10 knots or less) for at least the next three to four days, perhaps longer, along 97L’s path. In its 8:00 am EDT Monday tropical weather discussion, the National Hurricane Center gave 97L an 80% chance of development over the next two days. (On occasion, the NHC will bypass tropical depression status and upgrade a strong wave directly to tropical storm status.) 97L will be sweeping just south of Jamaica on Tuesday, perhaps as a moderately strong tropical storm (which would be named Earl], before approaching the Yucatan Peninsula later in the week (see below).


Figure 1. Enhanced infrared image of Invest 97L as of 1245Z (8:45 am EDT) Monday, August 1, 2016. Image credit: CIRA/RAMMB/CSU.


Figure 2. Low-level winds (in knots) detected by the Advanced Scatterometer (ASCAT) aboard the MetOp-A satellite as of 1302Z (9:02 am EDT] Monday, August 1, 2016. Image credit: NOAA/NESDIS/STAR.

The outlook for 97L
It’s now clear that 97L will survive its trek through the eastern Caribbean, which has long been known as the “hurricane graveyard” thanks to the climatological minimum in tropical cyclone formation over the area. According to a 2010 study in the Bulletin of the American Meteorological Society led by Owen Shieh (University of Oklahoma), this phenomenon is largely because of a predominant low-level southerly jet in the central Caribbean and an accompanying tendency toward low-level divergence over the eastern Caribbean. This pattern is most prevalent early in the season, peaking in July.

With a sprawling area of upper-level high pressure to its north, 97L should continue on a fairly straightforward west to west-northwest path for at least the next couple of days. There is strong agreement among our best longer-range track models, the ECMWF and GFS, that 97L will move south of Jamaica on Tuesday and approach the Yucatan Peninsula around Thursday. There are no signs of any major changes to the upper-level pattern that would divert 97L from this longer-term path, which could eventually bring it across the Yucatan Peninsula and into the Bay of Campeche. We will have to watch for any signs of a northward bend in 97L’s path later this week, but no members of the most recent GFS and ECMWF ensemble runs are indicating a track that would bring 97L into the heart of the western Gulf. Residents of Nicaragua and Honduras, and especially Belize, Guatemala, and eastern Mexico, will need to keep a close watch on 97L. Toward the end of the week, 97L could pose a threat to the western coast of the Bay of Campeche if it crosses the Yucatan Peninsula and survives the trek.


Figure 3. Oceanic heat content over the Caribbean on July 30, 2016, in kilojoules per square centimeter. The value is produced by integrating the vertical temperature from the ocean surface to the depth of the 26°C contour. 97L will be traveling over a large area of heat content greater than 100 kJ/cm2 on Tuesday and Wednesday. Image credit: NOAA/AOML.

Extremely warm water lies ahead of 97L
As it sweeps toward the western Caribbean, 97L will encounter a more favorable atmospheric regime as well as ominously warm water both at the surface and below. Sea-surface temperatures (SST) along 97L’s expected path from the area around Jamaica westward are around 29-30°C (84-86°F), which is roughly 1°C above average for this time of year. These warm waters extend to great depths, with large amounts of oceanic heat in the uppermost 200 meters (660 feet) of the northwest Caribbean (see Figure 3). Across the Caribbean and adjacent waters, the breadth and depth of oceanic heat content has been at near-record levels in recent weeks, as we discussed in a post on July 18. High levels of oceanic heat content are a major boon to tropical storm and hurricane intensification, because the storms do not churn up as much cold water as they otherwise would as they intensify. Rapid intensification is often associated with regions of high oceanic heat content, assuming that other conditions are favorable as well.


Figure 4. Infrared satellite image of Tropical Storm Howard at 1300Z (9:00 am EDT) Monday, August 2, 2016. Image credit: NOAA/NESDIS.

Howard becomes the Eastern Pacific’s eighth tropical storm of the season
Tropical Storm Howard was christened at 5:00 am EDT Monday, August 1, just a few hours too late to become a record-setting eighth tropical storm in the Eastern Pacific during the month of July. When it became a tropical depression on Sunday morning, Howard managed to tie the record, set in July 1985, for the eighth tropical cyclone of at least depression strength to form in the basin. The basin’s record of seven named storms for July, also set in 1985, was tied on July 22 with the formation of Tropical Storm Georgette (which went on to become the basin’s second Category 4 hurricane of the year).

Apart from extending the Eastern Pacific’s remarkable month-long string of tropical cyclones, Howard is unlikely to make much of a splash. At 5:00 am EDT Monday, Howard was located about 1000 miles west-southwest of the southern tip of Baja California. As it moves along a west-northwest path well offshore during the next 24 to 36 hours, Howard will have a window of light to moderate wind shear (10-15 knots) and SSTs of around 27°C, but it is not exceptionally well-organized and will have trouble strengthening much beyond mid-range tropical storm intensity. After Tuesday, Howard will encounter much cooler waters and greater shear, and the NHC outlook diminishes Howard to a post-cyclone remnant low by Thursday. The remains of Howard should arc leftward and may pass near or north of Hawaii by next weekend.


Figure 5. VIIRS visible satellite image of Typhoon Nida at 0509Z (1:09 am EDT) Monday, August 1, 2016. The Hong Kong area is outlined in a box along the yellow border denoting the southeast China coast. Image credit: RAMMB/CIRA/CSU.

Nida heads toward Hong Kong
Typhoon Nida continues to chug toward the southeast China coast, where it could make landfall very near Hong Kong. Nida dumped rains of more than 11 inches over parts of northern Luzon as it swept through the northern Philippines over the weekend. Fortunately, Nida has failed to intensify very much, with sustained winds of 70 knots (80 mph) as of the 5:00 am EDT bulletin from the Joint Typhoon Warning Center. Nida is unlikely to strengthen much further before making landfall on Tuesday night local time (midday Tuesday EDT). Even as a minimal typhoon, Nida may cause a fair bit of disruption for the seven million residents of the Hong Kong area. A number of destructive typhoons have struck in or near Hong Kong, including Wanda (1962), which produced wind gusts of 161 mph in Hong Kong Harbor.



Figure 6. Workers gather by street damage in Ellicott City, Md., on Sunday, July 31, 2016, after Saturday night's flooding. Historic, low-lying Ellicott City, Maryland, was ravaged by floodwaters that killed at least two people and caused devastating damage to homes and businesses, officials said. Image credit: Kevin Rector/The Baltimore Sun via AP.

Two killed as historic Maryland city ravaged by flash flooding
High water poured through the streets of historic Ellicott City, Maryland, on Saturday night in a dramatic flash flood (see YouTube embedded video at bottom) that killed at least two people and left a trail of damage and debris through the historic downtown area. Destruction in the city was reported to be severe, on par with the damage reported in the floods of 1868 and 1972 (the latter associated with Hurricane Agnes). While heavy rains were observed across much of the Washington/Baltimore area on Saturday evening (see Figure 7), the worst was focused near Ellicott City (Figure 8), where a rain gauge maintained by Howard County recorded the following amounts:

DURATION AMOUNT TIMEFRAME
----------------------------------------
1 minute ..0.20” - from 7:51pm-7:52pm
5 minutes ..0.80” - from 7:50pm-7:55pm
10 minutes..1.44” - from 7:50pm-8:00pm
15 minutes..2.04” - from 7:46pm-8:01pm
20 minutes..2.48” - from 7:44pm-8:04pm
30 minutes..3.16” - from 7:36pm-8:06pm
60 minutes..4.56” - from 7:30pm-8:30pm
90 minutes..5.52” - from 7:00pm-8:30pm
2 hours.....5.92” - from 6:45pm-8:45pm

Based on data from NOAA’s Atlas 14, which estimates recurrence intervals for heavy rains, the rainfall amounts within this two-hour interval have a less than one-in-a-thousand chance of occurring in any given year. Across many parts of the nation and globe, the heaviest precipitation events have become increasingly intense in recent years, with more water vapor available in an atmosphere warmed by increasing greenhouse gases paving the way for the potential for heavier rains when conditions are otherwise favorable. A Climate Central analysis shows that the Baltimore area has seen a particularly sharp increase in the heaviest downpours since the 1990s. See my post from July 30 for background on the Big Thompson Flood of July 31, 1976, a disaster that influenced our current warning and safety process for flash flooding.

I’ll be back with an update by Tuesday morning at the latest.

Bob Henson


Figure 7. 24-hour rainfall amounts through 8 am EDT Sunday, July 31, 2016, in the Washington-Baltimore area. Image credit: NWS DC/Baltimore.


Figure 8. Preliminary analysis of 3-hour rainfall totals between 6 and 9 pm EDT on Saturday, July 30, 2016, in the vicinity of Ellicott City, MD. Image credit: Greg Carbin, NOAA/WPC.


Video 1. Astounding video of the Elliott City, MD, flash flood of July 30, 2016. The videographer is looking down from a restaurant as cars float through the road! The video is a powerful reminder of the need to avoid being in vehicles during high water. Two feet of flowing water is enough to float most vehicles (even SUVs). Video credit: Nezacant.