97L a threat to become a tropical depression on Wednesday

A tropical wave (Invest 97L) near the east coast of Puerto Rico has become more organized overnight and is a threat to develop into a tropical depression as early as Wednesday. The disturbance has brought heavy rains of 8+ inches to Culebra and Vieques islands over the past day (Figure 1), and all of Puerto Rico and the Virgin Islands are under flash flood watches today. The storm could bring an additional 3 - 6 inches of rain to the islands over the next two days. The upper level low centered a few hundred miles north of the Dominican Republic is no longer bringing high levels of wind shear to 97L; wind shear has fallen to the moderate range, 10 - 20 knots, which may allow 97L to continue to develop today. Satellite images of 97L show a moderate area of disorganized thunderstorms, but no signs of a surface circulation, no low-level spiral banding, and no upper-level outflow. There is a large amount of dry air to the northwest of Puerto Rico that will interfere with development of 97L. Surface observations show only light winds over Puerto Rico, with no signs of a surface circulation. The Hurricane Hunters are on call to investigate 97L this afternoon, if necessary.


Figure 1. Total radar-estimated rainfall from Invest 97L.

Forecast for 97L
The storm is in a fairly straightforward steering current environment, and 97L should progress steadily to the west-northwest through Saturday. The rains from 97L's thunderstorms will bring the threat of flooding to the Dominican Republic today and Wednesday, and to Haiti on Wednesday and Thursday. Heavy rains from 97L will begin moving into eastern Cuba, the Turks and Caicos Islands, and the eastern Bahamas on Wednesday, and South Florida can expect heavy rains to arrive as early as Thursday night. We do have several models developing 97L into a tropical depression or tropical storm. The GFS and HWRF both take 97L to tropical storm status over the Bahamas by Thursday, with the storm then tracking over South Florida on Friday and entering the Gulf of Mexico on Saturday. The NOGAPS is similar, but portrays a weaker system. All of these models foresee a threat to the oil spill region by Saturday night or Sunday, with the storm making a second landfall somewhere between the Florida Panhandle and Louisiana. One factor potentially aiding the storm will be the Madden-Julian oscillation, which currently favors upward motion over the Gulf of Mexico and Caribbean. The Madden-Julian oscillation is a pattern of enhanced rainfall that travels along the Equator from west to east. The pattern has a wet phase with large-scale rising air and enhanced thunderstorm activity, followed by a dry phase with large-scale sinking air and suppressed thunderstorm activity. Each cycle lasts approximately 30 - 60 days. When the Madden-Julian oscillation is in its wet phase over a hurricane-prone region, the chances for tropical storm activity are greatly increased. Also in favor of development are the warm ocean temperatures of 29°C. The SHIPS model predicts shear will remain moderate, 10 - 20 knots, over the next 4 - 5 days, and I believe the primary detriment to development of 97L over the next two days will be the presence of dry, stable air in its path over the Bahamas, thanks to the upper-level low to the north of the Dominican Republic. NHC is giving 97L a 40% chance of developing into a tropical depression by Thursday, which is a reasonable forecast. I think there is a 60% chance 97L will eventually become Tropical Storm Bonnie, sometime in the next five days. Sudden rapid development today or on Wednesday is unlikely, due to the dry air over the Bahamas, and I put the odds of 97L making it to hurricane strength before reaching Florida at 10%. There is a better chance that 97L could attain hurricane strength in the Gulf of Mexico, perhaps 20%. These probabilities will depend heavily upon how long 97L (or Bonnie) spends over land or interacting with land over the next four days, which is very uncertain.

Time to cut the forecast numbers for the coming hurricane season?
Here are the number of Atlantic named storms, hurricanes, and intense hurricanes predicted by the various forecast groups in their late May or early June forecasts:

23 named storms: PSU statistical model
20 named storms: UKMET GloSea dynamical model
18.5 named storms, 11 hurricanes, 5 major hurricanes: NOAA hybrid statistical/dynamical model technique
18 named storms, 10 hurricanes, 5 intense hurricanes: CSU statistical model (Phil Klotzbach/Bill Gray)
17.7 named storms, 9.5 hurricanes, 4.4 intense hurricanes: Tropical Storm Risk (TSR), hybrid statistical/dynamical model technique (Note: TSR increased their numbers to 19.1, 10.4, and 4.8 with their July 6 forecast)
17 named storms, 10 hurricanes: FSU dynamical model
10 named storms, 6 hurricanes, 2 intense hurricanes: climatology

The group forecasting the lowest activity was the Florida State University group led by Dr. Tim LaRow. They use a new dynamical forecast model called COAPS, which is funded by a 5-year, $6.2 million grant from NOAA. This year's June forecast by the COAPS model called for 17 named storms and 10 hurricanes. However, Dr. LaRow emailed me yesterday to say that the COAPS model is now calling for reduced activity. Using the state of the atmosphere and ocean as of July 15, a new run of the COAPS model was performed over the weekend. The new forecast is now calling for two fewer hurricanes--a total of 15 named storms and 8 hurricanes (including Alex.) The COAPS model generated an "ensemble" of five different forecasts, done by varying the initial sea surface temperatures by a few percent at the beginning of the model run. These five forecasts came up with a range of 12 - 16 named storms (including Alex), and 7 - 10 hurricanes. It will be interesting to see when CSU issues its August 4 forecast if they also cut their numbers. With only one named storm (Alex) thus far this year, it's getting pretty hard to have a season with 19 or 20 named storms. Only four hurricane seasons since 1851 have had as many as nineteen named storms. These four seasons--1887, 1933, 1995, and 2005--all had at least three named storms by July 20.

"Hurricane Haven" airing again this afternoon
Tune into another airing of my live Internet radio show, "Hurricane Haven", at 4pm EDT today. Listeners will be able to call in and ask questions. The call in number is 415-983-2634, or you can post a question in the comments area on my blog during the show. You can also email the questions to me today before the show: jmasters@wunderground.com. Be sure to include "Hurricane Haven question" in the subject line. Some topics I'll cover today on the show:

1) Invest 97
2) A look ahead at the coming two weeks

Today's show will be about 45 minutes, and you can tune in at http://www.wunderground.com/wxradio/wubroadcast.h tml. The show will be recorded and stored as a podcast.

Jeff Masters