Tropical weather analysis - July 23, 2010

Bonnie

Bonnie weakened to a tropical depression while traversing southern Florida this afternoon. Thelast of the visible satellite images place the center at 26N 83W, over the eastern Gulf of Mexico. The National Hurricane Center had it at 26.4N 82.5W just a little over an hour ago, so this motion merely serves to reflect an extrapolation of that position. In any case, the aforementioned imagery also reveals a highly disorganized tropical cyclone, with the small area of deep convection displaced to the north and northwest of the poorly defined surface center, which appears to be located about 40-50 miles southeast of the aforementioned, shapeless convective activity.

This extreme disorganization is due to strong southeasterly vertical wind shear in excess of 30 kt. This shear is associated with a large, persistent, and powerful upper-level low centered near 27N 93W, and a large and powerful upper-level ridge centered across the mid-Atlantic states. Bottom line? Bonnie is embedded within a very hostile environment for intensification, with water vapor imagery depicting a large swath of dry air to the west. This dry air is gradually being entrained into the poorly defined surface center due to clockwise flow around the aforementioned ridge. Given the noticeable lack of upper-level divergence and low-level convergence associated with the cyclone, along with the aforementioned dry air and a continued hostile upper wind environment, I am not expecting Bonnie to be able to intensify. In fact, the most likely scenario is for the system to gradually weaken, ultimately opening up into a disorganized trough of low pressure prior to moving ashore along the northern Gulf Coast in the next 24-36 hours.

The track forecast is rather straightforward. Water vapor imagery depicts a strong and solid southeasterly flow impinging upon the cyclone. Embedded within general southeasterly flow between the upper low and the aforementioned anticyclone, Bonnie should continue moving WNW for the next 12 hours or so, after which point a turn to the NW should occur. By this time, a weakness should become apparent within the western Atlantic subtropical ridge as a mid- to upper-level trough and its associated cold front will be nudged a bit to the southeast by a well-defined upper-level low currently moving across northwestern North Dakota. This should effectively erode the ridge and allow the cyclone to turn more poleward during this time. That motion should continue until well after landfall occurs along the northern Gulf Coast in about 24-30 hours, and the cyclone could still be moving NW as it enters eastern or central Arkansas in about 60 hours.

The intensity forecast is also relatively simple. The 18z GFS indicates that the current southeasterly flow will gradually veer to the northeast, associated with the aforementioned ridge. There could be a brief relaxation of the shear over the next six hours, as indicated by the GFS, and then again in 30 hours as the system approaches the northern Gulf Coast. I'm in general agreement with the NHC's 5:00 PM EDT forecast track, albeit slightly farther to the right when landfall occurs. I expect that landfall will occur along the Louisiana/Mississippi border in just around 30 hours.

Watches and warnings

A tropical storm warning is in effect from:

- Destin, Florida to Morgan City, Louisiana, including New Orleans and Lake Pontchartrain

However, I imagine that these warnings were given as a course of least regret, just in case the cyclone unexpectedly strengthens due to an unforeseen relaxation of the vertical shear. I am not actually expecting tropical storm force winds anywhere within the warning area. I still expect Bonnie to open into an elongated trough within the next 12 hours, given the hostile environment.

Bonnie's main legacy will be 4 to 6 foot waves atop the oil spill region. A persistent fetch of SE to E winds will begin to blow across that region beginning in about 12 hours. This could potentially bring oil into the already devastated Louisiana marshlands, along with other areas of coast that have already experienced oil on their shores.

Elsewhere

There are no other threat areas to discuss in the tropics at the moment. Former Invest 98L moved inland near Tampico early this morning. A weak area of low pressure could still form near the Cape Verde Islands on Sunday, as indicated by both the GFS and the ECMWF, but both, particularly the latter, have become far less aggressive in this forecast than they were previously. The ECMWF brings this system off Africa at a higher latitude than the GFS does. I am inclined to agree with the solution of the GFS, given current convective trends over western Africa. Though upper-level winds favor development in this area, a dry environment awaits the system in the eastern Atlantic, which will act to inhibit any significant development. In fact, this is probably why the aforementioned models do not strengthen this feature, and quickly lose it within 12 hours of exiting the coast.

It should be noted that we are forecast, as per the EWP model, to enter a rather dry period across the Atlantic basin as the downward motion pulse of the MJO -- a 30 to 60 day eastward propagation of moisture originating over the Indian Ocean. The upward MJO pulse tends to favor greater atmospheric instability, and hence, enhances convection. The downward motion tends to do the exact opposite. However, tropical cyclogenesis can still occur during downward MJO pulses -- for instance, 2005's Harvey and Irene all formed during the downward MJO, the latter of which ultimately became a Category 2 hurricane that threatened the Carolinas.


However, given that the upward MJO has been rather hesitant to leave our area throughout most of the season, I rather am skeptical of the EWP's forecast. We will see, though.

On the other hand, the CFS model foresees a far less robust and long-lasting downward pulse.

Finally, the GFS is forecasting a rather significant downward MJO pulse, but an extrapolation of this forecast points to a general lessening of this motion by the second week of August. I rather like the forecast from the CFS, given current convective trends across the basin. However, given that both the EWP and the GFS are forecasting a significant downward pulse, I cannot simply ignore them. Hence, I will split the difference and forecast a downward MJO pulse for the next two weeks, after which point we should return to a neutral MJO pulse, followed by an upward motion pulse sometime during the second week of August.

Remember, just because we have only seen one significant tropical cyclone this year: Alex, doesn't mean that the season is a bust. Please do not be lulled into a false sense of security. In 1998, the most similar analog I can think of, the second storm didn't develop until August 19, and we ended up having 14 named storms, 10 hurricanes, and 3 major hurricanes that season, including notable ones like Bonnie, Georges, and Mitch.

The season isn't over by a longshot, and I still fully anticipate an above average season, even if we don't necessarily end up getting the 18 storms that CSU predicts.