Will the 2013 storm season see a furious finish after a quiet start?
After a very slow start, tropical cyclone activity has picked up in the Atlantic over the past few days.
September 11, 2013 | By Michael Watkins, Stormpulse contributor
Hurricane Humberto is spinning over the east Atlantic, Tropical Storm Gabrielle is near Bermuda, and a third system has a high chance of development in the southwest Gulf of Mexico as we move toward the weekend.
Here are a few quick hits on what to look for as we move through the remainder of September:
- We have reached the climatological peak of the Atlantic hurricane season: September 10th, historically, has seen more hurricane activity than any other day of the year.
- September is also the busiest month of the season.
- The Madden-Julian Oscillation, a global circulation pattern in the deep tropics, is in a favorable phase for Atlantic hurricane development.
- Atmospheric instability, which had been below normal for this time of year, has increased across the tropics over the last several days.
It is still very possible we could still see an active September/October pattern develop.
On Wednesday morning, Hurricane Humberto became the first hurricane in the tropical Atlantic so far in the 2013 season, narrowly averting the aircraft reconnaissance era record of the latest date for the first hurricane forming in the Atlantic basin. This “record” was set by the 2002 Atlantic season, which did not see a hurricane develop until 8:00 AM EDT on September 11th. After Humberto became a hurricane at 5:00 AM EDT this morning (just three hours shy of the 2002 mark), 2013 observed the second-latest “first hurricane” going all the way back to 1944.
It is interesting to note the 2002 season ultimately produced nine hurricanes and four major hurricanes (winds of at least 115 MPH). October, not September, is the peak month for Florida peninsula hurricane impact. These facts tell us a late start doesn’t necessarily imply a quiet season. The big question is: will the 2013 season see a furious finish after such a quiet start?
This is the Time of Year to Watch the Tropics Closely
September and October are the busiest months of the Atlantic season for a reason. Usually, calm upper level winds and very warm sea surface temperatures make tropical development possible almost any place in the basin. Seemingly benign systems can fire up quickly, even if not previously detected by forecast models. Subtle differences in the steering pattern could make the difference between a Florida or east coast landfall and a harmless recurve out to sea.
Forecast models can also change quickly. For example, mid-range global models have been forecasting a large trough of low pressure to stay in place off of the east coast of the United States for the next 7-10 days. A persistent trough of low pressure would work to deflect systems away from the U.S. if they approached from the east out of the deep tropics. However, the most recent model runs (as of Wednesday morning) are forecasting a slightly weaker trough and a slightly stronger area of high pressure in the western Atlantic.
Seasonal forecasters had predicted this to be a busy season even though it looks like we may observe an average, or even below average season in terms of total tropical activity. Then again, the hyper-active 2005 season was similarly under-forecast through the August updates that year.
1992 was an exceptionally quiet season overall. Residents of south Florida, directly impacted by a ferocious category five Hurricane Andrew, will tell you 1992 was anything but a quiet season. One storm can make a huge difference, even if we can’t currently see it on the Stormpulse map.
Currently, in the Tropics
Environmental conditions around Humberto are favorable for continued intensification through Wednesday, and Humberto has a very slight chance to become the first major hurricane of the season, although that appears very unlikely at this time.
Global models show Humberto moving mostly northward through the next few days, with a turn back to the west-northwest by the weekend. However, because the storm will be at such a high latitude when it moves back to the west, it has virtually no chance of reaching the United States at this point.
Increasing vertical wind shear and cooler sea surface temperatures will also work against Humberto maintaining hurricane status beyond a few days time, and the forecast is for the system gradually lose strength and become a tropical storm by the end of the weekend.
Looking further out into next week, forecast models suggest Humberto could restrengthen over the central Atlantic while moving very slowly. This means Humberto could be sticking around for an extended period of time.
Tropical Storm Gabrielle regenerated just south of Bermuda on Tuesday morning. This evolution was somewhat well-forecast by the GFS computer model, which has been suggesting a closed low of some sort would redevelop for several days in a row. Unlike Humberto, Gabrielle is embedded in a less-conducive environment for intensification and is not expected to become a hurricane before moving quickly to the north and eventually northeast. It may make landfall in Canada later this week as a weak tropical storm.
Finally, an area of low pressure, currently labeled Invest 93L is crossing the Yucatan Peninsula and will soon move into the Bay of Campeche in the extreme southwest Gulf of Mexico. Conditions there should be favorable for development, and there is a high chance that a tropical storm will form by Friday.
Global models were previously weakening the ridge of high pressure over the Midwest as we move into the weekend. In turn, the models were predicting a weak steering environment and suggesting 93L could move further north than Fernand and Tropical Depression Eight did earlier in the season. However, the most recent overnight model runs are showing a stronger ridge of high pressure over the central US in the extended period. This is good news for the United States, but bad news for Veracruz and other places in Mexico previously impacted by tropical storm Fernand. At a minimum, more flooding rains could be on the way for Mexico. Hopefully, some of the moisture will move far enough north to bring some much-needed moisture into south-central Texas, which would be great news for the drought-stricken area.
It is too early to speculate on how strong this system could become. There is considerable uncertainty because development is forecast to occur several days from now, when forecast skill and resolution are less reliable. This picture will come into better focus in a couple of days when/if the center of this system reaches the warm water of the southwest Gulf of Mexico.
A Look Back at August
August came and went across the Atlantic ocean without a hurricane developing within the month, despite predictions for a busy season from most of the experts including NOAA and Colorado State University. In fact, last month was the quietest August measured in the Atlantic since 1997. Incidentally, the uneventful 1997 season was not a surprise, a massive El Nino in the east and central Pacific shut down the Atlantic basin. There has been no such El Nino this year, and experts simply do not know why conditions have been so unfavorable for development so far.
Tropical Storm Erin formed from a tropical wave on August 12th in the far eastern Atlantic. It spun up into a tropical storm but struggled against strong upper-level winds, which ultimately separated the convective activity from the low-level circulation. Very dry, stable air prevented the development of new thunderstorms near the low-level center. Erin dissipated on August 18th, although the remnant swirl was evident in satellite imagery for a few more days as it moved westward in the low-level trade winds.
Tropical Storm Fernand spun up rapidly in the extreme southern Bay of Campeche on August 23rd before making landfall near Veracruz, Mexico. It likely would have reached hurricane strength if it had another 24 hours over water.
Why Has This Season Been Slow So Far?
Between strong directional shear, stable air in the atmosphere and very dry pockets Saharan air moving across the Atlantic, conditions have conspired against tropical cyclone formation so far this season.
From year to year, the Atlantic is the most variable of all of the major hurricane producing basins in the world. Sea surface temperatures, background surface pressures, low/mid level humidity and upper-level winds change from season to season. Yearly output can also be impacted by development close to land masses, just like Fernand earlier this season. Had it formed further over water, it likely would have added more Accumulated Cyclone Energy (ACE) to the seasonal total (and likely would have become a hurricane).
Accumulated Cyclone Energy (ACE) is one measurement of the total energy expended by a tropical cyclone. It’s a more useful than counting named storms or hurricanes because it accounts for the energy expended by a storm over time. A hurricane could last a day or a week, but counting named storms would give a day and a week equal weight. Looking at the ACE reveals a much better read on which storm expended more energy.
The total ACE for a storm or season is calculated by squaring the wind speed (in knots) every six hours, and summing those totals (then dividing by 10,000 to get to a relatively manageable number).
The below graph shows the seasonal ACE for each season since 1995 (blue bars) on one axis, compared with the corresponding 3 month average of the AMO (Atlantic Multidecadal Oscillation) from May, June and July for each season (orange line). August AMO values are not available at this time.
The AMO is an index of relative Atlantic Sea Surface Temperatures. Warm (positive) AMO seasons tend to produce lots of Atlantic hurricanes, and cold (negative) AMO seasons produce less hurricanes. The Atlantic has been in the “warm AMO phase” since 1995, but there is still variability from year to year within this metric.
Notice the AMO average for May, June and July (MJJ) of 2013 has dropped relative to 2012 (+0.14), a little more than half of the previous year. Looking back within the recent data set, there appears to be a loose relationship between the AMO and corresponding seasonal ACE, but the relationship is not absolute. However, over recent seasons like 2007 and 2009, we’ve seen a drop-off in seasonal ACE vs. the previous year, with the May/June/July index has been in a similar range to this year.
A closing thought: The 1988 season also had a very slow start, without any significant development in August. In September that year, hurricane Gilbert made history with the lowest pressure ever recorded in a hurricane as it spun through the western Caribbean and ultimately into Gulf of Mexico. The season isn’t over, and it won’t be over until November 30th. Anything can happen between now and then.
Michael Watkins is the founder of Hurricane Analytics and a Stormpulse contributor.