Yale Climate Connections - Jeff Masters Weather Blog

Less frenzied, but still above average » Yale Climate Connections


Two of the most closely watched outlooks for the Atlantic hurricane season have brought down their predicted levels of activity a notch or two. However, both are still predicting the season to end up busier than the 30-year average. The latest updates from NOAA and Colorado State University were released on Thursday, August 4.

Although 2022 has not yet spawned any Atlantic hurricanes, the season typically picks up in August and peaks in September, and several factors that augur an active season still apply. Most notably, a weak but persistent La Niña event continues across the eastern tropical Pacific. Odds now favor La Niña (which supports Atlantic hurricane development) remaining in place throughout the rest of the year.

NOAA’s revised Atlantic hurricane outlook is only slightly less worrisome than its initial forecast for the 2022 season, which was issued on May 24. The probability of a season that’s more active than the 1991-2020 average is now 60%, compared to 65% in the May outlook.

Figure 1. NOAA’s 2022 Atlantic hurricane season probability and number of named storms, as updated on August 4. (Image credit: NOAA)

Here’s how other revisions to the NOAA outlook stack up, including accumulated cyclone energy (ACE), which is calculated based on the intensity and longevity – but not the size – of each tropical cyclone.

May 24 August 4
Named storms 14-21 14-20
Hurricanes 6-10 6-10
Major hurricanes 3-6 3-5
ACE 115-120% [no update]

“Although it has been a relatively slow start to hurricane season, with no major storms developing in the Atlantic, this is not unusual and we therefore cannot afford to let our guard down,” said FEMA administrator Deanne Criswell. “This is especially important as we enter peak hurricane season—the next Ida or Sandy could still be lying in wait.”

Colorado State University has also tempered its outlook for a high-octane 2022 season. Here are the numbers issued from the CSU forecast group so far this year, in both capsule and more detailed form.

 July 7 August 4
Named storms 20 18
Hurricanes 10 8
Major hurricanes 5 4
ACE 180 150
Figure 2. Evolution of CSU’s Atlantic hurricane outlook for 2022 as of August 4. *The total forecast for named storms as of August 4 includes the three already observed this year: Alex, Bonnie, and Colin. (Image credit: CSU, with background image courtesy NOAA.)

“One of the reasons for the reduction in our forecast was due to anomalous cooling in the subtropical eastern and central Atlantic,” said the CSU team. “Cooler-than-normal [sea surface temperatures, or SSTs] in this region have been associated with enhanced wavebreaking into the tropics, which could potentially somewhat counteract the anticipated reduction in wind shear associated with La Niña.”

Even so, the group noted, “the current SST pattern is tracking between SSTs typically experienced in above-average Atlantic hurricane seasons and hyperactive seasons of the past 40 years.”

The CSU August outlook is based largely on a statistical scheme – newly updated this year – that takes into account three factors, as observed from mid-June to late July, that tend to correlate with Atlantic tropical cyclone activity from August onward:

  • Trade wind strength over the Caribbean (slightly weaker than usual this past July, so slightly favorable for tropical cyclones)
  • SSTs over the eastern subtropical North Atlantic (cooler than average, so unfavorable)
  • Westerly upper-level wind shear over tropical North Africa (slightly below average this year, so slightly favorable)

CSU has also teamed with the Barcelona Supercomputing Center to evaluate August-September forecasts of wind shear and SSTs for the tropical Atlantic and Caribbean produced by three modeling centers: the European Centre for Medium-Range Weather Forecasts (ECMWF), the UK Met Office (UKMET), and the Japan Meteorological Agency (JMA). Output from these models lends further support to the case for a busier-than-usual Atlantic season.

The years identified by CSU as analogs for the 2022 season are 1999, 2000, 2011, and 2021. All of these had above-average activity, with ACE values ranging from 119 to 177 and named-storm counts ranging from 12 to 21. All of the analog years had a La Niña event firmly in place (see Figure 3).

While there is little to no skill in seasonal hurricane forecasts issued early in the calendar year, CSU’s August forecasts have demonstrated considerable skill.

Figure 3. Average sea surface temperature anomalies (August-October) for the four analog years cited by CSU, showing a classic composite signature of cool La Niña conditions in the eastern tropical Pacific. (Image credit: CSU)

The Atlantic is dragging its heels

One reason for the slightly less bullish updates from NOAA and CSU is the relatively tepid pace of the 2022 Atlantic season thus far, at least compared to the last few seasons. Only three short-lived named storms have occurred this year (Alex, Bonnie, and Colin). None has reached hurricane strength, and none had major impacts during and after landfall. (Technically, Colin never made landfall; it developed just inland over South Carolina and dissipated over North Carolina just 24 hours later.)

NOAA’s National Hurricane Center (NHC) reported on August 1 that the seasonal total of ACE to date was running slightly below the 1991-20 mean. That’s an understatement: as calculated by CSU, the total North Atlantic ACE as of August 4 was a mere 2.8 units, compared to the 1991-2020 average up to this point of 10.8. (These totals can be especially variable so early in the season.)

The last time the first week of August came without the first Atlantic hurricane of the year having formed was in 2017, when Franklin became a hurricane on August 9 in the Bay of Campeche.

As of August 4, the Atlantic had gone more than a month since Colin dissipated on July 3, and the basin remained tranquil. No tropical waves were expected to develop over the next five days (per the NHC Tropical Weather Outlook), and there was no consistent signal in model ensembles of any development over the next week-plus. The latest Global Tropical Hazards and Benefits Outlook, issued by NOAA’s Climate Prediction Center on August 2, has no risk areas for Atlantic development highlighted for either of the weeks of August 2-9 or of August 10-16.

“If we make it until August 13 with no named storm formations (fairly likely), it’ll be the first time since 1999 that we went from July 4 to August 13 with no named storm formations,” said CSU’s Philip Klotzbach in a Twitter direct message. He added: “Of course, 1999 ended up a hyperactive season, so there’s that!” After a gap between Arlene, which formed on June 12, and Bret, which formed on August 19, “we had four named storm formations between the 19th and 24th of August that year, and the season was off and running.”

UK Met Office trims its outlook

Similar to CSU, the August 2 forecast update from the UK Met Office is calling for an active Atlantic hurricane hurricane season, but less of one than its May outlook did. The office is now calling for 16 named storms (down from 18), 6 hurricanes (down from 9), 4 major hurricanes (down from 6), and an Accumulated Cyclone Energy (ACE) of 151 (down from 176). Interestingly, the Met Office outlook is now for fewer hurricanes than the 1991-2020 average of 7; the other three variables are all still projected to come in above their long-term average.

Most of the seasonal hurricane forecast groups include at least some statistical element in their work, but the UK Met Office leans entirely on UKMET dynamical modeling. Its forecast is made using GloSea6, a high-resolution ensemble prediction system that draws on the office’s HadGEM3 family of climate prediction models. Each week GloSea6 is used to generate a 42-member ensemble seasonal forecast for the next six months, blending and bias-correcting the ensemble runs from the prior three weeks.

TSR’s July update: Above-average season, with 18 named storms

British private forecasting firm Tropical Storm Risk, Inc. (TSR) won’t be updating its seasonal outlook until Tuesday, August 9, but on July 5 it bumped up the numbers released in its May outlook. The outlook increased to 9 hurricanes (versus the forecast of 8 issued in May) and an Accumulated Cyclone Energy (ACE) of 150 (compared to 138 in the May outlook). The long-term averages for the past 72 years are 12.2 named storms, 6.4 hurricanes, 2.7 intense hurricanes, and an ACE of 105.

Over the period from 2003 to 2021, TSR’s April forecasts have averaged only about 10% better (mean square skill score) than a 10-year climatology for ACE. By August, the skill increases to about 40-50%.

Website visitors can comment on “Eye on the Storm” posts (see comments policy below). Sign up to receive notices of new postings here.



Source link