Please login to view this content.
05/13/2019, 7:35 pm EDT

North Atlantic 2019 Tropical Cyclone Season Forecast Update

Highlight: Weak El Nino still expected – slightly suppresses activity. But ACE index forecast increases as 2 hurricanes affect the U.S. Executive Summary: The 2019 Climate Impact Company North Atlantic seasonal tropical cyclone outlook is updated. The seasonal outlook is essentially unchanged from the early April forecast indicating 12 tropical storms, 5 hurricanes and 3 intense hurricanes. The accumulated cyclone energy index increases to 97 caused by 2 projected hurricanes into the U.S. Based on an analog…the projected hurricane paths for the 2019 season Forecast summary and regional discussions: Climate Impact Company has updated the early April initial 2019 tropical cyclone forecast for the North Atlantic basin with “Update 1” issued May 13, 2019. The outlook has changed only slightly maintaining a projection of 12 tropical storms and 3 intense hurricanes while lowering the number of hurricanes from 6 to 5. The accumulated cyclone energy forecast is raised from 84 to 97. In summary, the seasonal activity is adjusted downward slightly but the intensity of the seasonal activity increases slightly as 2 (potential major hurricanes) are forecst to strike the U.S. The outlook is based on the slightly suppressing influence on seasonal activity by a persistent weak El Nino, warmer than normal western North Atlantic basin and concern regarding stalling storms once inland in the Southern U.S. due to an expected blocking high pressure area over the Northeast U.S. Gulf of Mexico: Despite a weak El Nino a normally active season is forecast for the Gulf of Mexico. The leading support to compensate El Nino for Gulf activity is a warmer than normal SSTA pattern in this region. Of concern is one projected major hurricane into Louisiana later in the season not dissimilar from Lili (2002). High pressure over the Northeast U.S. could cause a stalling storm that could produce unusually long-lasting heavy rain and high wind for Louisiana. Mexico: Westerly shear caused by weak El Nino should prevent hurricanes from striking Mexico. The position of the Bermuda High and the attendant steering currents bend most systems northward before they can extend far to the west to reach Mexico. East Coast Virginia and north: Blocking high pressure over the Northeast U.S. raises concern for any system offshore that slows down and may have a tendency to turn west. For now, hurricanes are projected to stay offshore the Coastal Northeast Corridor. East Coast south of Virginia: There is concern that at least one major hurricane will move inland in the Southeast U.S. likely into Florida similar to Frances (2004). Due to blocking high pressure over the Northeast U.S. this system could stall bringing excessive rainfall to Florida and Georgia. Caribbean: Each of the 5 projected hurricanes for the 2019 season will affect the islands in the northern Caribbean Sea mostly just-missing the islands to the north. A favored area for minimal storm pressure is just north of Haiti. El Nino southern oscillation: Well-known is the influence of the El Nino southern oscillation (ENSO) on North Atlantic seasonal tropical cyclone activity. When warm ENSO (El Nino) is present, the tropical East Pacific experiences warmer-than-normal ocean temperatures fueling abundant tropical cyclone activity. The latent heat released at the top of the weather atmosphere (troposphere) by these convective events causes an increase in westerly wind shear downwind across the tropical North Atlantic basin. The upper shear suppresses development of North Atlantic tropical cyclone activity. The opposite scenario occurs when cold ENSO (La Nina) is present as the East Pacific is void of tropical cyclone activity and westerly shear in the North Atlantic tropics is absent allowing tropical cyclone activity to flourish. The relationship between ENSO phase and North Atlantic tropical cyclone activity (since 1950) is vividly established (Fig. 1). Interestingly, and not until recently a neutral ENSO can be nearly as active as La Nina. Seasonal (and single storm) intensity is measured by accumulated cyclone energy (ACE) index which measures storm intensity on a 6-hourly basis. Not surprisingly, seasonal ACE index also follows the ENSO phase relationship to seasonal tropical cyclone activity (Fig. 2). Fig. 1: North Atlantic basin tropical cyclone season climatology. Fig. 2: North Atlantic basin ACE climatology according to phase of the El Nino southern oscillation. Changing ENSO: The somewhat straight-forward relationship between ENSO phase and seasonal tropical cyclone activity across the North Atlantic was established primarily due to the relationship to ENSO in the 1980’s and 1990’s inspired by the historic 1982-83 and 1997-98 El Nino episodes. However, the character of sea surface temperature anomalies (SSTA) outside of the equatorial East Pacific ENSO System has changed. Nowadays despite ENSO the global oceans tend to be warmer than normal (Fig. 3) whereas strong ENSO events such as the 1982-83 (and 1997-98) historic El Nino’s consolidated all the available global ocean heat mostly to the tropical East Pacific (Fig. 4). As a result, other aspects of the ocean-atmosphere environment, beyond just the tropics is reviewed and applied to seasonal forecasts. This new approach to seasonal forecasts of tropical cyclone activity is new and not as well-correlated as ENSO and seasonal activity. Also not well-understood and poorly reported is the tendency for ENSO-related climate regimes to fail. A recent example is the much-needed but absent Southern California rains expected with the 2015-16 El Nino to end drought. The warmer global oceans are present due to increased north-to-south heat transport primarily by the global oceanic thermohaline conveyor belt accelerated by constricting polar ice caps. Fig. 3: A weak El Nino trying to organize in December 2018 is surrounded by a warmer-than-normal global ocean. Fig. 4: El Nino 1982-83 inspired ENSO/climate relationships and numerical prediction of ENSO events. Note that in December 1982 all of the global oceanic warmth was consolidated in the tropical East Pacific.   ENSO 2019: Because of the increase in global oceanic warmth ENSO events have become more volatile, generally weaker and harder to predict. During the past year upper ocean heat in the equatorial Pacific to fuel El Nino has been plentiful (Fig. 5). However, the warming of the East Pacific Ocean has struggled to inspire the atmosphere to gain El Nino character (Fig. 6). The lack of coupling between the atmosphere and ocean warming is preventing El Nino from developing beyond the very weak signature occasionally present in recent months. Despite the reluctant El Nino of 2018-19 forecast models generally continue to forecast not only a robust El Nino but widespread warm SSTA across most of the oceans for the core (JUL/AUG/SEP) of the 2019 tropical cyclone season (Fig. 7). Diagnostics supporting the evolution of El Nino forecast are mixed as upper ocean heat in the equatorial Pacific is diminishing (Fig. 8) while a westerly wind burst in the equatorial West Pacific (to shut down trade winds and allow the surface to warm) is forming (Fig. 9). The Bureau of Meteorology/Australia tends to doubt El Nino in 2019 (Fig. 10) while NOAA favors El Nino (Fig. 11). Forecasts issued in early April by Climate Impact Company, Colorado State University and Tropical Storm Risk in the U.K. indicated near to slightly below normal seasonal tropical cyclone activity and ACE index (Table 1). In April odds favored at least a weak to moderate El Nino for the 2019 season suppressing seasonal activity although numbers were increased slightly toward normal due to the expected warm North Atlantic surface. So where does this all leave us for 2019? Forecast models are agreeable for lingering El Nino through most of meteorological summer (JUN/JUL/AUG). Models are split between neutral ENSO/weak El Nino and a weak-to-moderate El Nino for the core of the 2019 tropical cyclone season (AUG/SEP). Therefore maintaining the slightly suppressing effect on seasonal activity by weak El Nino is maintained in the updated forecast. A leading component to the warmer global oceans of the past 1-2 decades is the emergence of the warm phase of the Atlantic multi-decadal oscillation (+AMO) which is an Atlantic-born phenomenon but having global implications (increasing the transport of anomalous warm water in the tropics pole ward). The +AMO regime for tropical cyclone season is a key component of the forecast. A warm North Atlantic leads to more and stronger tropical cyclones. The NMME forecast for JUL/AUG/SEP 2019 is interesting. There is not a basin-wide warm signature as frequently seen the past 2 decades. Strong warming is forecast in the western North Atlantic while the deep tropics where tropical cyclones are born is normally warm. A cool pool south of Greenland caused by Greenland ice sheet melt persists through the summer season. The signature described indicates tropical cyclone activity in the deep tropics is completely dictated by the ENSO regime, a slightly suppressing effect. However, the warmer ocean surface in the subtropics indicates tropical cyclones are more likely to produce their strongest intensity north of the tropical latitudes. Of interest is the very strong warmth indicated off the Northeast U.S. which correlates to high pressure ridging in the upper atmosphere. This warm SSTA/high pressure ridging connection was present last year forcing Florence to slow down and stall in the Carolinas and also present in 2017 causing Harvey to slow down and stall in eastern Texas. Florence and Harvey were historic rainfall/flood producing events. The dynamics leading to those events is likely to regenerate this season. Fig. 5: NOAA/CPC upper ocean heat index for sectors of the equatorial Pacific Ocean for the past 1 year indicating persistent El Nino support. Fig. 6: Multivariate ENSO index identifies the reaction by the atmosphere to ENSO phase. Note that despite plenty of oceanic warming in the equatorial Pacific Ocean the past 12 months the atmosphere has been unable to produce and sustain a weak El Nino. Fig. 7: The leading global SSTA forecast model (NMME – North American Multi-Model Ensemble) indicates a weak El Nino and widespread warmth in the northern hemisphere outside of the tropics. The warm Pacific decadal oscillation and warmth off the U.S. East Coast offer clues as to where hurricanes will track in 2019.   Fig. 8: Upper ocean heat across the equatorial Pacific Ocean remains in the upper 100 meters although less warm than MAR/APR while strong cooling has emerged below 100 meters. The “trend” is away from warm water El Nino support. Fig. 9: A surge of westerly wind has developed in the equatorial West Pacific and is extending eastward which would shut down trade winds and allow El Nino oceanic warming to increase. Fig. 10-11: The Bureau of Meteorology/Australia weakens El Nino by August (left) while NOAA/CPC maintains El Nino in 2019 (right). CIC CSU TSR 30-Year Normal Last Year Tropical Cyclones 12 13 12 13.6 15 Hurricanes   6 5 5 7.0 8 Intense Hurricanes 3 2 2 3.0 2 ACE Index   84 80 81 115.7 129  Table 1: The early April seasonal forecasts of North Atlantic tropical cyclone activity by leading forecast providers. The updated 2019 forecast: The keys to the seasonal forecast update are the projected ENSO pattern and regional SSTA regimes adjacent to North America (Fig. 7). As previously discussed ENSO will remain in weak El Nino phase for the tropical cyclone season. The tropical North Atlantic in-between the Caribbean Sea and northwest coast of Africa (known as the main development region for hurricanes) is forecast near neutral SSTA (using tropical North Atlantic index). The basin-wide North Atlantic SSTA is likely marginally warm however the vast majority of this warmth is off the U.S. East Coast (using the Atlantic multi-decadal oscillation index). Finally, the northeast Pacific is forecast warmer than normal by the NMME SSTA forecast (using the Pacific decadal oscillation index). The forecast is based on analog years taken from the modern climatology. The modern climatology extends from the late 1990’s to present and is based on a long-term cycle shift of ENSO/PDO (to the cool phase), AMO (to the warm phase) and escalating CO2 rise leading to constricting polar ice caps affecting the north-to-south distribution of heat originating in the tropics. The modern climatology has caused an increase in tropical cyclone activity in the North Atlantic basin (Table 2). The analog years are NOT perfect but the best available within the defined climatology. Tropical Storms Hurricanes Intense Hurricanes 15-Year 15.3 7.5 3.3 20-Year 15.0 7.4 3.4 30-Year 13.6 7.0 3.0 50-Year 12.0 6.3 2.5 Table 2: North Atlantic basin tropical cyclone season climatology. The ENSO analog years are 2002 and 2015 and based on the new multivariate ENSO index (released earlier this year). Each analog approximates the 2019 ENSO regime in that El Nino is weak early in the year and forecast to become more organized and possibly stronger during the tropical cyclone season. This part of the analog forecast is least confident as while most dynamic models indicate El Nino the climate diagnostics are by no means confidently supportive of El Nino. Equally important is the North Atlantic basin SSTA regime. The projection is a slightly warmer AMO versus the TNA (index). The warmer AMO versus TNA relationship for the tropical cyclone season in the modern climatology is somewhat rare only present in 2004 and 2014. Finally, only 2002 has a reasonably similar northeast North Pacific SSTA pattern compared to the 2019 NMM SSTA forecast. Combining the tropical cyclone activity for each analog year yields an average amount of 11.6 tropical cyclones, 5.4 hurricanes and 2.8 intense hurricanes (Table 3). The averaged ACE index is 96.6. The forecast is slightly less active than the initial outlook issued in April although slightly more intense as defined by ACE index. Tropical Storms Hurricanes Intense Hurricanes ACE Index ENSO Analog 2002, 2015 11.5 4.0 2.0 63 AMO Analog 2002 12.0 4.0 2.0 67 PDO Analog 2004, 2014 11.5 7.5 4.0 145 Average (Forecast) 11.6 (12) 5.4 (5) 2.8 (3) 96.6 (97) Previous Forecast 12 6 3 84 Last Year   15 8 2 129 30-Year Climatology 13.6 7.0 3.0 115.7 20-Year Climatology 15.3 7.5 3.3 122.4 Table 3: The early April seasonal forecasts of North Atlantic tropical cyclone activity by leading forecast providers. Fig. 12-13: Based on analog years the 500 MB anomaly pattern for AUG/SEP/OCT 2019 (left) versus the 2017/2018 pattern (right). Of interest is the FEARED projected upper air pattern for AUG/SEP/OCT 2019 (Fig. 12) similar to 2017/2018 (Fig. 13). Combining the analog years the 500 MB anomaly pattern for AUG/SEP/OCT yields a blocking high pressure ridge over the Northeast U.S. Implied is above normal risk of a slowdown of tropical cyclones tracking inland over the U.S. East Coast and Gulf of Mexico region not dissimilar from Harvey (2017) and Florence (2018). Projected hurricane tracks: Historically, the North Atlantic tropical cyclone season has a tendency to produce on average about 3 persistent paths for the strongest systems each year. This tendency is due to the positioning of the North Atlantic subtropical ridge (“Bermuda high”) which produces the steering currents guiding the forward path of hurricanes. The Bermuda ridge is quasi-stationary therefore the steering currents produced to guide hurricane paths of travel do not change (much). Last year there were 3 primary paths of hurricanes around the relatively low latitude Bermuda High guiding hurricanes quickly north then northeastward (Fig. 14). The projected hurricane paths by Climate Impact Company prior to last season were generally correct showing above normal activity risk in the Gulf of Mexico and off the U.S. East Coast (Fig. 15). (Last year the May 31 CIC forecast predicted 4 hurricanes for 2018. The verified number of hurricanes was 8. The unexpected late season big warm-up of the North Atlantic tropics caused an up-tick in hurricanes.) Using the analog years to generate the 2019 forecast combined with the influence of the projected upper air pattern an estimate of where and what month the 5 hurricanes projected for this season will occur is indicated in Fig. 16. The forecast reveals 2 land-falling hurricane risks to the U.S. similar to Lili (striking Louisiana in October 2002) and Frances (striking Florida in September 2004). Each system could be a major hurricane with lowest minimal pressure between 930 and 940 MB just offshore Louisiana and in the southern Bahamas. Several hurricanes turn north around the Bermuda ridge but stay out-to-sea in the western and/or central North Atlantic basin. Due to the projected high pressure area over the Northeast U.S. there is concern that the projected hurricanes into Louisiana and Florida could slow down and stall in the Southeast U.S. (Please see regional implications in summary on Page 2). Fig. 14: Tropical cyclone activity for the 2018 season with 3 primary hurricane paths which identify the relatively low latitude axis of the Bermuda high pressure ridge and the faster than normal north to northeast turn of last year’s largest systems.   Fig. 15: Based on an analog the Climate Impact Company projection of an anticipated 4 hurricanes for the 2018 season. Fig. 16: Climate Impact Company seasonal forecast and projected hurricane paths for the 2019 North Atlantic tropical cyclone season. Verification: Climate Impact Company projection of seasonal tropical cyclone activity for 1999-2018 and forecast error of intense hurricanes, hurricanes and tropical storms for each season (Fig. 17). Fig. 17: Climate Impact Company early April seasonal forecast verification 1999-2019 is indicated (observed minus forecast).