RA I/TCC-22/Doc. 4.1, p 7
WORLD METEOROLOGICAL ORGANIZATION______
RA I TROPICAL CYCLONE COMMITTEE
FOR THE SOUTH-WEST INDIAN OCEAN
TWENTY-SECOND SESSION
MAHE, SEYCHELLES
25-29 SEPTEMBER 2017 / RA I/TCC-22/Doc. 4.1
(31.VIII.2017)
______
ITEM 4.1
Original: FRENCH
REVIEW OF THE 2015/2016 and 2016/2017 CYCLONE SEASONS
Report of RSMC La Réunion
(Submitted by RSMC LA REUNION)
RA I/TCC-22/Doc. 4.1, p 7
THE 2015-2016 CYCLONE SEASON
IN THE SOUTH-WEST INDIAN OCEAN
Five tropical storms and three cyclones developed over the south-west Indian Ocean basin during the 2015-2016 cyclone season, making it slightly less active than normal in terms of both the number of events and the number of days of disturbed activity (a low-pressure system at the tropical storm/cyclone stage over the basin for 45 days, of which 14 were cyclonic days, that is, a tropical cyclone was present over the region). However, this account is somewhat misleading, largely owing to the impact of a single system, namely cyclone Fantala. This was the last event of the season, and it alone made up for the significant lack of activity that had so far characterized the season. Without its (late) arrival, this season would have looked quite different, and 2015-2016 would have been judged one of the basin’s least active seasons, as the second least disturbed season of the last decade (and ranking sixth in the past thirty years).
However, in the end things turned out quite differently, and Fantala, the defining event of the season, was to blame, striking out of the blue at the very end of the season and turning everything upside down. Its 10-day lifespan as a cyclone was a record for the basin and sent the statistics for this 2015-2016 period shooting up; for example, the accumulated number of cyclonic days (that is, when a tropical cyclone was present over the basin) would otherwise have stood at a paltry total of four. What is more, the accumulated cyclone energy (ACE), calculated from the sum of the squares of maximum wind speeds every six hours, more than doubled, with Fantala alone accounting for over 60% of the seasonal total, and incidentally beating the basin’s record for an individual system.
Thus, Fantala completely skewed the picture of what was otherwise a relatively unproductive season. The early part of the season in particular was extremely sluggish. This was not because it started rather late, with the first tropical storm (Annabelle) emerging on 20 November, a fairly normal date (given that the median date for the start of the season is around mid-November – in other words, in one out of two years the first tropical storm of the season forms before or after 15 November). In fact, it was because the first four systems of the season were all tropical storms (two of which only just reached this stage), while the first tropical cyclone, Uriah, did not arrive until the fifth system formed on 17 February. This very late arrival made Uriah the third latest tropical cyclone in our south-west Indian Ocean region (after Farida, classified as a tropical cyclone on 26 February 1992, and Davina, which was classified as a tropical cyclone on 5 March 1999 and holds the record for the basin’s latest cyclone).
Thus, Uriah livened up what had until then been a rather dull season and heralded a second stage which was to prove distinctly more active, with the four events mentioned, three very intense cyclones, including one exceptional event, Fantala, a very intense tropical cyclone which numbers among the five most severe cyclones observed over the south-west Indian Ocean basin since the start of the satellite era. Thus, Fantala brought the season to an astonishing close. Besides the exceptional peak intensity reached, and after mid-April, making it the basin’s latest very severe tropical storm, it was exceptional in other respects: its wildly veering path saw it backtrack twice to follow almost the same path three times over a distance of almost 1,000 km, going back and forth in a totally unlikely manner that was clearly unprecedented in the basin’s history. Its record longevity at the cyclone stage (almost ten full days, briefly interrupted by a short break of about twelve hours) was mainly made possible by a trajectory maintaining it for a sustained period at low latitudes (remaining entirely north of the 14th parallel South). This made Fantala (only) the second very intense tropical cyclone observed in the basin north of the 10th parallel South (after Bento in November 2004) and, finally, as a result of the aforementioned record longevity, an individual ACE record-holder for the basin.
Before the start of this 2015-2016 season, calculations on its level of activity were going well. The southern spring was progressing under the influence of a major El Niño episode developing in the Pacific, one of the strongest of recent decades. Bearing in mind that over the past 40 years three of the basin’s least active cyclone seasons had occurred in years associated with significant El Niño episodes (including the major episodes of 1982-1983 and 1997-1998), it would have been reasonable to wonder if this pattern – significant El Niño episode coinciding with quiet south-west Indian Ocean cyclone season – might be repeated. In this context, the first projections of the numerical models for seasonal cyclone activity forecasts were awaited with particular interest. Surprise! Contrary to the overly hasty and simplistic conclusions, they projected a fairly normal cyclone season ahead with no significant features, and apparently unaffected by the powerful El Niño episode occurring in the Pacific.
The Indian Ocean basin is known to be the most complicated in terms of seasonal predictability, particularly since, unlike other cyclone basins such as the Atlantic and the North Pacific, or the Australian region, where the Southern Oscillation (El Niño/La Niña) has a proven influence, with a fairly clear response in terms of cyclone activity, the signs here are much less clear and are harder to interpret.
Recent studies have shown, among other things, that the uncertain response to El Niño episodes in our basin is probably linked to the latter’s contradictory influence on certain large-scale parameters that could affect cyclone activity: although the above-normal surface sea temperatures observed in our region during El Niño events (and therefore associated with increased available moisture) tend to encourage the development and strengthening of systems, the northward shift of the subtropical westerly jet stream, generally associated with an El Niño episode, causes westerly wind anomalies in the tropical zone, where cyclone events generally reach peak intensity. This can potentially be associated with an unfavourable increase in vertical wind sheer in the western sector, which is detrimental to the development and maintenance of cyclone events.
The complexity of the Indian Ocean is also due to the fact that this basin is subject to other internal factors which can, more or less directly, exert control over cyclone activity. Besides the famous Indian Ocean Dipole, which is to some extent the Indian Ocean’s own counterpart to El Niño, there is also the Subtropical Indian Ocean Dipole. These two dipoles are not independent of each other, and the less well-known subtropical dipole plays an important role in modulating the relative strength or weakness of subtropical high-pressure systems. This has an impact on cyclone activity, and in particular on the tracks of cyclone events. The combined effect of all these factors on complex interactions makes it particularly difficult to identify clear predictors of cyclone activity in the basin
In reality, this 2015-2016 season’s cyclone activity was ultimately near normal, bearing out the numerical models’ seasonal predictions. However, as has already been mentioned, this success was something of a fluke: their initial prediction was rescued from failure only by the arrival in extremis of Fantala, which, it should be pointed out, emerged in a somewhat different context to the rest of the season, given that by April the El Niño episode had already weakened significantly and was coming to an end.
Furthermore, it was perhaps not incidental that Fantala was exceptional in being the season’s only system to have followed a zonal trajectory. Before the start of the season, predominantly meridional trajectories had been anticipated, owing to the expected weakness of subtropical high pressure systems throughout the hot season (an anomaly related more to the influence of the Subtropical Dipole). This, in fact, proved to be the case, since the majority – six of the season’s eight systems – fell into this category.
As regards cyclogenesis regions, it was announced before the start of the season that the central area of the basin (between 60° East and 77° East) would be the most affected. This proved to be the case, since five of the season’s eight systems reached the stage of tropical storm in the 10° band situated between longitudes 68° East and 78° East. Consequently, the southern sector of the Chagos archipelago was extremely active, with all these systems passing through in succession in their early stages. Conversely, there was almost no activity in the western part of the basin: only Daya, a tropical storm of negligible proportions, formed in this region, while the Mozambique Canal remained totally unproductive.
As a result of the almost total absence of cyclogeneses over the western half of the basin and the predominantly meridional tracks, the cyclone risk was minimal for the inhabited regions, concentrated in the western part of the basin. With only two systems developing west of 65° East, just one of which, Fantala, was of any real significance, the season almost proved to be free of any impact on land. In the end though, this was not the case: there was indeed an impact. And what an impact! The unfortunate Farquhar Atoll was right in the path of Cyclone Fantala, just when it was about to reach peak intensity. This gave it the unhappy distinction of becoming the second region of the basin (inhabited in this case, since a few people had not yet been evacuated) to have been hit by a tropical cyclone categorized as very severe. Needless to say, the island was devastated.
This small Seychelles possession situated near the 10th parallel South, and very rarely affected by cyclones, would certainly not have been the first to spring to mind as the territory most likely to be hit by the only landfall of the season. An impact of such extreme intensity was even less likely, since it is in itself exceptionally rare for a cyclone categorized as very severe to be observed north of the 10th parallel South (just one precedent documented since the start of the satellite era, almost half a century ago, namely Cyclone Bento in 2004); And, as chance would have it, this occurred in the Farquhar region, despite it not being the area of the basin most affected by cyclones. In fact, Fantala was only the second mature tropical cyclone to strike this archipelago since the start of the satellite era, the other being Bondo almost 10 years ago in December 2006.
The other territories were much more fortunate, being totally spared by this season’s meteors. For example, the Mascarenes archipelago on the African continent enjoyed an almost total lack of activity in the Mozambique Canal. Even more noteworthy was Madagascar, where the calm period prevailing on the eastern coast of Grande Ile for the past four seasons continued. This area too had been spared the least impact since 2012 (Cyclone Giovanna).
______
THE 2016-2017 CYCLONE SEASON
IN THE SOUTH-WEST INDIAN OCEAN
The 2016-2017 cyclone season in the south-west Indian Ocean was both quiet (ranking among the five least active seasons of the last 30 years) and at the same time highly unusual in the way it developed. With the extremely early emergence in the middle of the southern winter of the first significant low-pressure system of the season (Abela), the season was off to a flying start. Furthermore, this system reached exceptional intensity, becoming the first severe tropical storm to be observed in the basin in July. It was followed by Bransby, the first subtropical depression ever recorded in October. However, the basin then fell into a period of total inertia lasting almost four months. Thus, it was well into the height of the southern summer before disturbed activity decided to wake up and shake off its lethargy, allowing the season’s third significant low-pressure system, and the very first of the hot season, to emerge finally at the beginning of February.
Such a long phase of inactivity at this time of year is quite exceptional, and unprecedented in recent history: since 1967 (the start of the satellite era in the basin half a century ago), the 2016-2017 cyclone season is the first not to experience any significant cyclonic episode (that is, a tropical storm or cyclone) whatsoever during the November-December-January quarter. We note in passing that a totally "blank" January (in terms of storms or cyclones) is in itself a quite remarkable event, although not unique, since there was a similar precedent (one and only!) in 2011. The slight difference was that January 2017 could justifiably be described as even less "active" than January 2011, which experienced a (brief) tropical depression and a tropical disturbance over the course of the month.
This lack of cyclonic events during the first half of the hot season was due largely to the establishment since the southern spring of abnormally and sustainably stable, dry conditions over a large area of the south tropical Indian Ocean, including the region generally most affected by cyclonic events. A significant positive Subtropical Indian Ocean Dipole (SIOD) phase emerged, and was probably the main contributing factor in this situation (see details in the insert below: "A broad-scale environment unfavourable to cyclogenesis").