Average 20 year old weather nerd. Plymouth State University Meteorology, Class of 2018. NOAA Hollings scholar. Summer 2016 intern at NWS Boston.
By: MAweatherboy1, 9:03 PM GMT on March 17, 2015
Happy Spring! It’s that time of year again: time for predictions to start being made for the upcoming 2015 Atlantic Hurricane Season, and that is exactly what I’m doing in this entry. Without further ado, I’ll jump right in.
The 2015 Atlantic season will begin on June 1 of this year and run through November 30, with a climatological peak from mid-August through late September, when the vast majority of tropical storms and hurricanes have formed throughout history. There are a lot of interesting aspects about this upcoming season, and a lot of factors that are going into the forecast. Seasonal hurricane forecasting, and for that matter any seasonal forecasting in meteorology, is a very inexact science. It’s hard enough to forecast accurately for 5 days out. 5 months is a whole different animal. So, with that, I offer a disclaimer that there are no guarantees at all with this forecast, and this forecast by me does not represent the opinion of any professional agency. However, through analyzing trends in key atmospheric and oceanic variables, it is usually possible to come to a general understanding of how the upcoming season will play out.
Since 1995, the Atlantic basin has been in what has been labeled an “active era” of tropical cyclone activity. We have seen many of the most active hurricane seasons and strongest hurricanes on record during this period. While this is partly due to technological advances, it can also be largely traced to one of the important variables we look at in determining Atlantic cyclone activity. That would be the Atlantic Multidecadal Oscillation (AMO). The AMO refers to a pattern of sea surface temperature variability in the Atlantic basin, which, as its name suggests, oscillates on a time scale of decades. In 1995, the AMO flipped to its “positive” phase after around 30 years in the negative phase, a period during which quiet hurricane seasons dominated the Atlantic. Lately, however, things with the AMO have become a bit muddled. We know little about the AMO, but our research and historical data suggests the index should stay in the positive phase for at least another 5 years, and perhaps as much as 10-15 years. But in the past two years, we’ve seen the positive AMO wane, and it has now turned negative. This doesn’t necessarily mean the positive phase is over; brief periods of multidecadal indices turning out of their predominant phase pattern are not unusual. However, some have argued, and this idea may have some merit, that we are in fact on the verge of an AMO flip. This would have major implications on hurricane seasons for many years to come. Let’s get back to this year, however. Regardless of whether the AMO flip is for real, it is true that at least for this year, the AMO is likely going to be neutral to negative, which is not good for tropical cyclone formation. The current Atlantic sea surface temperature anomalies can be seen below. We’re still half a year away from the peak of hurricane season, so these values could change, and short term fluctuations are a certainty, but this is about what I expect the Atlantic SST anomalies to look like for most of this season, in accordance with the near neutral AMO.
Figure 1: North Atlantic SST anomalies. The general theme is well above normal SSTs in the subtropics and near to below normal SSTs in the Atlantic MDR. Short term fluctuations will occur, however, and should not be given much thought. (Credit: Levi Cowan.)
Note the expansive region of warmer than normal water in the subtropical latitudes of the central Atlantic. This is a negative for traditional Atlantic cyclone formation, although it does increase the likelihood of storms forming in those subtropical regions. Also of note is the near to cooler than normal water off the west coast of Africa, near the Cape Verde islands, and extending several hundred miles west of there. This is also a negative for cyclone formation; this region is known as the Atlantic’s Main Development Region (MDR). If waters in this region are cool, long track, “Cape Verde” style hurricanes become less likely. Also of note, the Caribbean Sea and Gulf of Mexico are warmer than average. This is a positive for cyclone formation in these regions, which can be of particular concern due to the natural proximity of large, populated landmasses to those bodies of water. More on that later. The main point of this section is that SST anomalies in the tropical Atlantic are generally unfavorable for tropical cyclone formation, which would favor a quieter hurricane season in 2015.
Another important factor we look at in attempting to determine the prospects for tropical activity in the approaching season is ENSO, the El Nino Southern Oscillation. These are two separate indices, but are closely connected, hence the common joining of their names. The terms El Nino, and its opposite form, La Nina, refer to a pattern of sea surface temperature anomalies in the central and eastern equatorial Pacific Ocean. El Nino refers to anomalous warming of these waters, while La Nina is anomalous cooling. Since early last year, there has been a continuing saga of El Nino’s attempt to develop. Last winter and spring, many indicators, as well as climate models, pointed towards a potentially moderate to strong El Nino developing for the summer and fall of 2014. This never happened; despite repeated attempts by the Pacific to warm, no sustained anomalous warming (greater than or equal to a 0.5C anomaly for three consecutive months) ever occurred. However, towards the end of 2014 and into early 2015, just enough warming did develop that the Climate Prediction Center declared an official weak El Nino. Now, for the second year in a row, several important climate indicators as well as climate models are signaling that El Nino will strengthen into the summer of 2015. Very anomalous warm subsurface water has been detected moving into the important ENSO regions, and the surfacing of this warm water could lead to large SST spikes. In addition, the Southern Oscillation Index, a Pacific pressure pattern (specifically the difference in sea level pressure between the island of Tahiti and Darwin, Australia), has been trending negative, which is a usual indication of El Nino development. El Nino can have a major impact on the Atlantic hurricane season. Anomalous warm water in the East Pacific generally leads to enhanced convection in that region. This tends to create a suppressed region over the western Atlantic (Caribbean Sea), and this enhanced Pacific convection also tends to produce outflow shear that can keep western Atlantic convection disorganized. Consequently, El Nino years tend to produce below normal tropical activity in the Atlantic basin. El Nino is not easy to predict, but based on the signals we’re seeing and the model projections, I would expect El Nino to reach at least moderate levels by the peak months of the hurricane season, so this is another factor favoring below normal Atlantic activity.
Figure 2: Nino Region 3.4 anomalies. This is the "critical region" for determining ENSO state. This chart is very typical of a weak El Nino, with the three month average anomaly around 0.5C. (Credit: Levi Cowan.)
Figure 3: Equatorial sub-surface water temperature anomalies. Note the expansive region of very warm water that may produce a more moderate to strong El Nino by mid summer. (Credit: NOAA CPC.)
Factors like the AMO and ENSO are long term, multi-year or multi-decadal phenomena. They can produce a general mold for an Atlantic hurricane season, but the specifics of the season will often depend on more short term factors. Naturally, these are close to impossible to predict months in advance of the season. Wind shear, for example, can make the difference between a cluster of thunderstorms being ripped apart or developing into a hurricane. You can never predict what something like shear will be like in a certain place at a certain time months in advance. However, we have clues that can point us in certain directions on these short term factors. The likely El Nino, for instance, should lead to a higher baseline level of shear basin-wide, particularly in the Caribbean and Gulf of Mexico. Instability in the Atlantic is also an important factor, and a lack of vertical instability over the main development region has been a major reason why the past two seasons have been so quiet. Once again, thanks in part to the cooler than normal east Atlantic waters, it appears very likely Atlantic instability will run below average again this year. We will also have to watch for the extent of Saharan Air Layer (SAL) outbreaks during the early to middle portion of the season. These large plumes of dust and dry air that emerge from the Sahara can create a dry, stable layer of air in the Atlantic that inhibits tropical cyclone formation. The past two years have seen several significant SAL events that have kept a lid on cyclone formation through mid-late August.
Figure 4: Tropical Atlantic vertical instability, which is well below normal for this time of year and likely to remain that way for the peak of hurricane season. (Credit: NOAA SSD.)
So far I’ve offered a lot of doom and gloom, reasons why I think this will be a less active than average Atlantic hurricane season. So what are some potential positives for tropical development this year? Well, we can go back to the Atlantic SST configuration. While the current pattern is not favorable for “classic” tropical cyclones, that expansive region of warm water in the subtropics could be a breeding ground for several storms this season. Any systems that form this year will have an increased likelihood of reaching their peak intensity at a higher than normal latitude. A region I’m particularly concerned about is off the Southeast coast of the United States, extending down into the Bahamas, and even towards the waters north of Hispaniola and Puerto Rico. These are areas where SSTs are well above normal, and they are also less likely to be impacted by El Nino induced shear. Any disturbance that is able to get into those areas this year will have to be watched. I could see something similar to Hurricane Arthur of last year happening. Another area to watch will be the Caribbean Sea and Gulf of Mexico. These areas are generally seeing warmer than average water temperatures, and will possess the greatest potential energy of anywhere in the Atlantic for the peak of the season. The problem with these regions, as mentioned, will be wind shear; anything that tries to tap into the warm waters will likely face stiff resistance.
Another way we can try to get a picture of the upcoming hurricane season is by looking into the past, with analog based forecasts. It is possible to take present conditions and compare them against past conditions, with the hope that results of similar configurations in the past will be repeated again in the present. This will never produce a perfect forecast, but it can be interesting to look at what has happened in years similar to this in the past. I took a look through all the years since 1950 that featured El Nino events for the peak of the hurricane season. The best analogs I came up with were 1976, 1991, and 2014 (though this wasn’t technically an El Nino year). These years all featured limited activity in the deep tropics, but enhanced activity in the subtropics. 1991 is probably the best of them.
Figure 5: 1991 Atlantic hurricane season storm tracks. I think this represents the best analog for this season. A mainly quiet El Nino year, 1991 did feature Hurricane Bob, which caused major impacts in southeast New England, as well as Hurricane Grace which helped spawn the "Perfect Storm". Past results do not guarantee future ones, but I think the 2015 map will look similar to this in a general sense. (Credit: Levi Cowan.)
So, with all the key factors discussed, it’s now time for my personal forecast. Based on everything listed above, I am predicting a below average 2015 Atlantic Hurricane Season, with my specific forecast numbers as follows:
* 8 tropical storms
* 4 hurricanes
* 1 major hurricane
* ACE 40-60% of normal
That is a forecast for number of storms. I am fairly confident that I will come close on those numbers. However, as is true every year, there is no forecaster nor computer model that can forecast the tracks of any storms months or even weeks in advance. We can highlight areas that may be more at risk than others. But we can never know whether that 1 major hurricane I’m forecasting will pass harmlessly out to sea, or be a potential catastrophe. It only takes one, and that’s a phrase that simply can’t be overused. The United States has not been hit by a major hurricane in nine and a half years. Statistically speaking, we are way, way overdue. With a quiet season likely, the odds are in our favor that our luck holds another year. But you just never know, and it is essential that anyone in a hurricane prone area do their yearly rundown of emergency hurricane procedures. Better safe than sorry, always.
And that’s all folks. We’re still two and a half months away from the start of this season. It’s time to sit back and wait now, and let the chips fall where they may. I’m very excited for another season of storm tracking here on WU. I can’t believe it’ll be my fourth one.
Thank you for reading! I don’t blame you if you just skimmed it as I got a bit longwinded in spots. Feel free to leave comments and questions!
The views of the author are his/her own and do not necessarily represent the position of The Weather Company or its parent, IBM.