Analog Forecast For March and Hurricane Season 2014

January to February 2014 500mb height anomaly

January and February 2014 were much warmer and wetter than normal in Bermuda. This is quite unusual, based on the period of records at the airport few of the January-February periods are both wet and warm. So looking back through climate data and finding little overlap, I decided to make two analog groups: a “warm-winter” group and a “wet-winter” group. With a mean January-February temperature of 67.3°F (19.6°C) and total precipitation of 14.31″ at the airport, and using the climate data on the Bermuda Weather Service’s Climate page, and the SC ACIS database, I came up with 7 years that had similarly warm January and Februaries, and then 7 years that had similarly wet January and Februaries.

Warm Winter Analog March height anomaly

The “warm-winter” group of analog years – 1975, 1989, 1984, 1991, 1994, 1990, and 1974  when averaged together, show continued anomalous 500mb ridging over the western Atlantic when averaged over March. This likely allows for continued warmer than climate mean monthly temperatures at 1.2°C above climate as an average of the mean monthly March temperatures for those years which is 0.60 standard deviations above climate. Looking at precipitation, March turns drier than normal – likely a product of the anomalous ridging overhead. When averaged together, these years had a March that was about 0.90″ drier than climate for the month, which is 0.48 standard deviations below climate.

Wet Winter March analog years

The “wet-winter” group of analog years – 1988, 1963, 1980, 1993, 1972, 1958, and 1977 when averaged together show no anomalous ridging across the Atlantic, but in fact start to show weak anomalous toughing extending from the central north Atlantic into Western Europe. These analog years show mean monthly temperatures in Bermuda near climate normal at 0.2°C below (0.1 standard deviation from climate), and for precipitation, March is again drier than climate at 0.20″ (0.11 standard deviations) drier than climate for Bermuda.

Using the “warm-winter” analog years, the following March is significantly warmer and drier than climate. However, using the “wet-winter” analog years the following March is near normal for both temperatures and precipitation. So, based on this, March 2014 should be near or above normal for temperatures and near to below normal for precipitation.

Looking forward to the Atlantic Hurricane Season, from June 1st to November 30th,  we can make the sweeping (and potentially very bad) assumption that conditions in Bermuda during January and February are a predictor for the following Atlantic Hurricane Season. I will again use the two groups of analogs: the warm-winter and wet-winter analogs. This analog based forecast should definitely be taken with a grain of salt and should be taken as something out of curiosity.

Hurricane Season is generally split into two periods, the first half (June to August) where much of the tropical cyclone activity is focussed in the Southwestern Atlantic, Caribbean and Gulf of Mexico.  The second half of the Hurricane Season sees a general increase in activity and shift in activity towards the Central Atlantic with continued activity in the Southwestern Atlantic, Gulf of Mexico, and Caribbean. Although the Atlantic Hurricane Season lasts through November, that month rarely sees significant tropical cyclone activity and for the purposes of this we’ll exclude that month.

JJA warm sea level pressure anomaly JJA wet sea level pressure anomalyu

Looking at the first half of Hurricane Season (June-August), the warm winter analogs (left) show that sea-level pressures over the Western Atlantic, Gulf of Mexico, and Caribbean are higher than normal.  Indicating less than normal tropical cyclone activity during this time. The wet winter analogs (right) show higher than normal sea-level pressure over the southeastern US  and lower than normal sea-level pressures over the Caribbean. These lower sea-level pressures are likely an indicator of the Central American monsoon trough shifting northwards into the Caribbean where disturbances could break off from the trough and develop into tropical cyclones. Sea-level pressures are significantly lower than normal over north Africa in both analogs, this could indicate that tropical waves coming off of Africa are stronger than normal or there are more of them further north than normal.  This could be an indicator of an earlier start to the Cape Verde season where tropical waves develop in the central Atlantic and track westward.

ASO warm sea level pressure anomaly ASO wet sea level pressure anomaly

For the second, meatier part of the Hurricane Season, the warm analogs (left) continue to indicate that sea level pressures are higher than normal across much of the north Atlantic. Anomalies peak over the eastern US and over the north central Atlantic. This pattern appears to be a very suppressive one for tropical cyclone development, but any that do form might tend to stay far to the south or recurve out to sea before or near Bermuda. The wet winter analogs show very different anomaly patterns. Sea-level pressures are higher over the southeastern US, extending to Bermuda and to the Azores.  This combined with the lower than normal sea level pressures over the south central Caribbean could be an indicator stronger than normal trade winds in the Caribbean and the Central Atlantic. Strong trade winds typically interfere with the organization of disturbances into tropical cyclones. Again, both analogs show that sea-level pressures over North Africa are lower than normal indicating continued above normal disturbance activity there.

Overall, based on these analogs, tropical cyclone activity should be near or below normal. Higher than normal sea-level pressures in the warm analogs, and stronger than normal trade winds in the wet analogs are strong inhibitory factors for tropical cyclone development. But this is offset a little by a more active tropical disturbance pattern. The warm winter analog years average out to have 10 tropical storms 5 hurricanes and 2 major hurricanes with a season total ACE of 74.4. The wet winter analog years average out to have 9 tropical storms, 6 hurricanes, and 2 major hurricanes and a season total ACE of 78.6. An average year, based on 1981-2010 climatology would have 12 named storms 6 hurricanes and 3 major hurricanes. ACE values usually add up to 93.5.

Although the analog years would indicate near normal or below normal tropical cyclone activity, they only take into account the air temperature at Bermuda and the precipitation here. They do not take into account the distribution of sea-surface temperature anomalies in the Atlantic, in the equatorial East Pacific (related to El Nino), or upper atmospheric winds that could be inhibitory towards the development of tropical cyclones. Keeping in mind these factors, a forecast of more El Nino like conditions could support a less active season, while warmer than normal Atlantic sea surface temperatures (should they continue above normal into hurricane season) could support a more active season.

It is important to remember that it takes one hurricane impact to ruin a hurricane season. 1963 (Hurricane Arlene), 1975 (Hurricane Faye), and 1989 (Hurricane Dean) are analog years that feature hurricane impacts on Bermuda, it is impossible to say based on these analogs whether or not a hurricane will impact Bermuda in 2014, but the message is to be prepared.


2 thoughts on “Analog Forecast For March and Hurricane Season 2014

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s