The National Oceanic and Atmospheric Administration (NOAA), Colorado State University (CSU) and Tropical Storm Risk (TSR) have each published projections for storm activity during the 2018 North Atlantic hurricane season, which officially runs from the beginning of June to the end of November. A summary of the forecasts produced by each can be seen below:
|Forecaster||Named Storms||Hurricanes||Major Hurricanes|
|1950 – 2017 average||11||6||3|
|2008 – 2017 average||15||7||3|
|NOAA: National Oceanic and Atmospheric Administration|
|CSU: Colorado State University|
|TSR: Tropical Storm Risk|
|Named Storm:||Sustained Winds > 33 knots = Tropical Storm status|
|Hurricane:||Sustained Winds > 63 knots = Hurricane Category 1 to 5|
|Major Hurricane:||Sustained Winds > 95 knots = Hurricane Category 3 to 5|
Alpha comment: whilst we tend not to attach too much credence to these forecasts, it has been the case that the Atlantic has noticeably cooled in recent months, with CSU yesterday (31 May) reducing the amount of storm activity contained in their initial 2018 forecast published in April. Nevertheless, these forecasts taken together still predict an average amount of hurricane and major hurricane activity this year.
Recent history has shown active seasons can come in pairs following prolonged periods of inactivity (eg 2004 and 2005). The volatile nature of hurricanes, however, means that computer models struggle to accurately predict the track of an active hurricane when only a few days away from the coast (eg Hurricane Irma in 2017), let alone the number of landfalling hurricanes which may occur 3 months in advance of the season’s peak. Unsurprisingly, the track record of such forecasters in predicting the number of landfalling major hurricanes (ie the ones that will bother Lloyd’s) is underwhelming. Furthermore, and most importantly, forecasts do not and cannot predict which section of a coastline the anticipated hurricanes will strike.
To reiterate a comment made a few years ago:
“I find these forecasts to be nearly meaningless. They contain no actionable information. I am all for the research, but not happy about the hype and scepticism it brings. In 1992 we had four Atlantic hurricanes from six named storms. It was a light season. The “A” storm came in mid-August. That was Hurricane Andrew. As with real estate, the three most important factors are location, location, location. Nothing else, including number of storms, matters.”