Posted: 8:50 am Tuesday, April 10th, 2018
By Kirk Mellish
A very PRELIMINARY outlook for the coming tropical cyclone season was issued last week by some of the pioneers in the field at Colorado State University.
It was incorrectly characterized by many in the news business as a prediction for a “bad” season or “another busy season”.
I would not describe the outlook that way number one, and secondly there are multiple sources that make forecasts not just one! There is no “Royal decree” on the matter.
What is more, the track record of outlooks issued this early is not good. However, the ones that are issued in the coming months show useful accuracy.
The April/May sea surface temperatures and ENSO trends have better correlations with better initial conditions to project the tropical season ahead, with the Summer outlooks better still.
Past history has shown what we call the El Nino/La Nina “Spring predictability barrier”. This is because in the Spring the conditions often change rapidly. Performance testing shows model ENSO forecasts are more reliable after the Spring transition season.
For example the April outlook issued last year foresaw a normal to below normal season which wound up very active:
(Image: Capitol Weather)
(Image: Tropical Storm Risk)
But by the Summer the updated forecasts correctly saw an above-normal season coming.
In fact, the CSU outlooks issued by June have shown considerable skill over the years.
The 2017 Atlantic hurricane season had the highest ACE value (223) since 2005 and ranks in the top 5 years since 1900. ACE correlates with named storm frequency and intensity during a hurricane season.
As of now, most of the numerical equation variants suggest we transition from the long La Nina we’ve had in the Pacific to a weak El Nino by the Autumn. The CSU group believes neutral conditions are the best call.
As a rule El Nino is a sign of a below normal season and La Nina with a normal or above-normal season, although sea surface temperatures and wind shear in the Atlantic and Caribbean also play a very significant role.
As I said there are multiple hurricane season forecast groups. Some are based on rigorously tested systems and years of experience with peer-reviewed science, others are not.
At this point some groups are predicting a slightly above normal season and others a slightly below-normal season. So you can see with the uncertainty high in the Spring and with low skill in predicting something 5 months away it’s too soon to focus on hurricane season.
The CSU group outlook:
Follow me on Twitter @MellishMeterWSB.