It’s spring here on Roatan, we don’t have to deal with all those cherry blossoms, and the whole new life growing outta the dead of winter scenario, but it is spring none the less. Subtle changes in the environment give clues such as new growth on the mango trees, the drying up of the pot holes, and a drop in the gross tonnage of cruise ship visitors as the ship routes veer north.
The weather becomes more regular with the cessation of the winter storms which are the tail ends of the cold weather fronts that sweep across el Norte until early spring. By the beginning of May the weather is less volatile and you can look to the 5 day forecast with some confidence (?). Normally we receive the trade winds from the east varying up to 30 mph at times. However situated on the leeward side of the island we are in the wind shadow and the dives site remain calm while the prevailing wind connects with the surface a mile or two off shore, and we can see the waves and whitecaps in the distance. This is great for almost all of our diving, but can be problematic on the Blackwater dives.
The Blackwater dives take place about 2 miles off shore in the open ocean. With prevailing winds often in the 20 to 30 mph range and waves 4 to 6 feet, we keep an eye on the five day forecast for those intervals when the wind dies, and the seas calm down and start behaving.
These five day forecasts can be problematic. It seems that when that calm clear day that was forecasted just four days ago becomes tomorrow, it loses all resemblance to the day you planned for. It can be maddening in the tourism business.
So what gives? Less than 100 years ago forecasters just had thermometers barometers wind speed/directions, the Farmer’s Almanac, and messages from others in the direction of the prevailing weather by telegraph, or pony express. Dicey weather prediction could be excused. However today’s forecasters, ahem Meteorologists, have up to the minute precise digital global weather information in real time. Instantaneous communication globally at the speed of light via the internet brings a constant data flow. GPS now accurate to a fraction of an inch give precise locations of phenomena on live digital projection maps. Doppler weather radar tracks cloud systems as they approach giving information about cloud densities, speeds, and precipitation. Weather satellites supply images of weather systems as they track across the globe in infra-red, ultra- violet, and extreme High def. You can watch the weather come over the hill on your smart phone, and yet any long range prediction (over a day) remains iffy.
To be fair to meteorologists, theirs is an impossible job. Weather, broken down to its simplest denominator is turbulence. The way the air comes outta your ceiling fan, the action created by the mixer in your cake batter, the blending of fuel and air molecules in your carburetor, the swirling smoke in your bong, as well as surface winds and ocean currents, are all examples of turbulence.
Turbulence by its own definition is unpredictable. It is best explained with a quote from Wikipedia:
“The onset of turbulence can be predicted by a dimensionless constant called the Reynolds number, which calculates the balance between kinetic energy and viscous damping in a fluid flow. However, turbulence has long resisted detailed physical analysis, and the interactions within turbulence create a very complex situation.”
Just to be sure we don’t understand this; Richard Feynman has described turbulence as the “most important unsolved problem of classical physics.”
The whole field of Meteorology is an impossibility. Weather is Turbulence. Turbulence is unsolvable and unpredictable. Yet the weather forecasters abound. What kind of success rate can they have?
I found a website where a gentleman has plotted the weather accuracy of the BBC forecasters over the course of a few years, and we all know from asking them that the British must have proper forecasters. After all, they are British….
Extrapolating the data we find that the BBC when forecasting out five days has about a 35% success rate, a one chance outta three of being correct. Whereas the forecast for the next day’s weather is usually just under 60% accurate most anytime of the year. As a base point the author of this page (I can’t find his name anywhere) made his own weather predictions by forecasting that the next day would be the same as today, and hit the mark about 33% of the time.
These are dismal performances seen in the most optimistic light. With every tool at their command meteorologists are able to tell the weather 5 days out correctly 1 out of 3 times, almost. Tomorrow’s weather, using all the science at their disposal, can be foretold 2 out of three times. Averaging this out we can rely on 50% accuracy in weather forecasting at best at any time or place.
Even odds, like flipping a coin, so why not?