Supercomputers Are Driving a Revolution In Hurricane Forecasting (arstechnica.com)
Ars Technica's Eric Berger reports of how dramatic increases in computer power have helped improve the accuracy of hurricane forecasts: Based upon new data from the National Hurricane Center for hurricanes based in the Atlantic basin, the average track error for a five-day forecast fell to 155 nautical miles in 2017. That is, the location predicted by the hurricane center for a given storm was just 155 nautical miles away from the actual position of the storm five days later. What is incredible about this is that, back in 1998, this was the average error for a two-day track forecast. In fact, the annual "verification" report released Wednesday shows that for the hyperactive 2017 Atlantic hurricane season -- which included the devastating hurricanes Harvey, Irma, and Maria -- the National Hurricane Center set records for track forecasts at all time periods: 12-hour, 24-hour, and two-, three-, four- and five-day forecasts.
So where is this "revolution" you speak of?
Even now, when you see the predicted tracks of a hurricane from a bunch of different computer models, there's often one or two that are wildly different.
Yet all the climate models agree very closely? Where's the outlier climate model?
There are a lot of 'super-computers' nowadays
Japan has it
US has it
Europe has it
Even backward shithole countries like China has it
What do they use the supercomputers for?
Would Approve.
You mean "AI" not "forecasting"
The 'error' is due to real hurricanes failing to comply with the NHC's demands. Young, budding hurricanes will be imaged and scrutinized in extreme graphic detail to force the fully-grown hurricanes to step into line. Don't even ASK about the NHC's usage of things like 'spinning', 'blowing' and 'torrential showers' to bend poor hurricanes to their will. What, you thought they did those things of their own free will?! The freedom fighters end up on Youtube, but the revolution will not be televised on The Weather Channel.
Corruption is convincing someone that the selfless ideal is the same as their selfish ideal.
So we have 3 extra days to save our stuff before our house gets destroyed.
Irma was forecast to go up off the east coast of Florida. Then the track moved westward until it was in the Gulf of Mexico. Then Irma's path moved back east before the storm finally came up the west coast of Florida. This caused many people from the east coast of Florida to evacuate to the west coast of Florida which was directly in the hurricane's path.
The forecasting is still not accurate enough for you to safely evacuate and stay in Florida. The majority of hotel rooms in Florida are on the coast. You can evacuate and find yourself in the direct path of the storm in a hotel on the beach.
While I applaud the efforts of all those involved in improving forecasts, when it comes to evacuations, this still isn't enough. Remember the last hurricane that hit Florida? If you do, you will also remember the gridlock on the highways heading north. The problem was the hurricane was predicted to hit south Florida but it veered north, and rolled over all the people that evacuated or were trying to evacuate. Luckily the storm had weakened.
My point is that 155 miles could mean the difference between some living and some dying because the potential affected area is so large our infrastructure can't handle the millions that are trying to get away from the storm. Additionally, if folks in a major metropolitan area like Miami or Jacksonville are motivated to leave, but the hurricane hits north or south, the folks that really need to leave can't.
I'm not sure what the solution is but we can't kid ourselves into thinking "yay, we are masters of the weather" because we aren't. Not even close.
can make a big difference in the impact.
A self-referential AI feedback loop.
Let us take a moment to remember the Earth Simulator, at its time it was the fastest super computer in the world and built specifically for global weather pattern simulations. Super computers obviously have come a long way since then, but this one specifically marked a major milestone in large computing power.
https://en.wikipedia.org/wiki/...