Michael Field writes for Stuff about NIWA’s quarterly seasonal climate outlook, and why accurately forecasting weather out past 16 days is difficult for even the fastest weather super-computers.
However, while scientists can only predict 50% of weather three months out, NIWA’s James Renwick believes it could be possible to improve that figure.
An excerpt: (read in full here)
“The skill would be to maximise the 50 per cent signals they could predict.
“”Half of the variability is not predictable and then if we were totally onto the other half we could be getting it right 70 percent of the time and we’re getting it right 40 to 50 percent of the time now. So there is definitely scope to do it better.”
“A 70 per cent success rate would “be incredibly useful” to energy and agriculture sector.
“No one, though, is ever going to be able to give an accurate weather forecast for a day beyond a two week period.
“”The `butterflies’ get in the way on that one. Predicting the weather on a given day is essentially impossible outside two weeks in advance.””