Re: The Mother of All Snow Ramping Threads.....
Hindcasts are used to check whether forecasting methodolgy can correctly predict past weather events in order to verify the potential accuracy of the forecasting method. A forecast for the past is exactly what it is, because we know what happened, we can check the forecasts accuracy without waiting for it to happen.
I'm not going to explain the in's and outs, but basically the TWO community and The Weather Outlook are two seperate entities. One of the members of the TWO forum produced this forecast:
Quote:
Number of matching years: 10 (medium)
Confidence in forecast: Moderate
Exceptionally mild (>6.5c): 0%
Very mild (>6c): 5%
Mild (>5c): 5%
Average (+/- 4.2c): 50%
Rather cold (<3.5c): 30%
Very cold (<2.5c): 5%
Severe (<1.5c): 5%
Trend: Average or rather cold
Average CET forecast: 3.7c
Coldest month: February (3.2c)
Having used the same method to hindcast the winters of the 70's, 80's and 90's the success rate was 63%, generally the method worked more accurately with winters that were closer to the average, rather than ones which were forecast to be exceptionally mild, or cold.
[%sig%]