Select to expand quote
ginger pom said...
windjunky -I was going to ask this on another thread but is there anyway to assess certainty from a forecast. Forecasters say what they think is going to happen but never how much evidence there is for it eg they never say "this low could go either way, if it goes north they'll be nothing"
Not really, or rather, not explicitly. The forecasters (for Victoria only at the moment) now take a weather forecast model output grid and manipulate that (with some fancy smoothing-out code making things consistent) to create a new map. See:
www.bom.gov.au/australia/meteye/The text is automatically generated from the maps they create. Groovy eh...
The manipulating they do is in accordance with how they know the model might be deficient (for instance, models generally have too low a surface wind speed), their local knowledge, and for what other models from around the world might be showing. This last aspect is what can give some idea of the uncertainty, but isnt in a product per se. (The only one i know of is the "chance of" rainfall products at:
www.bom.gov.au/jsp/watl/rainfall/pme.jsp ) It may be that one day we do get probability forecasts, because the better models actually make multiple runs every day to try and sample the range of "possible states" of the atmosphere - and from those multiple runs we can get some idea of probability too.
So my advice is... look at several different model outputs rather than just one if you really want to get an idea of probability. If they all line up, good chance, if they dont, less chance.