tag:blogger.com,1999:blog-1367053740188758246.post1942504532347433620..comments2023-06-21T00:39:34.443-07:00Comments on Dosbat: 2015 September NSIDC Extent PredictionChris Reynoldshttp://www.blogger.com/profile/16843133350978717556noreply@blogger.comBlogger2125tag:blogger.com,1999:blog-1367053740188758246.post-14217110154462678012015-06-01T09:45:45.416-07:002015-06-01T09:45:45.416-07:00Thanks Kevin,
Other methods may give different re...Thanks Kevin,<br /><br />Other methods may give different results. I prefer volume as the priming initial condition because of the implied increase in open water formation efficiency. It also allows me to hindcast the method to see if it is of use.Chris Reynoldshttps://www.blogger.com/profile/16843133350978717556noreply@blogger.comtag:blogger.com,1999:blog-1367053740188758246.post-30932240102914403152015-05-31T09:24:27.256-07:002015-05-31T09:24:27.256-07:00Chris,
I haven't spent much time trying to pr...Chris,<br /><br />I haven't spent much time trying to predict extent, but a cursory examination leads me to believe that there's a 30% chance there will be less than a 5% change from last year; an equal 30% chance that the change will be between 5 and 10%; and a 40% chance that it will be greater than 10%.<br /><br />Given that we've seen extent grow in two consecutive years and that each time this has happened in the past the result has been a subsequent average decrease of 12%, I'd vote for an average September extent of 4.59 Mkm^2.<br /><br />The standard deviations over the dataset are so large (13%) that a 95% confidence interval becomes almost meaningless (+/- 1.68 Mkm^2)<br /><br />Based on the fact that variability has increased in recent years, if I were to make an adjusted guess I'd probably go even a few tenths lower.Kevin O'Neillhttps://www.blogger.com/profile/06692943768484857724noreply@blogger.com