The fit for the land ocean temperature anomaly assuming a cubic term plus a simple sinusoidal term has a very small value for the cubic coefficient.
The projection was made with the smoothed 20-year average plus its unsmoothed 10-year end segments. The data is still withing the 3σ error bounds of the projection so it can't be rejected at this time but this is likely to change since the coefficients of the polynomial indicate that the projection will start to rise and then later decrease over time. One could also rate projections on how well they follow the data in order to determine the best method.
I'd like a chance to ask you some questions but I don't see an email contact. My questions don't really relate directly to some of your posts but your areas of interest seem like they would coincide with some of mine.
ReplyDelete@TCullen I don't see an email contact in your profile.
ReplyDelete