|
Post by steve on Dec 5, 2009 13:49:33 GMT
The current flat level is within the window of a model forecast made in 2005 - in fact for 2008 and 2009 it is pretty close. The following also shows the same model being run against previous periods. This is essentially the same model as the one used for some of the IPCC climate projections. The only difference is that they try harder to get the starting conditions of the ocean right because they think it will give them a "skillful" forecast for a few years out. The jury is still out on whether it will.
|
|
|
Post by hilbert on Dec 6, 2009 0:26:51 GMT
There is a way to shrink pictures so that they fit better. I don't recall exactly how to do this, but kiwi knows.
|
|
|
Post by magellan on Dec 6, 2009 18:10:24 GMT
So now a climate model predicts within a window the current "trend".
Who sees the folly in steve's graph? It is a statistical anomaly.
steve, once you find a model that can correctly identify ENSO events, post it. Otherwise, it is GIGO.
|
|
|
Post by magellan on Dec 9, 2009 3:51:43 GMT
|
|
|
Post by sentient on Dec 9, 2009 5:30:38 GMT
I was just thinking, and although it did hurt, I came upon the notion that perhaps what we are witnessing is the evolution of, yet another, NEW maths....
I have decided to christen this startling new discovery the "algoreithm"
You heard it here first, and I will not release the code or the data that underpins this paradigm shift (note I did not say which way).
Copyright pending.......
|
|
|
Post by nautonnier on Dec 9, 2009 11:34:42 GMT
The current flat level is within the window of a model forecast made in 2005 - in fact for 2008 and 2009 it is pretty close. The following also shows the same model being run against previous periods. This is essentially the same model as the one used for some of the IPCC climate projections. The only difference is that they try harder to get the starting conditions of the ocean right because they think it will give them a "skillful" forecast for a few years out. The jury is still out on whether it will. Steve, First I note that the observed 'global temperature' does not continue to current time. Second as pointed out by Willis Eschenbach in wattsupwiththat.com/2009/12/08/the-smoking-gun-at-darwin-zero/ someone is actually fiddling the base observations (not to mention only using 3 temperature sites in the whole of Australia?). It has come to the point where the 'adjustments' to temperatures cannot be trusted. Therefore there is not a lot of trust in the so called 'observations'. Finally, I note from the graph that the 'Confidence' level (Bayesian trust level) is graded so that the trust increases as the shading lightens this is perhaps the one thing I agree with on this chart that we can have 100% confidence in the actual temperature (not the adjusted observations) being furthest away from the forecast. I would also point out that a 'global temperature' is meaningless - like the average color of cars on a road. What should be used is global heat content if that is measurable.
|
|
|
Post by trbixler on Dec 10, 2009 2:13:02 GMT
|
|