CSU's Klotzbach and Gray Suspend December Hurricane Forecast

This study is deeply flawed and is likely outside of the unanimous consensus. As all of you should know by now beyond reasonable doubt, the new null hypothesis is it may be worse than we thought. If you are not able to positively demonstrate it can’t be the case under any circumstances, the precautionary principle requires you to transfer all the money you have in your possession and some more to our bank account or else 10:10
And now /sarc off

Latitude says:
December 13, 2011 at 1:35 pm
Yes. Because I am not intimately familiar with the model used by Gray and Klotzbach, I took the conservative approach of referring to extreme events.

Joe Bastardi is the one to watch. He predicted early most of the paths of last seasons hurricanes. He may not be perfect ,but show me 1 person or group who is better.

Being on the east coast of the US I follow the Atlantic storms. There is a much cooling of the sea temps behind the storms.
Has anyone considered that as the number of named storms increases that the overall power may be reduced as multiple storms remove heat from the oceans to space and therefore would be less energy for the next. Kind of like a self regulating atmospheric/ocean relationship. I would guess that the ACE would be the best indicator but would simply reflect surface temps. I would think that fewer storms would produce stronger ones.
I made a personal prediction that Irene would be a washout (collaspe) as far as wind once it began moving north along the Gulf Stream at about the same speed as the Gulf Stream. It pulled the heat from the sea and as it moved at the same rate there was basically no source of additional energy except at the rear or southern portion of the storm when it slowed. That was confirmed by satilite views showing the eye wall behind the main cloud/storm body. The track it followed and the speed would seem to be able to only produce a tropical storm. It was a big tropical storm but far from a hurricane at landfall.
I believe that Irene was some politician’s fantasy maneuver. The media sure has been quiet since with regards to the big scare along the east coast and a few inches of rain in New England. I can still see that film clip of the covered bridge washing away after it was broadcast over and over and over.
Did New England get any rain from any systems approaching from the west or Gulf of Mexico? I don’t believe I heard about any on the news.

To provide some context here, we do attempt to leave years out when we develop our statistical modeling on historical data. For example, in one of our early August forecast schemes, I built the forecast model on data from 1949-1989 and then tested it on data from 1900-1948 and 1990-2005. The model showed similar levels of skill on both “independent” data sets.
For all of our models, we at least check to make sure that the correlations between predictors and hurricane activity remain statistically significant over various periods of time. For those interested, check out our forecasts at http://tropical.atmos.colostate.edu or in our various peer-reviewed publications for how we go about developing a statistical forecast model.
Our forecasts in June and August have shown promising levels of skill since they started being issued in the mid 1980s, while our early April forecast has shown improved skill in recent years. As longer-period historical datasets come on line, we hope to improve the skill of all of our statistical models. I think the primary challenge with the December outlook is that there has been no proven model (either statistical or dynamical) with skill at predicting ENSO at this extended range.
Obviously, as one goes back further in time, there is more uncertainty, both in historical datasets of atmospheric and oceanic parameters (such as sea surface temperature, sea level pressure, etc.), as well as in historical calculations of hurricane statistics. Prior to the mid 1960s, there was no satellite data, and prior to the mid 1940s, no aircraft reconnaissance, so certainly systems in the open Atlantic were either underestimated or perhaps missed completely.
Hopefully, this adds a little clarity to our decision to suspend quantitative guidance for the time being for this early December outlook.

@Ron Cram December 13, 2011 at 7:11 am:

It is nice to see scientists admitting that models with terrific hindcast skill have absolutely no predictive skill. I’ve been saying this for years. Orrin Pilkey and his daughter wrote a book about it titled “Useless Arithmetic.”

Please. This would be nice if true. But I am not disagreeing with your sentiment, only the hindcasting aspect. If you’ve seen even ONE climate model that can hindcast terrifically, point me at it. Over 10 years ago a meteorological friend pointed out that not one then could, and I haven’t heard of one since.
So far there is no 20/20 hindsight in climate models.
The next one will be the first one.

Phil Klotzbach says:
December 13, 2011 at 6:13 pm

Phil;
Your effort to limit data snooping is commendable. What do you consider to be “significant”; i.e., how many sigmas? Climatology currently has the lowest standards of any purported science, it would seem.