Jump to content
Join the Unexplained Mysteries community today! It's free and setting up an account only takes a moment.
- Sign In or Create Account -

Projections of sea level are underestimated


questionmark

Recommended Posts

why have you cherry picked your start point?

you're only trying to find a 15 year flat period.

what is the trend from April 1997 to present in the HADCRUT3 dataset?

...i can't believe this nonsense has gone on for 7 pages!

Little Fish - have you lost your mind - that is a complete homogenous dataset (the satelite record) - that ain't cherry picking that's just data.

I can't believe you are arguing over a cherry picked slice of a complete dataset - over 7 pages.

Br Cornelius

Link to comment
Share on other sites

 

I never said the trend has slowed, I said the trend has been zero. and CRUTEM3 as well as RSS show a flat trend over the last 15 years.

I have to confess to a mistake in the calculations: when I deleted the 1996 datapoint to make my list match yours, I forgot to change n = 16 to n=15. That kind of messed things up.

At any rate, the corrected figures are:

NCDC: Average annual temperature change 1997-2011, inclusive, is +0.0096 degrees C. That produced an F-value of 3.968. F(0.95,1,14)=4.600. The straight-line model does not fit the data. The rate of slope is changing.

Hadcrut3: Average annual temperature change 1997-2011, inclusive, is -0.0061 degrees C. F=0.017. F(0.95,1,14) = 4.600. The straight-line model does not fit the data. The rate of slope is changing.

RSS, lower stratosphere over USA: Average annual temperature change 1997-2011, inclusive, is +0.013 degrees C. F=37.392. F(0.95,1,14) = 4.600. The straight-line model fits the data. The rate of slope is constant.

SO: your contention that temps during that time interval have a constant value of zero might be right for the lower stratosphere over the US, but neither global surface temperature dataset can be fit with a straight line. And means that at least sometime in that interval, the value is NOT zero.

if the trend was positive and statistically significant, then all the datasets would show it. I'm just looking at the data, you are denying and ignoring part of the data. you have to deny the trend has been flat for 15 years, because to accept it would mean letting go of your fear mongering predictions.

As I just showed, the surface temps HAVE NOT been flat for 15 years. If that were so, we could fit a straight line to the data. That is not the case.

Sorry about the calculation mistake earlier, but your basic premise is still false.

Doug

Link to comment
Share on other sites

Little Fish - have you lost your mind

I know what year and month it is. Doug doesn't.

Link to comment
Share on other sites

I know what year and month it is. Doug doesn't.

OK. So what do your calculations say? Mine say there isn't a fifteen-year straight line anywhere in the surface data's 1997-2011 time frame. Reset the dates if you like, but at best that will only shift things by three months - not nearly enough to produce the fifteen-year zero slope you're looking for. How do I know that? The data can't be fit with a straight line model. You have to have a curve to achieve significance at the 95% level. And that means that in at least part of that period, there HAS TO be a non-zero rate of slope. QED.

Doug

Link to comment
Share on other sites

Sorry about the calculation mistake earlier, but your basic premise is still false.

what basic premise is false?

this one maybe? which you just pulled out of your hat :

"your contention that temps during that time interval have a constant value of zero"

how about just calculating the trend from crutem3 between April 1997 and the present?

Link to comment
Share on other sites

what basic premise is false?

As I understand it, your premise is that there has been no change in global mean temps from 1997 through 2011 (or alternatively, April 1997 to "present."). As the dataset ends in October, that is the meaning of "present" in this context.

this one maybe? which you just pulled out of your hat :

"your contention that temps during that time interval have a constant value of zero"

If you are trying to say that there has been no change in global temps in that time interval, then that is your premise. If have merely stated it more clearly to facilitate testing and understanding of what it is that is being tested.

how about just calculating the trend from crutem3 between April 1997 and the present?

That is an involved calculation requiring adjustments for the fact that some months are represented 16 times, while others are represented only 15 times. That gives unequal weight to some parts of the year and distorts the results. Also, there is no reason that you can't do that yourself. So how about it?

It seems to me that you are grasping at straws here. Cherry-picking the time frame by shifting it three months forward will barely affect the numbers at all. Data from the first ten months of 2012 show it to be warmer than the average for 1997-2011, suggesting that 2012 will be the fourth or fifth hottest year on record. That will increase the slope of the line, weakening your contention even further. A better idea that gets us around the calculation problem is to wait until the December figures are published, then run the calculations using 1997 through 2012. Much easier to calculate and also more indicative of what is really happening.

Also, there is no straight line slope in the surface data. That means that at least once, the trend was NOT zero. And that means that the rate of temperature change changed during the 1997-2011 period. Shifting the reference period slightly will not have much effect as the F-values are nowhere near the critical values.

Doug

Link to comment
Share on other sites

if you can't do it, then use a spreadsheet or the online interactive calculator I showed you.

hadcrut3, april 1997-present (more than 15 years), you don't need a number, you just have to look at the red trend line, here:

http://www.woodfortr...gl/from:1997.25

if you want to look at the red trend line closely then delete where it says "series 2" on the right in the above link.

Edited by Little Fish
Link to comment
Share on other sites

if you can't do it, then use a spreadsheet or the online interactive calculator I showed you.

hadcrut3, april 1997-present (more than 15 years), you don't need a number, you just have to look at the red trend line, here:

http://www.woodfortr...gl/from:1997.25

if you want to look at the red trend line closely then delete where it says "series 2" on the right in the above link.

The online calculator does not correct for autocorrelation or seasonal (monthly) variability. Neither does it use the right model (The straight line doesn't fit.). Do I think they made a mistake? With those failures, what do you think?

As demonstrated above, the actual line curves. That means it is changing and your hypothesis of 15 years with no change in surface temperatures is false.

Remember the story about the guy with one foot in a tub of boiling water and the other frozen into a block of ice? On average, he was comfortable. That's what you're trying to do - apply an average to a situation that doesn't warrant it.

It is time for you to take a beginning course in statistics - the straight-line model is covered in high school stat courses. Until you do that, you will not even be aware of the number of unwarranted assumptions you are making.

Doug

P.S.: my questions about what you think are not just rhetorical. If you try to anser them, you will have to think about what you are doing/saying.

Doug

Edited by Doug1029
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Recently Browsing   0 members

    • No registered users viewing this page.