cc: John.Lanzante@noaa.gov, "Thomas.R.Karl" <Thomas.R.KarlatXYZxyza.gov>, carl mears <mearsatXYZxyzss.com>, "David C. Bader" <bader2atXYZxyzl.gov>, "'Dian J. Seidel'" <dian.seidelatXYZxyza.gov>, "'Francis W. Zwiers'" <francis.zwiersatXYZxyzgc.ca>, Frank Wentz <frank.wentzatXYZxyzss.com>, Karl Taylor <taylor13atXYZxyzl.gov>, Melissa Free <Melissa.FreeatXYZxyza.gov>, "Michael C. MacCracken" <mmaccracatXYZxyzcast.net>, "'Philip D. Jones'" <p.jonesatXYZxyz.ac.uk>, Sherwood Steven <steven.sherwoodatXYZxyze.edu>, Steve Klein <klein21atXYZxyzl.gov>, 'Susan Solomon' <susan.solomonatXYZxyza.gov>, "Thorne, Peter" <peter.thorneatXYZxyzoffice.gov.uk>, Tim Osborn <t.osbornatXYZxyz.ac.uk>, Tom Wigley <wigleyatXYZxyz.ucar.edu>

date: Fri, 28 Dec 2007 16:14:10 -0800

from: Ben Santer <santer1atXYZxyzl.gov>

subject: Re: [Fwd: sorry to take your time up, but really do need a scrub

to: Leopold Haimberger <leopold.haimbergeratXYZxyzvie.ac.at>

<x-flowed>

Dear Leo,

The Figure that you sent is extremely informative, and would be great to

include in a response to Douglass et al. The Figure clearly illustrates

that the "structural uncertainties" inherent in radiosonde-based

estimates of tropospheric temperature change are much larger than

Douglass et al. have claimed. This is an important point to make.

Would it be possible to produce a version of this Figure showing results

for the period 1979 to 1999 (the period that I've used for testing the

significance of model-versus-observed trend differences) instead of 1979

to 2004?

With best regards, and frohes Neues Jahr!

Ben

Leopold Haimberger wrote:

> Dear all,

>

> I have attached a plot which summarizes the recent developments

> concerning tropical radiosonde temperature datasets and which could be

> a candidate to be included in a reply to Douglass et al.

> It contains trend profiles from unadjusted radiosondes, HadAT2-adjusted

> radiosondes, RAOBCORE (versions 1.2-1.4) adjusted radiosondes

> and from radiosondes adjusted with a neighbor composite method (RICH)

> that uses the break dates detected with RAOBCORE (v1.4) as metadata.

> RAOBCORE v1.2,v1.3 are documented in Haimberger (2007), RAOBCORE v1.4

> and RICH are discussed in the manuscript I mentioned in my previous email.

> Latitude range is 20S-20N, only time series with less than 24 months of

> missing data are included. Spatial sampling of all curves is the same

> except HadAT which contains less stations that meet the 24month

> criterion. Sampling uncertainty of the trend curves is ca.

> +/-0.1K/decade (95% percentiles estimated with bootstrap method).

>

> RAOBCORE v1.3,1.4 and RICH are results from ongoing research and warming

> trends from radiosondes may still be underestimated.

> The upper tropospheric warming maxima from RICH are even larger (up to

> 0.35K/decade, not shown), if only radiosondes within the tropics

> (20N-20S) are allowed as reference for adjustment of tropical radiosonde

> temperatures. The pink/blue curves in the attached plot should therefore

> not be regarded as upper bound of what may be achieved with plausible

> choices of reference series for homogenization.

> Please let me know your comments.

>

> I wish you a merry Christmas.

>

> With best regards

>

> Leo

>

> John Lanzante wrote:

>> Ben,

>>

>> Perhaps a resampling test would be appropriate. The tests you have

>> performed

>> consist of pairing an observed time series (UAH or RSS MSU) with each one

>> of 49 GCM times series from your "ensemble of opportunity". Significance

>> of the difference between each pair of obs/GCM trends yields a certain

>> number of "hits".

>>

>> To determine a baseline for judging how likely it would be to obtain the

>> given number of hits one could perform a set of resampling trials by

>> treating one of the ensemble members as a surrogate observation. For each

>> trial, select at random one of the 49 GCM members to be the

>> "observation".

>> From the remaining 48 members draw a bootstrap sample of 49, and perform

>> 49 tests, yielding a certain number of "hits". Repeat this many times to

>> generate a distribution of "hits".

>>

>> The actual number of hits, based on the real observations could then be

>> referenced to the Monte Carlo distribution to yield a probability that

>> this

>> could have occurred by chance. The basic idea is to see if the observed

>> trend is inconsistent with the GCM ensemble of trends.

>>

>> There are a couple of additional tweaks that could be applied to your

>> method.

>> You are currently computing trends for each of the two time series in the

>> pair and assessing the significance of their differences. Why not first

>> create a difference time series and assess the significance of it's

>> trend?

>> The advantage of this is that you would reduce somewhat the

>> autocorrelation

>> in the time series and hence the effect of the "degrees of freedom"

>> adjustment. Since the GCM runs are based on coupled model runs this

>> differencing would help remove the common externally forced variability,

>> but not internally forced variability, so the adjustment would still be

>> needed.

>>

>> Another tweak would be to alter the significance level used to assess

>> differences in trends. Currently you are using the 5% level, which yields

>> only a small number of hits. If you made this less stringent you would

>> get

>> potentially more weaker hits. But it would all come out in the wash so to

>> speak since the number of hits in the Monte Carlo simulations would

>> increase

>> as well. I suspect that increasing the number of expected hits would

>> make the

>> whole procedure more powerful/efficient in a statistical sense since you

>> would no longer be dealing with a "rare event". In the current scheme,

>> using

>> a 5% level with 49 pairings you have an expected hit rate of 0.05 X 49

>> = 2.45.

>> For example, if instead you used a 20% significance level you would

>> have an

>> expected hit rate of 0.20 X 49 = 9.8.

>>

>> I hope this helps.

>>

>> On an unrelated matter, I'm wondering a bit about the different

>> versions of

>> Leo's new radiosonde dataset (RAOBCORE). I was surprised to see that the

>> latest version has considerably more tropospheric warming than I recalled

>> from an earlier version that was written up in JCLI in 2007. I have a

>> couple of questions that I'd like to ask Leo. One concern is that if

>> we use

>> the latest version of RAOBCORE is there a paper that we can reference --

>> if this is not in a peer-reviewed journal is there a paper in submission?

>> The other question is: could you briefly comment on the differences in

>> methodology used to generate the latest version of RAOBCORE as

>> compared to the version used in JCLI 2007, and what/when/where did

>> changes occur to

>> yield a stronger warming trend?

>>

>> Best regards,

>>

>> ______John

>>

>>

>>

>> On Saturday 15 December 2007 12:21 pm, Thomas.R.Karl wrote:

>>

>>> Thanks Ben,

>>>

>>> You have the makings of a nice article.

>>>

>>> I note that we would expect to 10 cases that are significantly

>>> different by chance (based on the 196 tests at the .05 sig level).

>>> You found 3. With appropriately corrected Leopold I suspect you will

>>> find there is indeed stat sig. similar trends incl. amplification.

>>> Setting up the statistical testing should be interesting with this

>>> many combinations.

>>>

>>> Regards, Tom

>>>

>>

>>

>

--

----------------------------------------------------------------------------

Benjamin D. Santer

Program for Climate Model Diagnosis and Intercomparison

Lawrence Livermore National Laboratory

P.O. Box 808, Mail Stop L-103

Livermore, CA 94550, U.S.A.

Tel: (925) 422-2486

FAX: (925) 422-7675

email: santer1atXYZxyzl.gov

----------------------------------------------------------------------------

</x-flowed>

## No comments:

## Post a Comment