Guest post by Bob Tisdale
My blog post October to December 2011 NODC Ocean Heat Content Anomalies (0-700Meters) Update and Comments was cross posted by at WattsUpWithThat here. (As always, thanks, Anthony.) Starting at the January 27, 2012 at 8:44 am comment by J Bowers, I was informed of a critique of my ARGO-era model-data comparison graph by the Tamino titled Fake Predictions for Fake Skeptics. Oddly, Tamino does not provide a link to my post or the cross post at WattsUpWithThat, nor does Tamino’s post refer to me by name. But I believe it would be safe to say that Tamino was once again commenting on my graph that compares ARGO-era NODC Ocean Heat Content data to the climate model projection from Hansen et al (2005), “Earth’s energy imbalance: Confirmation and implications”. If he wasn’t, then Tamino fooled three persons who provided the initial comments about Tamino’s post on the WUWT thread and the few more who referred to Tamino’s post afterwards. The ARGO-era graph in question is shown again here as Figure 1. Thanks for the opportunity to post it once again up front in this post, Tamino.
Figure 1
Take a few seconds to read Tamino’s post, it’s not very long, and look at the graphs he presented in it.
My graph in Figure 1 is clearly labeled “ARGO-Era Global Ocean Heat Content Model-Data Comparison”. The title block lists the two datasets as NODC/Levitus et al (2009) and Hansen et al (2005) Model Mean Trend. It states that the Model Mean Trend and Observations had been Zeroed At Jan 2003. Last, the title block lists the time period of the monthly data as January 2003 to December 2011.
In other words, my graph in Figure 1 pertains to the ARGO-based OHC data and the GISS model projection starting in 2003. It does not represent the OHC data or the GISS model hindcast for the periods of 1993-2011 or 1955-2011, which are the periods Tamino chose to discuss.
Do any of the graphs in Tamino’s post list the same information in their title blocks? No. Do any of Tamino’s graphs compare the climate model projection from Hansen et al (2005) to the NODC OHC observations? No. Does Tamino refer to Hansen et al (2005) in his post? No. Do any of Tamino’s graphs present the NODC OHC data or the Hansen et al model projection during the ARGO-era, starting in January 2003 and ending in December 2011? No.
In other words, Tamino redirected the discussion from the ARGO-era period of 2003-2011 to other periods starting in 1955 and 1993. ARGO floats were not in use in 1955 and they were still not in use in 1993. He also redirected the discussion from the projection of the GISS model mean to the linear trend of the data itself. Yet Tamino’s followers fail to grasp the obvious differences between his post and my ARGO-era graph.
In my post, I explained quite clearly why I presented the ARGO-era model-data comparison with the data zeroed at 2003. Refer to the discussion under the heading of STANDARD DISCUSSION ABOUT ARGO-ERA MODEL-DATA COMPARISON. Basically, Hansen et al (2005) apparently zeroed their model mean and the NODC OHC data in 1993 to show how well their model matched the OHC data from 1993 to 2003. Hansen et al explained why they excluded the almost 40 years of OHC and hindcast data. The primary reason was their model could not reproduce the hump in the older version of the Levitus et al OHC data. Refer to Figure 2, which is a graph from a 2008 presentation by Gavin Schmidt of GISS. (See page 8 of GISS ModelE: MAP Objectives and Results.) I presented that same graph graph by Gavin in my post GISS OHC Model Trends: One Question Answered, Another Uncovered, which was linked in my OHC update from a few days ago.
Figure 2
I zeroed the data for my graph in 2003, which is the end year of the Hansen et al (2005) graph, to show how poorly the model projection matched the data during the ARGO-era, from 2003 to present. In other words, to show that the ARGO-era OHC data was diverging from the model projection. Hansen et al (2005), as the authors of their paper, chose the year they apparently zeroed the data for one reason; I, as the author of my post, chose another year for another reason. I’m not sure why that’s so hard for Tamino to understand.
And then there’s Tamino’s post, which does not present the same comparison. He did, however, present data the way he wanted to present it. It’s pretty simple when you think about it. We all presented data the way we wanted to present it.
Some of the readers might wonder why Tamino failed to provide a similar ARGO-era comparison in his post. Could it be because he was steering clear of the fact that it doesn’t make any difference where the model projection intersects with the data when the trends are compared for the ARGO-era period of 2003 to 2011? See Figure 3. The trend of the model projection is still 3.5 times higher than the ARGO-era OHC trend. Note that I provided a similar graph to Figure 3 in my first response to his complaints about that ARGO-era OHC model-data comparison. See Figure 8 in my May 13, 2011 post On Tamino’s Post “Favorite Denier Tricks Or How To Hide The Incline”.
Figure 3
Figure 3 shows the ARGO-era OHC data diverging from the model projection. But the visual effect of that divergence is not a clear as it is in Figure 1. I presented the data in Figure 1 so that it provided the clearest picture of what I wanted to show, the divergence. That really should be obvious to anyone who looks at Figure 1.
Some might think Figure 1 is misleading. The reality is, those illustrating data present it so that it provides the best visual expression of the statement they are trying to make. The climate model-based paper Hansen et al (2005) deleted almost 40 years of data and appear to have zeroed their data at 1993 so that they could present their models in the best possible light. Base years are also chosen for other visual effects. The IPCC’s Figure 9.5 from AR4 (presented here as Figure 4) is a prime example. Refer to the IPCC’s discussion of it here.
Figure 4
The Hadley Centre presents their anomalies with the base years of 1961-1990. Why did the IPCC use 1901-1950? The answer is obvious. The earlier years were cooler and using 1901-1950 instead of 1961-1990 shifts the HADCRUT3 data up more than 0.2 deg C. In other words, the early base years make the HADCRUT anomaly data APPEAR warmer. It also brings the first HADCRUT3 data point close to a zero deg C anomaly, and that provides another visual effect: the normalcy of the early data.
Base years for anomalies are the choice of the person or organization presenting the data. Climate modelers choose to present their models in the best light; I do not.
Those familiar with the history of Tamino’s complaints about my posts understand they are simply attempts by him to mislead or misdirect his readers. And sometimes he makes blatantly obvious errors like using the wrong sea surface temperature dataset in a comparison with GISS LOTI data. His post Fake Predictions for Fake Skepticsis just another failed critique to add to the list.
ABOUT: Bob Tisdale – Climate Observations
SOURCE
The NODC OHC data used in this post is available through the KNMI Climate Explorer:
http://climexp.knmi.nl/selectfield_obs.cgi?someone@somewhere
jasonpettitt says: “Clue 2: Bob Tisdale, having decided to very crudely represent the model output with a straight line (so you’re being shown things at very different detail resolutions) of his own making…”
Apparently you’re not aware that the modelers themselves decide “to very crudely represent the model output with a straight line”. Refer to OHC model-data comparisons in the RealClimate posts here:
http://www.realclimate.org/index.php/archives/2009/12/updates-to-model-data-comparisons/
And here:
http://www.realclimate.org/index.php/archives/2011/01/2010-updates-to-model-data-comparisons/
George E. Smith; says: “The brown ‘trend line’ I take it is an alternative that you offer, or else is NODC’s own trend computation.”
The brown trend line is calculated by the spreadsheet EXCEL from the NODC’s Ocean Heat Content data available through the KNMI Climate Explorer.
Camburn
“And this begs the question, why was he using 2.7C when according to the AGW folks 3.5C is the metric?”
1. Hansen’s model has a sensitivity of 2.7. This is not a set parameter but rather is an result of running the model with given forcings.
2. Circa 2007 the models used by the IPCC ranged from 2.2 to 4.4
3. The average of the models is ~3.2C
Perhaps another way to look at this :
obtain the latest ARGO data set and the plot tool from the U of San Diego.
install
run plot for time span of current year, back to year that first represented the current set ( size ) of
ocean buoys in place.
plot is pacific ocean only
plot is surface to max depth ( 700 M ? )
plot is – 60 to 60 lat
plot is – long to + long that your computer main memory can support.
examine carefully for change in ocean thermal energy.
Steven mosher says;
1. Hansen’s model has a sensitivity of 2.7. This is not a set parameter but rather is an result of running the model with given forcings.
2. Circa 2007 the models used by the IPCC ranged from 2.2 to 4.4
3. The average of the models is ~3.2C
—————————————————————————————————————
That’s all good if we live in a model world, but we don’t.
Nail your flag to the mast Steven, are you a believer?
Bob Tisdale – “How did I give it an “incorrect offset”, when I am only looking at the period of 2003 to 2011? The OHC data and model mean before 2003, and any trend contained in that OHC data is not relevant to the presentation of the trend in my Figure 1.”
It’s entirely relevant when you criticize the Hansen 2005 predictions, which include a particular offset that is a fit to longer term data. You’ve changed that offset, meaning that you are no longer discussing the Hansen prediction, yet you mislabel your strawman (http://en.wikipedia.org/wiki/Straw_man) as Hansen’s.
That’s an egregious distortion – and that’s why everyone is calling you on it.
You’re welcome, Bob! Please let me applaud the WUWT trend toward posting:
(1) more data,
(2) higher-quality data,
(3) spanning more decades, and
(4) subtracting known short-term trends
(e.g., volcanoes, solar cycle, El Nino/La Nina), and
(5) comparing with “classic” nonskeptical and skeptical predictions.
(e.g., Hansen 1981, Lindzen 1989)
This (for me) represents the collision of the strongest science with the strongest skepticism.
Everyone wins from that.
Bob Tilsdale, what is your background and career area?
I don’t even know why you choose to debate with him. His language is that of a fanatic and is abusive … it does nothing to acknoweldge it.
Regarding the data, the divergence is clear. The only point I might add is that 10 years isn’t really enough to say anything. I don’t know what might be an appropriate timescale (as per your question) but when looking backwards at century scale data, only anomalies of 30 years plus are usually discussed (eg. the cool period from 1945-1975) etc..
Nick Stokes says:
January 28, 2012 at 2:30 pm
“I zeroed the data for my graph in 2003, which is the end year of the Hansen et al (2005) graph, to show how poorly the model projection matched the data during the ARGO-era, from 2003 to present.”
I think this quote captures it. The issue is simple. You say you are showing “how poorly the model projection matched the data during the ARGO-era”. But what you’ve plotted isn’t the model projection. You’ve altered it.
………………………………………………………………………………………………………………………
Nick you seem unwilling to cast a critical eye over the Hansen paper. The question seems to be whether a decade of OHC data can really give us a complete insight into what is causing the recent warming. That was essentially Hansen’s claim in this paper. Teh 1993-2003 data was doing a nice job of re-inforcing the claims being made with climate models. The slower rate of increase in the ARGO period is undermining that. You can raise doubt’s about the ARGO data or call for a greater role for aerosols or come up with whatever explanation you want for the present situation but the fact remains that the happy coincidence that occurred in that decade appears to be unraveling.
The greatest weakness in Bob’s argument seems to be .the same weakness in Hansen’s. A decade of data is only ever going to confirm our own prejudices or give us insight into a small part of the problem. At least Bob is man enough to admit to this.
To all who feel my zeroing the NODC OHC data and the model projection at 2003 is a problem:
I offered in a comment above to use the base years of 1955-2011 for my short-term ARGO-era model-data comparison. That way there can be no claims that I’ve cherry picked the base years or shifted the data inappropriately. I do not have the capability to process the GISS Model-ER OHC hindcast and projection data from the CMIP3 archive. So I cannot to create the ensemble member mean of the global data, on a monthly basis, for the period of 1955 to present. But some of you do have that capability. You could end the debate.
If you choose to do so, please make available online for all who wish to use it the Global GISS Model-ER hindcast/projection ensemble member data on a monthly basis from 1955 to present or as far into the future as you decide.
I will revise my recent OHC update and reuse that model data for future OHC updates. That way we don’t have to go through this every time I use that ARGO-era comparison graph as the initial graph in my OHC updates.
Does it really matter where the lines start?
It’s really very simple:
1. The model predicts 0.17 rise in the ARGO-Era.
2. Observations show 0.03 rise in the ARGO-Era.
Call it whatever you want, show it to start at whatever level you want. But the model predicts a rise 5.66 times more than actually occurred. Warmists just have their nickers in a knot.
When do you call a model a failure?
As I commented elsewhere, I believe that Tamino chose to avoid using Bob Tisdale’s real name and linking to his post in order to avoid the terms of service issues he encountered when he posted a particularly vitriolic commentary on Donald Rapp. This is speculative on my part, perhaps Tamino will post here and clear up the air on why he chose to not link to the article he was commenting on (and perhaps pigs will learn to fly).
Regarding baselines, they don’t affect trends, and graphs are practically useless as a means of analyzing the significance of difference of trends. Statistical based methods are much better.
But if you are going to compare data, you need to use a common baseline when comparing data to model ensemble mean, and that baseline should be over the “verification period” of the model with data (that is not part of the forecast period)… in Bob’s case this should be 1993-2003.
And you should show the verification period together with the forecast period.
It was also informative to me to see that the models and data failed to converge during the period 1970-1980. This also tells us something about how unique (or not unique) the current period of disagreement between model and data is.
Gneiss: “1. The starting point for a physical model (e.g., Hansen’s). If that point is unusually high or low, but the physics are any good, the model itself should correct to more reasonable values as it runs forward.”
If we assume that (a) it is based on physics, (b) that it is not chaotic, and (c) that negative feedbacks dominate then yes, your point holds. That said, (a) is entirely unnecessary and (b) is in error. Such that there is no prior reason to expect that Hansen’s model will converge. Lastly, if we accept (c) then we, to the best my knowledge, disallow Hansen’s model and the bulk of the other climate models.
“2. The y intercept in a regression. That usually is not the starting point of the data, and the trend may well be unrealistic if you force it to be. … ”
This is a non sequitor. If we’re talking about linear regressions then it is certainly the case in y = mx + b, that m is independent of b. In the non-linear case y = f(x) + b cannot create an inappropriate trend either. It is only in the cases in which we have y_n = f(x, y-(n-1)) + b that your point carries any merit and only on the presumption that f(x, y_(n-1)) is non-linear and that we are ‘zeroing’ the graph at the same time that we are producing the graph. Zeroing afterwards can not reach back though time and create a different trend line.
“3. Anomalies. Shifting the baseline for anomalies can be done for many reasons but should have no effect on how steep either a physical or statistical model trend is. The line should be at an appropriate height as long as the model knows about the new baseline too.”
This also is a non sequitor. If we are talking about zeroing the scale units of the output — after the output has been generated — then the model is entirely independent of this correction. You freely interchanging a view of the produced data (Tisdale) and *producing* the data (new baseline in the model). These are two entirely different subjects and if we accept that you meant the first then your last sentence is nonsense and your argument is in error. If we accept that you meant the latter then your first sentence is in error, and your argument with it.
All that said however, there are a number of sketchy issues involving the use of linear regressions generally and specifically on this topic and for these uses. But it is very much a tu qoque issue. If it is a valid argument for the AGW crowd then it is a valid argument for Tisdale — because valid arguments do not rely on your metaphysics, religion, or income source. If it is an invalid argument for Tisdale, then it is also for the AGW crowd, because it’s simply an invalid argument.
It is beyond contest that the AGW crowd has, and continues to, use arguments such as the one here. And if it is valid then Tisdale is right to ask when we accept that falsification has occured on the basis of those arguments. If it is not valid then the AGW crowd doesn’t *have* any arguments and we aren’t at all worried about the political, religious, and economic fallout from AGW. Nor are we worried about the philosophy of science since we haven’t gotten ourselves around to performing any.
Bob Tisdale says: January 28, 2012 at 5:04 pm
‘Nick Stokes says: “I think this quote captures it. The issue is simple. You say you are showing “how poorly the model projection matched the data during the ARGO-era”. But what you’ve plotted isn’t the model projection. You’ve altered it.”
The model projection is 0.6 watts/m^2 according to Hansen’s reply to Pielke and Christy here:
http://pielkeclimatesci.files.wordpress.com/2009/09/1116592hansen.pdf
All I’ve done is change it to GJ/m^2.’
No, what Hansen said was:
” Our simulated 1993-2003 heat storage rate was 0.6 W/m2 in the upper 750 m of the ocean.”
It’s not a projection for the Argo era. But more seriously it isn’t a projection of OHC. That needs an offset value, which is yours, not theirs.
KR , you quoted Willis, highlighting a portion of Willis’s closing sentence, “So, for this reason I am still not comfortable with the pre-2005 estimates of heat content.”
It appears you are saying we should dismiss the OHC model-data comparison in the Hansen et al (2005) paper due to the errors in the ARGO and XBT data before 2005. That would leave only the ARGO OHC data from 2005 to present for my comparison. If we’re dismissing the Hansen 1993-2003 comparison, how should the model projection and the OHC data be compared from 2005 to present?
All this argument about offsets is pointless because all you need to look at is the different gradients of the trends. You could plot them in different charts and still see the gradient difference,
The most informative plot I have seen (albeit for SST rather than OHC) compares gradients derived from 30 year trends calculated on a moving basis every month.
http://climate-change-theory.com/360month.jpg
I have added the yellow trend line which shows a decrease in the gradient over the 100+ year period. No one can call this cherry picking – it’s nearly all the data we have.
What it shows is that SST were increasing at a rate of about 0.06 deg.C per decade between 1900 and 1930 and are now only increasing at about 0.05 deg.C per decade, heading for a maximum in the actual long-term temperature trend by about the year 2200. You could expect OHC to follow a similar trend in the long-term.
So, yes the world is still warming, but an end to the warming is in sight, followed by perhaps 500 years of cooling.
The reason for the decline in gradient is that the ~1000 year trend is roughly sinusoidal and is approaching a maximum within 200 years, the previous maximum being the MWP. (It has passed the flex.) The decline should increase a little so the trend itself is cyclic and should pass through zero when that maximum occurs, for which temperatures look like being about 0.8 to 1.0 degree higher than at present – in 2200 that is.
You could construct a similar plot of 30-year moving trends for all the OHC data available, and this would not be open to criticism. It is more important to demonstrate however that temperatures won’t increase more than a degree or so, than it is to argue about OHC, because there will come a time when some of the OHC could seep under the floor of the oceans or into the deep depths of the shorelines. This would demonstrate the stabilising effect of the huge amount of energy in the rest of the Earth all the way down to the core – see ‘Explanation’ page at http://climate-change-theory.com
This is hard to follow. Tamino’s first graph clearly shows Ocean Heat Content increasing and the units are Joules. Tisdale’s first graph is GigaJoules per meter squared and does not have the same high density of sample points, making it harder to follow. Tisdale then goes on to show a graph in Watt years per meter squared and then converts to Temperature anomalies in degrees C. So chip away at Tamino’s ankles if you wish but you have left many behind as we can’t see which band wagon we are supposed to be jumping onto.
There are two issues here. Firstly Bob is quite right in saying that the slope over the period of the ARGO data is lower than the slope that Hansen calculated. It is however quite wrong in to that the data has diverged from a projection of Hansen’s regression line.
Hansen’s regression line when projected forward matches the ARGO data remarkably well – see the graph in Tamino’s post which Bob linked to (kudos for that Bob.) Whatever motives that Hansen had for choosing his starting point they were clearly not a fudge to help him match future data!
Mooloo says “time will be the winner.” It is arguable that in Hansen’s case it has already won – the data that were not available to Hansen fit neatly around his trend line. It will be interesting to see how closely the data for the next nine years fits Bob’s trend line.
Nick Stokes says: “It’s not a projection for the Argo era. But more seriously it isn’t a projection of OHC. That needs an offset value, which is yours, not theirs.”
It appears to be common practice by the climate science community in their presentations of data at their blogs to extrapolate model projections and to shift or offset data. Note the extrapolation of the Model Mean and the location of the Lyman et al (2010) OHC data in the OHC model-data comparison from the RealClimate post here:
http://www.realclimate.org/index.php/archives/2011/01/2010-updates-to-model-data-comparisons
Notice the base years of the data in Lyman et al (2010) and how their data actually sits in their graphs:
ftp://ftp.soest.hawaii.edu/coastal/Climate%20Articles/Ocean%20warming%202010.pdf
Sure does look like RealClimate simply shifted or offset the Lyman et al (2010) data when they cut and pasted that Lyman et al curve to their graph. And here’s another RealClimate comparison graph of the same datasets. Obviously, they used different base years, because the Lyman et al curve has been shifted or offset to another location:
http://www.realclimate.org/index.php/archives/2010/05/ocean-heat-content-increases-update
Bob Tisdale -Regarding Willis, I was pointing out that your 2003 baseline didn’t have strong support.
And that doesn’t change the fact that, by shifting the offset, you are misrepresenting your trend as Hansens.
Tamino. Isn’t that a defunct failed auto attempt. El tamino. Fail.
And that doesn’t change the fact that, by shifting the offset, you are misrepresenting your trend as Hansens.
He may well be misrepresenting the prediction, as made in the model.
However, he can’t possibly be misrepresenting the trend, which is invariant of offset.
Kr:
Is the slope of the trend presented different than the slope of the trend from Hansen etal?
No sir…..it is not.
What is mis-represented here?
OHC, on a short timescale, is not following the trend of Hansen etal. Can you accept that? I know you can as it is just the facts.
Kim
Congratulations, you’ve now reached the fabled ‘He who must not be named’ status.
==================
ROTFLMAO