Bob Tisdale on GISS land/sea ratios

Land Surface Temperature Contribution To Non-Polar Temperatures

A recent version of global land-sea temperature variations in the  last 130 years.

Guest Post by Bob Tisdale

INTRODUCTION

Frank Lansner’s post Did GISS discover 30% more land in the Northern Hemisphere? at Jo Nova’s blog created a recent stir. Watts Up With That ran a similar post by Frank, Tipping point at GISS? Land and sea weight out of balance. Both posts spawned rebuttals/explanations, including Zeke Hausfather’s post The GISTemp Land Fraction at Lucia’s The Blackboard and my post Notes On The GISTEMP Ratio Of Land To Sea Surface Temperature Data, which Anthony Watts co-posted as GISS land and sea ratios revisited, on the same day that he ran Frank Lansner’s post.

Basically, Frank Lansner’s post contends that GISS has increased the ratio of land to sea surface data with time, from zero percent early in the 20th century to near 70% in recent years. With a close examination of the graph that was being presented as a reference, Frank’s land surface data looked unusual, and I believe Frank’s observations are skewed by his choice of base years, and possibly by his smoothing method. I discussed this with him in a detailed comment that I posted at WattsUpWithThat and Jo Nova’s website. Refer to:
https://wattsupwiththat.com/2010/07/17/tipping-point-at-giss-land-and-sea-out-of-balance/#comment-435038

OVERVIEW OF THIS POST

There are lingering beliefs that there’s something unusual about the way GISS handles land surface data. In an effort to dispel those misunderstandings, the land surface data contribution to combined land and sea surface temperature data will be illustrated in this post, using a very simple method. Sea surface temperature data will be subtracted from GISS, Hadley Centre, and NCDC combined (land and sea) surface temperature products. The remainders, which are the contributions of the land surface temperature data to the combined products, will be compared. Personally, I was surprised with the results. But first, we need to eliminate the effects of known differences between the GISS and the other two global temperature datasets.

GISS treats the polar regions differently than the Hadley Centre and NCDC. GISS has better land surface temperature data coverage than the Hadley Centre and NCDC in the Arctic and Antarctic. And the Hadley Centre and NCDC include Arctic and Southern Ocean sea surface temperature data as seasonal sea ice melts, while GISS deletes sea surface temperature from areas where there is seasonal sea ice. The treatment of polar data by GISS was discussed in GISS Deletes Arctic And Southern Ocean Sea Surface Temperature Data, which was also co-posted at WattsUpWithThat, GISS Deletes Arctic And Southern Ocean Sea Surface Temperature Data. So, due to those differences, this post will only examine the global temperature data between the latitudes of 60S and 60N. These latitudes represent approximately 85% of the surface area of the globe.

Note: It is also known there is little to no Antarctic land surface temperature data prior to the 1950s. But this can’t explain the results Frank Lansner was reporting.

DATA SOURCES

GISS and Hadley Centre combined (land and sea) surface temperature anomaly data were downloaded from the KNMI Climate Explorer:
http://climexp.knmi.nl/selectfield_obs.cgi?someone@somewhere

The KNMI Climate Explorer was also the source of Hadley Centre sea surface temperature (SST) anomaly data (HADSST2) used in its combined product. It also served as the source of the two SST components of the GISS combined product, HADISST from January 1880 to November 1981 and Reynolds (OI.v2) from December 1981 to present. The method employed to merge the two SST datasets used in the GISS product is discussed under Step 4 of the GISS current analysis webpage. The base years (1982 to 1992) used for splicing are different than those presented by the KNMI Climate Explorer for the GISS combined product, so I shifted the merged SST anomaly data to account for this.

I wanted to compare NCDC data in this post also, but it is not available through the KNMI Climate Explorer, and since the NCDC does not break down its standard combined surface temperature product into the desired latitude band (60S-60N) on its Global Surface Temperature Anomalies page, I used a second source. The NCDC also has SST, LST, and combined temperature anomaly data available through its ERSST Version 3/3b webpage, and it is available in multiple latitude bands, including 60S-60N. Scroll down to their link ftp://eclipse.ncdc.noaa.gov/pub/ersstv3b/pdo under the heading of “ASCII Time series Tables”.

Figure 1 compares the combined global (land and sea) surface temperature anomalies of the standard NCDC product and the data available through the ERSST.v3b webpage. The trends are identical at 0.057 deg C/decade. The difference appears to be caused by the use of different base years. The standard NCDC product uses 1901 to 2000 for base years while the data available through the ERSST.v3b webpage appears to be based on the NCDC climatology of 1971 to 2000. So this post uses the NCDC SST and combined (LST and SST) data that’s available through the ERSST.v3b webpage:
ftp://eclipse.ncdc.noaa.gov/pub/ersstv3b/pdo
http://i28.tinypic.com/vrsikk.jpg
Figure 1

Links to the data presented in Figure 1:
“Standard” NCDC Global combined product:
ftp://ftp.ncdc.noaa.gov/pub/data/anomalies/monthly.land_ocean.90S.90N.df_1901-2000mean.dat
NCDC Global combined product through ERSST.v3b webpage:
ftp://eclipse.ncdc.noaa.gov/pub/ersstv3b/pdo/aravg.mon.land_ocean.90S.90N.asc

Note: There are two different series of data available through the ERSST.v3b webpage. Those ending with .gv3.asc are recent additions, and since they have a slightly different trend and I have not found any mention of them in any other webpage, I have not used them in this post.

NON-POLAR SST AND COMBINED (LST & SST) SURFACE TEMPERATURE ANOMALY COMPARSONS

Figures 2 through 4 are comparison graphs of GISS, Hadley Centre, and NCDC SST and Combined (SST&LST) datasets for the latitudes of 60S-60N. All data has been smoothed with a 13-month running-average filter. They are being provided as references.
http://i26.tinypic.com/dmd2k1.jpg
Figure 2
XXXXXXXXXXXXXXXXXXXXX
http://i32.tinypic.com/2gvitja.jpg
Figure 3
XXXXXXXXXXXXXXXXXXXXX
http://i25.tinypic.com/258bcs6.jpg
Figure 4

CONTRIBUTION OF LAND SURFACE TEMPERATURE ANOMALY DATA TO NON-POLAR SURFACE TEMPERATURES

As discussed in the overview, to determine the contribution of land surface temperature anomaly data, the SST anomaly data was subtracted from the combined data. Before subtraction, the SST data was scaled by a factor of 0.755 to represent the ratio of ocean to land between the latitudes of 60S-60N. The remainders for each dataset are shown in Figures 5 through 7. Note that these graphs represent the remainder of subtraction, not the actual land surface temperature anomalies. To convert them back to land surface anomalies, they would have to be scaled.
http://i32.tinypic.com/14izupt.jpg
Figure 5
XXXXXXXXXXXXXXXXXXXXX
http://i29.tinypic.com/2qak75i.jpg
Figure 6
XXXXXXXXXXXXXXXXXXXXX
http://i30.tinypic.com/2ymc2vk.jpg
Figure 7

COMPARISON OF LAND SURFACE TEMPERATURE CONTRIBUTIONS

Figure 8 compares the remainders resulting from the subtraction of the scaled SST data from the combined (LST&SST) GISS, Hadley Centre, and NCDC products. As shown in Figures 5 through 7, the land surface residuals are noisy, so for this comparison, the data was smoothed with a 37-month running-average filter. While there are slight differences in the yearly and decadal variations in Figure 8, the linear trends for the three datasets are basically the same, differing only 0.001 deg C/decade.
http://i31.tinypic.com/33cp74w.jpg
Figure 8

This suggests, for the latitudes of 60S-60N, the differences between the combined products from GISS, Hadley Centre, and NCDC result from differences between the SST data they employ. Refer to An Overview Of Sea Surface Temperature Datasets Used In Global Temperature Products. And the most significant differences in SST anomalies occur before 1940. From 1880 to 1940, the SST anomaly data used by the Hadley Centre and NCDC have significant dips and rebounds, as shown in Figure 3 and 4, while the dip and rebound is much less pronounced in the SST data used by GISS, Figure 2.

SO HOW DOES GISS LAND SURFACE DATA DIFFER FROM THE HADLEY CENTRE AND NCDC?

The Hadley Centre and NCDC land surface temperature anomaly datasets represent continental land masses only, and on a global basis, both of those datasets exclude Antarctica. The land surface data presented by GISS, on the other hand, includes continental land mass data plus much more. First, looking at Figure 9, continental land mass data is extended out over the oceans in the GISS land surface temperature product with 1200km radius smoothing. (Figure 9 is a trend map available through the GISS Global Maps webpage, and it shows the regional changes in temperature anomaly from 1880 to 2009. I’ve cropped the map to show the latitudes, 60S-60N, used in this post.)
http://i31.tinypic.com/4kd9ns.jpg
Figure 9

Second, GISS also includes data from island surface stations and from “Ship stations,” and these values are also extended out over the oceans. This GISS dataset is a carryover from the methods developed by GISS back in the 1980s, when SST datasets were incomplete. They were attempting to simulate global temperature anomalies without using SST data. This is explained further in a WUWT comment from Zeke Hausfather to Frank Lansner in which Zeke quotes from a correspondence from Dr. Reto Ruedy of GISS:
https://wattsupwiththat.com/2010/07/17/tipping-point-at-giss-land-and-sea-out-of-balance/#comment-434647

In part it reads, “The curve NCDC and most likely you are computing shows the mean temperature over the land area (which covers about 1/3 of the globe, a large part of it located in the Northern hemisphere).

“None of our graphs represents that quantity. We could obtain it by creating a series of maps, then averaging just over the land areas (similar to what we do to get the US graph).”

It continues, “Since our interest is in the total energy contained in the atmosphere which correlates well with the global mean surface temperature, all our graphs display estimates for the global mean, the ones based on station data only as well as the ones based on a combination of station and ship and satellite data. Obviously, the latter is the more realistic estimate and we keep the first one mostly for the following historical reason:

“When we started out in the 1980s analyzing available temperature data, historic ocean temperature data were not yet available and we did the best we could with station data. As soon as ocean data compilations became available, we used them to refine our estimates (calling it LOTI). But we kept the earlier estimates also, mostly for sentimental reasons; they are rarely if ever mentioned in our discussions (see also the ‘note’ in the ‘Table’ section of our main web site).”

And continuing this post, the “‘note’ in the ‘Table’ section of [the GISTEMP] main web site” reads, “Note: LOTI provides a more realistic representation of the global mean trends than dTs below; it slightly underestimates warming or cooling trends, since the much larger heat capacity of water compared to air causes a slower and diminished reaction to changes; dTs on the other hand overestimates trends, since it disregards most of the dampening effects of the oceans that cover about two thirds of the earth’s surface.

And again, LOTI represents the GISTEMP combined land and sea surface data and the dTs represents the land surface data.

CLOSING REMARKS

The GISS Global-mean monthly, seasonal, and annual means dataset does not represent continental land mass temperature anomalies as many believe. It, therefore, cannot be employed in analyses like the one Frank Lansner is attempting to perform.

Dr. Ruedy’s statement that GISS could create a continental land temperature anomaly dataset similar to NCDC “by creating a series of maps, then averaging just over the land areas,” is another way of saying they could create it by masking the areas where the land surface data extends out over the oceans. And this was noted in the post Notes On The GISTEMP Ratio Of Land To Sea Surface Temperature Data.

Then there’s the similarity in the linear trends of the land surface contributions for the three combined datasets, Figure 8. It confirms the findings of the independent researchers who are creating land surface temperature anomaly datasets: the results are pretty much the same as the GISS, Hadley Centre, and NCDC data.

Gust Post by Bob Tisdale
Advertisements

104 thoughts on “Bob Tisdale on GISS land/sea ratios

  1. As an “alarmed” layman trying to make sense of the conflicting claims of the AGW controversy, I proposed to my Senator a way to sort the whole matter out, once and for all.

    In particular, I proposed a “High Carbon Lifestyle” tax for those with family incomes greater than $250,00/year and businesses with similar incomes. My fantasy bill would impose a surtax that would proportionately apply to taxpayers whose residential properties exceed 1,2000 sq. ft./resident (all residences, worldwide, considered and rising to a 100% surcharge if the ratio exceeds 5,000 sq. ft./resident). A similar surcharge for private aircraft travel. Also, elimination of the business deduction for business travel by airplane and elimination of all government funding for travel to conferences (to encourage video conferencing). Other provisions may be usefully added to the above.

    The rationale for my proposal is that if our wealthiest and most powerful citizenry are willing to make draconian sacrifices to their lifestyle to reduce carbon consumption, then the AGW concerns are likely well-founded–otherwise, an alarmist hustle. Also, my proposal would allow our “betters” to lead from the front, for once, and lead by example.

    A “little guy” way to sort out the AGW business, but I think it’s once that has something to recommend it.

  2. I am reading this as: The land surface temperature is the smaller portion of the dataset warming anomaly. The SST is the dominant factor here.
    So, when I look at SST data and anomaly, I see the oceans cooling, and with it goes the global temperature.
    ABC World News Tonight was trying to poke fun at Sen. Inhofe tonight, suggesting that “NOAA says” hottest ever is the trump card.
    The Senator says the globe is cooling, and he’s correct. The newsman kept cutting the Senator off.
    The interviewer’s parting suggestion of “NOAA says” was a cheap shot off interview.

  3. Well of course the land contribution will be warmer now with all the rural sites disappearing from the dataset leaving more urban/airport readings in the mix, MMTS’s in bad sites, etc. Skewed raw data (with some that just magically appears to fill the gaps) that goes through a meat grinder with the past seemingly getting colder. Before you know it, I’ll be thinking my grandpa was slinging spears at woolly mammoths.

  4. But the fundamental error BEGINS with Hansen’s changing/editing/”updating” of the raw data differently in his original 1990 plot of land temperature records since 1910, and the more recent highly skewed results of 2002 and 2008.

    Before, in his original”research” – which I only trust because that was before he began making his money and reputation selling CAGW – the 1940’s temperatures were higher than the 1990 temperatures. Later, as he needed more and more dramatic “evidence” of CAGW to maintain his lifestyle and his “religion” of CAGW, the 1940’s got cooler an cooler with new paper.

    What is the raw data trends of sites that have no UHI? What is the tend in sites that have NOT been corrupted by (false ?) TOBS changes by interested observers?

    What are the real thermometer readings across the 20th century US countryside?

  5. Zeke’s post wasn’t and explanation, it was making excuses for some pretty lousy work by Hansen/GISS

  6. 1. Technical point (and I don’t think that it sways the results, but can be important sometimes): Do you compute the trends on the smoothed graphs or the real data? you should do them on the latter, but use smoothing for display purposes,

    2. The recalcitrance of Lasner is typical with what I see in skeptic world and we kill ourselves with it. Way too eager to want to believe something on our side and DOGGED in giving it up instead of fair. Also, just a bit of a logic and mental gap and not quite as shrewd as we ought to be analytically.

  7. I have one nagging question about this all ordeal:
    What is called Normal temp. as a reference to Anomally? Who decides where the Normal point is and why? Beacause if we concider the long range, the Normal point is way way down, since most of the time Earth is in glacial periods, with short warm periods like now (which is also in its end).

  8. mike says:
    July 23, 2010 at 7:47 pm

    “As an “alarmed” layman trying to make sense of the conflicting claims of the AGW controversy, I proposed to my Senator a way to sort the whole matter out, once and for all………….”

    Mike, I like the sentiment, but it’s wrong. Your proposed tax, like many of the taxes today, punish the successful segment of our society for no other reason than they are successful. I propose that no new taxes be levied that carry an exemption for anybody. If the tax is just, then all should share the burden. It is unreasonable to expect the fruits of our nation without laboring for them.

    BTW, any significant “business” that I ever engage in requires up close and personal interactions. As far as video conferences go……I’ve never been part of one that wasn’t a complete waist of time. But that’s just my experience.

  9. I answered Bob on his “climate Observations” and Bob wrote back:
    ***************
    Frank Lansner said…
    Dear Bob.

    I offered you a bottle of wine if you would go through my new article:
    http://hidethedecline.eu/pages/posts/the-perplexing-temperature-data-published-1974-84-and-recent-temperature-data-180.php

    If you had done this, you would certainly not have written as you do.

    One more time. Please look at PART 2 of my article, chapter 3.4.
    This is where I explain that GISS inlcudes ocean in their station data series, and where I show a graphic of the ocean included.

    So if anyone is aware of this, its me. The fact that you and others keep writing that you think im not aware of ocean data in GISS station “land” data might be my fault due to bad communication.

    I wrote in my article PART 4:

    “I am sure that the algorithm or specific method used by GISS to combine Land temperature and SST explains some of these apparently odd findings. But whatever the “algorithm” used by GISS is, can it be justified that GISS gradually weights the warm NH-Land graph more and more? And ends up with around 67% NH land fraction in 2007 although NH only has 40% land? Maybe, this algorithm or method deserves some attention?

    And in the WUWT article i write: https://wattsupwiththat.com/2010/07/17/tipping-point-at-giss-land-and-sea-out-of-balance/#more-22126

    I write:
    “In general GISS defends use of larger land fraction due to their 1200km zones around land stations reaching some Ocean areas. But this does obviously not explain a land fraction that appears to go from near zero to around 70% globally during the 20th century.

    Now, Your article, Bob:
    You focus on the similarities between CRU and GISS – i suppose to say that the resulting GISS is ok?
    The thing is, CRU and GISS ends up rather alike. But in CRU data i find much more direct land data adjustment than for GISS. On the contrary for GISS, the direct land data adjustments are not so big at all (to my surprice) but in stead the GISS warming trend thats similar to CRU comes when combining the SST and “land”.

    SOmething thats messy in al this is, that you seem to trust that CRU land is not ocean while GISS is… Yes yes, GISS has ship and island data included, but a BIG part of the GISS ocean area in their “land” data is obviousy from coastal stations. These stations are exactly the same as for CRU. So its nonsense to say “CRU is just land data”.
    Just becasue CRU says that their coastal stations are land while GISS (the same) coastal stations covers huge ocean areas, you cant just treat the same data as if completely different.

    I have raised some serious problems in data, and I know you disagree strongly, but i have not seen conving arguments from you, its not bad will.

    K.R. Frank

    July 23, 2010 7:25 PM

    *****************
    Bob Tisdale said…
    Frank Lansner: You wrote, “I offered you a bottle of wine if you would go through my new article:
    http://hidethedecline.eu/pages/posts/the-perplexing-temperature-data-published-1974-84-and-recent-temperature-data-180.php

    “If you had done this, you would certainly not have written as you do.”

    It was my intent to eventually address your request, in a few days. I have other priorities. Also, I don’t drink alcohol. I gave it up.

    You wrote, “One more time. Please look at PART 2 of my article, chapter 3.4.
    This is where I explain that GISS inlcudes ocean in their station data series, and where I show a graphic of the ocean included.”

    But your analysis where you assume that GISS increases land surface area does not address this.

    You wrote, “I am sure that the algorithm or specific method used by GISS to combine Land temperature and SST explains some of these apparently odd findings. But whatever the ‘algorithm’ used by GISS is, can it be justified that GISS gradually weights the warm NH-Land graph more and more? And ends up with around 67% NH land fraction in 2007 although NH only has 40% land? Maybe, this algorithm or method deserves some attention?”

    But GISS does NOT weight “the warm NH-Land graph more and more.” You need to mask the areas where land surface data extends out over the oceans in your analysis.

    You wrote, “In general GISS defends use of larger land fraction due to their 1200km zones around land stations reaching some Ocean areas. But this does obviously not explain a land fraction that appears to go from near zero to around 70% globally during the 20th century.”

    Please provide a link to an article written by a member of GISS where “GISS defends use of larger land fraction due to their 1200km zones around land stations reaching some Ocean areas.”

    You wrote, “You focus on the similarities between CRU and GISS – i suppose to say that the resulting GISS is ok?”

    I also included NCDC data. Or did you miss that? This post was not about the accuracy of land surface temperatures; it was about the contribution of land surface temperature readings to combined land and sea surface temperature data. And there is basically no difference between those of GISS, Hadley Centre, and NCDC.

    July 23, 2010 8:34 PM

  10. So we have:
    Bob says :
    **
    You wrote, “One more time. Please look at PART 2 of my article, chapter 3.4.
    This is where I explain that GISS inlcudes ocean in their station data series, and where I show a graphic of the ocean included.”

    But your analysis where you assume that GISS increases land surface area does not address this.
    **
    Yes it does, i just gave you a few of the examples, again from WUWT:
    “In general GISS defends use of larger land fraction due to their 1200km zones around land stations reaching some Ocean areas. But this does obviously not explain a land fraction that appears to go from near zero to around 70% globally during the 20th century.

    To this you suggest that giss never said so, but thats another matter. GISS has used their 1200 km radius since Hansen and Lefebeff in1987 where they argue for this to be true.

    And thats why i stressed, the the ODD thing is that this “land” data from GISS goes from a near blank ZERO in twenty years 1900-1920 and is then weighted strongly in recent years. Thats the problem! And as I said, if its not clear from my article its my fault, this I admit. But my point is strong and important non the less.

    In my WUWT article
    https://wattsupwiththat.com/2010/07/17/tipping-point-at-giss-land-and-sea-out-of-balance/#more-22126
    i even showed a graph where the land trends and Ocean trens goes opposite 1970-74 (BOTH using smoothed and 5 yr trend graphs) . So honestly i think the land vs. ocean issue was adresses in my article.

    K.R. Frank

  11. Bob then to your article, you focus on that there are land similarities between GISS, CRU and NCDC, right?

    For me, a similarity between those three institution is simply not an argument in these climate gate times, im sorry.
    I may sound fanatic, but if eferything that these 3 institution agrees about is just a little true, i would have to turn alarmist pronto ;-)

    This is why I chose to do a large work to create this article:
    http://hidethedecline.eu/pages/posts/the-perplexing-temperature-data-published-1974-84-and-recent-temperature-data-180.php

    -where i compare data publisher BEFORE the GW movement started an with data after and more.

    K.R. Frank

  12. Hi Bob,

    Thanks..I enjoy your articles.
    Figure 8 fits with my life experience in tropical Queensland, Australia in the periods since 1940. The time up to 1955 had dense cloud cover, many cyclones, and huge floods in summer periods. It got dryer, then in 1965, we went again back to that very wet cloudy pattern which reduced slowly after about 1976. It has been less cyclones, less floods and warmer temperatures since then. Less water vapour in the air, for whatever reasons, around the Equator seems to be a possibility..Be it Volcanos or whatever. I don’t believe CO2 alone since 1976 can cause that change, contary to the CSIRO’s constant barrage on our media.
    Politics and science are like oil and water..!

  13. One of the more interesting points raised by Frank is the departure of land temps from SST in more recent years, whereas in earlier days, the SST and land temps track each other closely. This is very clear in the above graphs in Bob’s article as well. A very strong indication that land temps are strongly contaminated by UHI (or strongly “homogenized” in recent years.

    In his 4 part article linked to in his post above, Frank also compares land temps with UAH/RSS satellite records and find the same tendency. Its seems a strong indication that recent recorded warming in all the global temp products is for a large part down to UHI. This seems especially true for the GISS product.

    I recommend reading Frank’s 4 part article. He raises a number of interesting points and questions beside this point that Bob now focuses on.
    http://hidethedecline.eu/pages/posts/the-perplexing-temperature-data-published-1974-84-and-recent-temperature-data-180.php

  14. Frank Lansner: My comment at my blog was, “But your analysis where you assume that GISS increases land surface area does not address this.”

    Your reply here, “Yes it does, i just gave you a few of the examples, again from WUWT:
    “In general GISS defends use of larger land fraction due to their 1200km zones around land stations reaching some Ocean areas. But this does obviously not explain a land fraction that appears to go from near zero to around 70% globally during the 20th century.”
    ##################

    Frank: Let me rephrase my reply to you. But your analysis where you assume that GISS increases land surface area does not address this, because you do not mask the ocean areas where GISS has extended the land surface data out over the oceans. I’ve noted this numerous times. I wrote it in my post from last week…
    https://wattsupwiththat.com/2010/07/17/giss-land-and-sea-ratios-revisited/
    There I wrote, “So in order to properly create a weighted average of GISTEMP land and sea surface temperature data with 1200km radius smoothing, the land surface data where it extends out over the oceans would first need to be masked.”

    I also noted it in a detailed comment to you that I posted at Jo Nova’s and at WUWT that I linked in this post. Here’s the link again:
    https://wattsupwiththat.com/2010/07/17/tipping-point-at-giss-land-and-sea-out-of-balance/#comment-435038

    There, I ended with, “4.From an earlier comment at Jo Nova’s, you need to mask the land surface data.”

    Have you masked the land surface data where it extends out over the oceans? No.

  15. Bob, thanks for good answer.

    You write: “Have you masked the land surface data where it extends out over the oceans? ”
    No but I did a much better compare of UAH land vs. the different “land” graphs available.
    As GeorgeGr just wrote:
    “I recommend reading Frank’s 4 part article. He raises a number of interesting points and questions beside this point that Bob now focuses on.
    http://hidethedecline.eu/pages/posts/the-perplexing-temperature-data-published-1974-84-and-recent-temperature-data-180.php

    The idea you have to exactly retrieve land data from these GHCN based temperature series from GISS, CRU etc. is very very complicated an inaccurate as i told you. The very same temperature data from all the “coastal” stations (the ones less than 1200 km from a coast …. !!!) are interpreted as partly ocean by GISS and yet the very same data from CRU is not viewed this way..

    So if you really want to use GHCN based station data to learn about PURE land data, then you would have to mask using method “More than 1200 from a coast” – isnt that so?

    But I have found 2 methods to learn more about these things, to validate these “land” data further, I will publish within a few days.

    K.R, Frank

  16. Frank Lansner: You replied above, “To this you suggest that giss never said so, but thats another matter. GISS has used their 1200 km radius since Hansen and Lefebeff in1987 where they argue for this to be true.”

    To prove my point, that is a great example to use, Frank. In Hansen and Lebedeff (1987)…
    http://pubs.giss.nasa.gov/docs/1987/1987_Hansen_Lebedeff.pdf
    …they extended the 1200km smoothing out over the oceans because sea surface temperature data was so sparse. In other words, they did not merge it with sea surface temperature data. They extended the continental land, island, and ship surface station out over the oceans to simulate air temperature over the oceans.

    You, however, take that dataset and assume it represents only continental land mass temperature anomalies. It does not. It represent much more.

    You wrote, “And thats why i stressed, the the ODD thing is that this ‘land’ data from GISS goes from a near blank ZERO in twenty years 1900-1920 and is then weighted strongly in recent years. Thats the problem!”

    You are using the GISS land surface data incorrectly, Frank. Again, you need to mask it where it extends out over the oceans. In the post above, I even included a quote to Dr. Ruedy from GISS. Here it is again, “The curve NCDC and most likely you are computing shows the mean temperature over the land area (which covers about 1/3 of the globe, a large part of it located in the Northern hemisphere).” He continues, “None of our graphs represents that quantity. We could obtain it by creating a series of maps, then averaging just over the land areas (similar to what we do to get the US graph).”

    In other words, GISS could create a land surface dataset for continental land masses but they would need to mask where the data extends out over the oceans.

    You wrote, “i even showed a graph where the land trends and Ocean trens goes opposite 1970-74 (BOTH using smoothed and 5 yr trend graphs) . So honestly i think the land vs. ocean issue was adresses in my article.”

    And I noted in a comment that ran at Jo Nova’s and WUWT that your land surface data looked unusual. Again, here’s a a link to that comment:
    https://wattsupwiththat.com/2010/07/17/tipping-point-at-giss-land-and-sea-out-of-balance/#comment-435038

    You have since replied that you’ve tried other base years and see no difference. But I find that curious since I have shown you the differences are significant depending on the selection of base years. Here’s my attempt to reproduce your graph, using 1951-1980 as base years:

    And here’s the same graph but in this case the base years are 1900 to 2000:

    I will again ask you, What method of smoothing are you using? It is not a centered 60-month running-mean filter. Otherwise, your data would not have extended for the full term of the data.
    Is it weighted to one side?

    And have you tried a centered 60-month running-mean filter, as I asked, to see if that changes your values?

    You ended your comment with, “In my WUWT article
    https://wattsupwiththat.com/2010/07/17/tipping-point-at-giss-land-and-sea-out-of-balance/#more-22126
    i even showed a graph where the land trends and Ocean trens goes opposite 1970-74 (BOTH using smoothed and 5 yr trend graphs) . So honestly i think the land vs. ocean issue was adresses in my article.”

    I don’t believe you have since you have not redone your analysis with the land surface data masked over the oceans, Frank. You also haven’t determined the impact of your method of smoothing. And while you claim to have tried other base years, you have not presented them. Did you try using the base years of 1880 to 2009?

  17. trbixler says: “.3 degree C over 130 years?”

    I assume you’re referring to Figure 8. As I noted in the post, …these graphs represent the remainder of subtraction, not the actual land surface temperature anomalies. To convert them back to land surface anomalies, they would have to be scaled.

  18. # Frank Lansner says: July 23, 2010 at 11:02 pm (concerning your interesting paper, at hidethedecline/12th July, 2010)
    Part 4: Perhaps we can say: „Land Temp = Sea Surface Tem“; and
    Part 1: “Phil Jones, 1985, about the temperature decline after the 1930´ies: “No satisfactory explanation for this cooling exists, and the cooling itself is perplexing because it is contrary to the trend expected from increasing atmospheric CO2 concentration.”

    It would presumably more correct to say: __Sea Surface Temp govern Land Temp__, and if this principle would be more observed, the reason for the temperature decline after the 1930s could be found by analysing the impact of WWII, and the naval activities at sea since September 1939, which was immediately followed by the extreme winters 1939/40, 1940/41 and 1941/41, the coldest winters in Northern Europe for more than 100 years , and when the naval war went global after Pearl Harbour (07 Dec. 1941) , these activities may have contributed to the yet still not explained three decades global cooling (1940- mid 1970).
    Concerning winter 1939/40 in short (top left column): http://www.oceanclimate.de/A_Large_Scale_Experiment_with_Climate/A_Large_Scale_Experiment_with_Climate.pdf
    Concerning WWII: http://climate-ocean.com/

  19. PolyisTCOandbanned says: “Technical point (and I don’t think that it sways the results, but can be important sometimes): Do you compute the trends on the smoothed graphs or the real data? you should do them on the latter, but use smoothing for display purposes,”

    I computed them on the smoothed data. Only the y-intercepts are different than the original Figure 8:

    It made (and most times makes) no difference in the trends, so I took a shortcut and displayed only one graph.

    Regards

  20. rbateman says: “I am reading this as: The land surface temperature is the smaller portion of the dataset warming anomaly. The SST is the dominant factor here.”

    SST data represents 70% of the surface temperature anomalies globally, and in this post with the latitudes of 60S-60N, it represents 75.5% of the data. SST is the dominant dataset. Land surface temperatures mimic and exaggerate the changes in SST due to a number of reasons, polar amplification, UHI, poor station siting, etc.

  21. It is unscientific to discuss changes of tenths of degrees a century for the globe, when the errors are hugely larger.

    Part of the misuse of science involves the vexed “anomaly” method of expression.

    If you select a time period as a reference base, you have to change the reference value each time a station in the reference period is added, removed or altered. Since this is a near-continuous process, you can derive almost any “anomaly” story you wish, simply by choice of a reference set favourable to your argument. Or, for that matter, the temperatures outside the reference period, which are also added, dropped and adjusted almost continuously.

    Who is interested in a fight about moving targets that are presented as a stationary time series?

    Who is interested in fairy story 1200 km interpolations when stations a few km apart routinely report differences of more than a deg C?

    Who is interested in filling in missing data by averaging, when the very reason for the data being missing can be because of anomalous weather?

    Why, oh why cannot we simply work in real physical numbers, even if they are deg C rather than K? Why can we not select a set of truly rural stations – and then start the process of refinement and comparison?

    It is untenable that much past proxy work is (probably) wrongly calibrated because one cannot get an agreed and plausible instrumented temperature series.

    It’s fair amazing how the 1945-1970 decline has been steadily brought to near level over the years, by successive adjustments. Some must be phony.

    Time to lift the game. Clean slates, please.

  22. e_por says: “I have one nagging question about this all ordeal:
    What is called Normal temp. as a reference to Anomally?”

    Each of the global temperature anomaly data producers (GISS, Hadley Centre, and NCDC) selects the base years. And NCDC uses a number of base years, depending on the dataset. But as you’ll note in my Figure 1 above, the choice of base years has no real impact on the long-term trend. It shifts the data up and down, changing the years on the graph when the data intersects with zero degrees. There also are some minor changes in the month-to-month variations but they are the same each year.

  23. GeorgeGr says: “One of the more interesting points raised by Frank is the departure of land temps from SST in more recent years, whereas in earlier days, the SST and land temps track each other closely. This is very clear in the above graphs in Bob’s article as well. A very strong indication that land temps are strongly contaminated by UHI (or strongly “homogenized” in recent years.”

    The divergence of land surface temperatures from sea surface temperatures is not only impacted by UHI. Land surface temperatures are also affected by land use changes, polar amplification, thermal inertia, etc. Does Frank account for these in his analysis?

  24. I’m interested in Bob Tisdale’s work. He seems to be a thorough worker. But I can only concentrate on so much of the details before my mind starts to jump… often to the next commentator. If I try to force myself to “concentrate” I simply fall asleep. I know. I’ve fallen asleep more often at the keyboard doing Climate Science than I’ve fallen asleep in bed over the last couple years!

    I’m interested in Frank Lansner too. But I cannot disentangle which is “correct”, Frank or Bob, without falling asleep! However, my feeling is that while Bob may be more thorough on details in his area of expertise, Frank has a bigger picture… and that bigger picture, which includes the UHI/station degradation, is of great importance… and holds my interest. I trust my feelings, at least to indicate something worth examining a little more closely. Frank may also be right on the details too… or Bob may be. How can I tell – without falling asleep again?

    How many others have experiences like this? Yet the official science papers are far worse. No wonder scientists flounder outside their expertise – so that some will say, Miskolczi / G&T is spot on… and others will say, Garbage… and all I can manage is hunches which are more humble than many commentators’ comments, but which I still trust and just keep quietly to myself, looking for bite-size evidence, rather than open to the flood of “expertise”/”disproof” of a Nick Stokes or a Ferdinand Engelbeen, that just overwhelms me and sends me to sleep again.

    Courtesy, and not losing sight of human scale, are really important here. I have the feeling that Bob and Frank are gradually rubbing expertise off each other… and I look forward to the time when they will come to agreement in ways that I too can cope with, and understand.

  25. Frank Lansner asked, “So if you really want to use GHCN based station data to learn about PURE land data, then you would have to mask using method “More than 1200 from a coast” – isnt that so?”

    Are you saying that only land surface data inland 1200km from the coast should be evaluated? If so, why?

  26. ArndB says: “It would presumably more correct to say: __Sea Surface Temp govern Land Temp__, and if this principle would be more observed, the reason for the temperature decline after the 1930s could be found by analysing the impact of WWII…”

    If one assumes the global oceans integrate the effects of El Nino and La Nina events, the temperature decline after the mid-1940s could be explained as a change in the frequency and magnitude of ENSO events. Refer to my post “Reproducing Global Temperature Anomalies With Natural Forcings”:
    http://bobtisdale.blogspot.com/2009/01/reproducing-global-temperature.html

  27. Frank Lansner: You wrote, :The idea you have to exactly retrieve land data from these GHCN based temperature series from GISS, CRU etc. is very very complicated an inaccurate as i told you. ”

    First observation about your comment. you write abouut GISS and CRU land surface temperature data as if they represent the same thing. As I have written to you repeatedly, GISS “land surface temperarure” data includes much more than the continental land masses. It is not the same as CRUTEM data.

    There is nothing complicated about retrieving NCDC or CRU or GISS land surface data. They are readily available on a gridded basis through the KNMI Climate Explorer. The problems arise when one assumes that GISS land surface data represents the same data as CRU and NCDC.

  28. Hi Bob.

    “You, however, take that dataset and assume it represents only continental land mass temperature anomalies. ”
    Again, just one of the examples, from the WUWT writing:

    “In general GISS defends use of larger land fraction due to their 1200km zones around land stations reaching some Ocean areas. But this does obviously not explain a land fraction that appears to go from near zero to around 70% globally during the 20th century.

    Maybe its my English (??) but this text comes just after the essential graph in my WUWT article, it was ment to say:
    “YES, according to GISS the “land” data is supposed to be not only land , but it does not explain that from 1900-20 the GISS “LAND” was weighted zero, and then recently weighted strongly.”

    Again, i may not have communicated this well enough.
    I am writing on a follow up, and i think this will make more obvious some issues of GISS “land” temperatures, and weighting hereof.

    K.R. Frank

  29. Frank Lansner: You wrote, “Bob then to your article, you focus on that there are land similarities between GISS, CRU and NCDC, right?” And you continued, “For me, a similarity between those three institution is simply not an argument in these climate gate times, im sorry. I may sound fanatic, but if eferything that these 3 institution agrees about is just a little true, i would have to turn alarmist pronto ;-)”

    This post has nothing to do with “climategate.” Why are you introducing the topic? It appears as though you are trying to redirect the conversation to something other than the errors in your analysis.

    In this post I removed a known difference between GISS and the other two datasets by limiting the data to non-polar latitudes. I then subtracted the scaled SST anomalies from the combined land and sea surface temperature anomalies. The remainders from the three datasets (the contributions of the land surface data) are remarkably similar:

    What does that imply to you, Frank? To me, it implies that with all of the adjustments made to the data by GISS, Hadley Centre and NCDC, there is little effect on the long-term trend. To me it also implies that GISS does not weight the continental land surface temperature data differently than the Hadley Centre or NCDC. And this contradicts your posts.

  30. Frank Lansner July 24, 2010 at 5:21 am: You wrote in part, “But this does obviously not explain a land fraction that appears to go from near zero to around 70% globally during the 20th century.”

    You then rewrote it as, ““YES, according to GISS the ‘land’ data is supposed to be not only land , but it does not explain that from 1900-20 the GISS ‘LAND’ was weighted zero, and then recently weighted strongly.”

    And this is the basis for this post and the basis for my post from last week. GISS does not change the land fraction so that it “appears to go from near zero to around 70% globally during the 20th century.” You need to mask where the land data extends out over the oceans in order to perform the type of analysis you are attempting. You also need to assure that your base years and method of smoothing do not skew your results, and as far as I know you have elected not to do these things. Why?

    Also, I have presented a number of questions to you above, and you have not replied to them. Why? I reply to your questions.

  31. I propose an “alarmed layperson carbon lifestyle” tax. After all, the block of population that earns income above $250,000 is so small that it would not make a dent in our unenlightened carbon emitting ways. The layperson tax would hit a larger group of citizens (who needs to drive to work anyway?) and thus have more impact. Our alarmists should light the way through self imposed dracnonian sacrifices in their daily lives. We can then measure the effects and go from there.
    Sarcasm aside, it’s funny to see how some can just say that others shoud pay more for no other reason than that they have money.
    Mike, I need money. Please give it to me. I swear that I will use it for a more worthy cause than you will. I swear.

  32. Bob
    Yes, what I was struck by was the small difference and the possibility of in that period of time that the accuracy of the source put the results in the noise.
    Thanks for the response.

  33. Bob Tisdale says: July 24, 2010 at 4:59 am, (Rely to: ArndB, July 24, 2010 at 3:27 am )
    “If one assumes the global oceans integrate the effects of El Nino and La Nina events, the temperature decline after the mid-1940s could be explained as a change in the frequency and magnitude of ENSO events.”

    By ocean dimension the El Nino and La Nina events are “peanuts”, as El Nino represents a temporary warm water pool over a small area of the entire ocean space down to 50 depth, but they are the most significant of the known and investigated events caused by the sea. But the status of the oceans is driving the atmosphere at any time, which is quickly observed when in coastal areas the sea wind (afternoon) changes to land wind (early morning). If an El Nino can have a distant impact, any other change of the ocean surface layer of a few dozen meters alters the running weather as well. The claim that there had been a prolonged El Nino from 1939 to 1942 may have contributed, but does not explain the extraordinary three war winters in Europe 1939-1942, and it is a great pity that science has been reluctant to investigate their causation. The massive interference of naval war on the marine environment from 1939 to 1945 could be seen as a large-scale experiment concerning the impact of human activities on weather and climate. It would contribute significantly to understand the ocean-atmosphere process, and the relevance of the oceans better.
    Always interested in your fine work; Regards ArndB

  34. Please, Frank, you keep iterating the same nonsense over and over. The rebuttals of Bob, Zeke and other should have given you pause long ago about the correctness and relevance of your analysis. Please ponder what Bob Tisdale is saying.

    Or go back to the Lucia thread and read what eg carrot eater (eg. Comment#49424) has to say to you:

    http://rankexploits.com/musings/2010/the-gistemp-land-fraction/#comments

    You know that I do not not subscribe to CAGW but I wish you would abandon this topic. It is a dead end.

  35. Lucy Skywalker: I completely agree with you and I just don’t have the time or energy to delve into the data enough to make my own conclusion. It puts me to sleep just thinking about all the effort these two guys have given this subject.

    And while it is important we get all the technical details correct, I think this issue is just that, a technical one. It is not the most important one. I think we can trust that GISS, Hadley Centre and NCDC all come up with pretty much the same results. However, I don’t think those results are correct due the homogenization, adjusting, poor siting, UHI problems, etc. that has been done to the raw data. I’m still putting my faith in the satellite record, which clearly says, 2010 isn’t the warmest year of record. It is too bad we don’t have satellite data for the 30s. What we are experiencing now, at least in the US, isn’t anything like those years.

  36. As the “alarmed” layperson of an earlier posting, I can see from the couple of comments my “High Carbon Lifestyle Tax” proposal has garnered that I didn’t make my point clearly.

    Let me try again. My proposed tax is not intended to raise money and it is not intended to allow anyone to shirk a necessary shared sacrifice. Rather, it is intended to “smoke out” the truth of the conflicting AGW claims. If great sacrifices are required by society, in terms of carbon-reduced lifestyle, to save us all from being burnt to a AGW crisp, then I, as little guy, am more than willing to do my part. But before I “bite the bullet” at the say so of certain enlightened folks, at least some of whom are poised to make big bucks on alternative energy and carbon trading, I want to first observe our wealthiest and most powerful citizens showing the way as we transition to a world of rabbit hutches and bike paths. Leadership from the front and by example, in other words.

    My proposal would ensure proper leadership by our elite or confiscation of their wealth by taxation (which wealth could be better used outfitting their up-scale rabbit hutches and bicyles in ostentatious, low-carbon splendor). Then, once us “small people” see our best and brightest making such sacrifices we could be sure the planet really is in a “jam” and we need to shoulder similar sacrifices, as well. On the other hand, if such a tax is not palatable to the wealthiest and best connected of our citizenry, then most likely the whole AGW business is a hustle by some rich guys trying to get richer and us cannon-fodder can blow the whole scare off and get back to guiltless Sunday drives in our SUV’s. My proposal assumes, of course, that the elite tier of our society is enlightened enough to moderate their opulent life-style if they knew and truly believed such life-style downsizing was necessary to save Gaia from becoming a Venusian hell-hole. I hope I have not made an unreasonable assumption.

    To put a fine point on the matter, the science of AGW is of interest, but the public policy that flows from it is the item of greatest concern to us all–especially since so much doubt has been raised about climate science bias and politicization. Hence my proposal as a means to get to the heart of the public policy issue. Fish or cut bait.

  37. ArndB: You wrote, “The claim that there had been a prolonged El Nino from 1939 to 1942 may have contributed, but does not explain the extraordinary three war winters in Europe 1939-1942, and it is a great pity that science has been reluctant to investigate their causation.”

    I believe the European wartime winters have been studied. I recall reading something about them (a paper?) within the last five years, though I don;t have any idea when the paper was written. And recall Tamino writing something within that last two years, so that may be a place to start looking for links. But then you went on to discount ENSO events.

    You wrote, “By ocean dimension the El Nino and La Nina events are ‘peanuts’…”

    Peanuts? The equatorial Pacific extends approximately 170 deg longitude or almost half way around the globe. The tropical Pacific SST anomalies are directly impacted by ENSO events and they can peaked at 0.9 deg C (1997/98 El Nino), then dropped to -0.34 deg C during the subsequent La Nina, and that’s even with the dipole effect between the eastern and western tropical Pacific. The last time I checked the tropics represent more than one-third of the globe, so we’re looking at about one-sixth of the surface area of the globe that is directly impacted by ENSO events. Based on weekly SST data for the 1997/98 El Nino, NINO3 SST anomalies peaked at more than 3.6 deg C and NINO1+2 SST anomalies peaked at 4.4 deg C. What else? Through coastally trapped Kelvin waves, ENSO events also directly impact the sea surface temperatures of the American coast from Chile to Alaska. Through the strengthening of surface winds and currents and the reduction in cloud cover, La Nina events directly warm the PWP, SPCZ and Kuroshio extension. Through known changes in atmospheric circulation, ENSO events impact land surface and sea surface temperatures globally. The only other periodic natural forcing/variable that has a larger multiyear impact on global temperature is a catastrophic volcanic eruption. You call those peanuts?

    Regards

  38. C James: You wrote, “And while it is important we get all the technical details correct, I think this issue is just that, a technical one. It is not the most important one.”

    Agreed. This post was simply an attempt to dispel misunderstandings and suppress the chain reaction rumors.

  39. # Bob Tisdale says: July 24, 2010 at 9:09 am
    ___ I believe the European wartime winters have been studied. I recall reading something about them (a paper?)
    ___ The only other periodic natural forcing/variable that has a larger multiyear impact on global temperature is a catastrophic volcanic eruption.

    Presumably you have come across the:
    Letter to NATURE; “Extreme climate of the global troposphere and stratosphere in 1940–42 related to El Nino, by S. Broennimann, J. Luterbacher3, J. Staehelin, T. M. Svendby, G. Hansen & T. Svenøe; Vol 431, 21 October 2004, p. 971f.
    The paper concludes: “ that the global climate anomaly in 1940 to 1942—previously poorly documented—constitutes a key period for our understanding of large-scale climate variability and global El Nino effects”, based on the assumption that there had been a prolonged El Nino effect from 1939-1942. The claim (prolonged El Nino) is mainly based on sea surface temperatures, which are highly unreliable as discussed in a paper from 1997:
    http://www.oceanclimate.de/English/Pacific_SST_1997.pdf
    For example Henry Diaz and G. Kiladis (1992, p.18) conclude that there was an El Nino in 1939 and an El Nina in 1942. Even when one regards a size between Australia and Africa as a reasonable equivalent for the size of the pool of warm water representing an El Nino, the pool by cubic kilometres is small in comparison with the dimension of the ocean in space and depth.

    Concerning the impact of the oceans in case of a serious volcanic eruption, back in 1992 I analysed why the Krakatoa event in August 1883 did not resulted in a devastation decline of global temperatures, although the sun amount of solar energy received was clearly reduced for a period of several years.
    http://www.whatisclimate.com/conditions-for-the-protection-of-the-global-climate.html#_aa12
    Five years after the eruption of Krakatoa, the scientific work on the events of the year 1883 were temporarily brought to a close with the “Report of the Krakatoa-Committee of the Royal Society.” The most amazing aspect of the report is that it does not contain any mention of possible relevance of the oceans. The investigation shows that the most likely reason for the fairly stable air temperatures in the aftermath of Krakatoa had been the heat supply by the oceans.

    With regard to understanding the three extreme WWII winters little to nothing has been done. Since the 19th century, winters had become successively milder. “The present century has been marked by such a widespread tendency towards mild winters that the ‘old-fashioned winters’, of which one had heard so much, seemed to have gone for ever. The sudden arrival at the end of 1939 of what was to be the beginning of a series of cold winters was therefore all the more surprising,” reported the British scientist A. J. Drummond in the QJoR Met. Society as early as 1943. He also notes: “Since comparable records began 1871, the only other three successive winters as snowy as the recent ones (1939/40, 1940/41, 1941/42) were those of the last war, namely 1915/16, 1916/17, 1917/18.”
    Just a coincident? The role of the ocean and seas on the war winters and the global cooling could be well studied along the WWII naval activities in Northern Europe’s waters, and in the Northern Atlantic and Western Pacific since 1942.
    Thanks for your reply and best regards ArndB

  40. At the risk of turning this into a 2 discussion thread, I cannot let Mike’s comment go without rebuttle.
    First and foremost, why not let science smoke out the truth. How does a tax on non-scientists get us closer to this “truth” that you and the rest of us seek?
    Senondly, why does someone else have to go first before you make a sacrifice? Do you really think those that propose a carbon free energy world and those that profit from it will set the example for self sacrifice? Would that person be someone like Al Gore?
    You want leadership from the elite or you threaten confiscation of their wealth. I hope that one day you will see how easy such loigic can turn against you. Taxation used to be a means for funding the constitutional functions of the government. Now we sit back in our chair and say, they don’t live the right way, let’s take their wealth. I cannot make the connection between the two and it concerns me that you can.
    Lastly, you close with the hope that you have not made an unreasonable assumption. I think you have made several unreasonable assumptions. To assume that the “elite’s” lifestyle is the key to preventing the earth from becoming a “venusian hell hole” is naive in the extreme. Just as naive as assuming that public policy should flow before the facts are discerned. It may be done that way and you may be used to that methodology, but I would rather be on more firm ground scientifically before rearranging every facet of interaction between supposedly free people.

  41. ArndB says:
    July 24, 2010 at 6:31 am

    the 3 War Years make a great argument against letting the Green Agenda toy with the climate at scale.
    The millions of gallons of dispersant condoned by EPA in the Gulf with unknown long-term consequences are an example, along with Dept. of Fish & Game poisoning of Lake Davis, Plumas Co, Ca.
    They don’t care what goes wrong.

  42. Bob Tisdale says:
    July 24, 2010 at 3:54 am

    Could you do something on Polar Amplification?
    I’d like to know how that works and what it affects.

  43. “And while it is important we get all the technical details correct, I think this issue is just that, a technical one. It is not the most important one. I think we can trust that GISS, Hadley Centre and NCDC all come up with pretty much the same results. However, I don’t think those results are correct due the homogenization, adjusting, poor siting, UHI problems, etc. that has been done to the raw data. ”

    1. It is a technical detail that Frank got wrong. It points to a couple things. Its very difficult to speak about GISSTEMP with any authority UNLESS you have gone through the code. When you do go through the code you’ll see that Frank is just flat wrong. he’s misinterpreted a graph. Its also clear that there are dangers in posting data without the code. GISS post a “legacy” graph. One that has been misinterpreted. Its important to those of us who lobby for open code and data, that people who make mistakes about freely posted material be candid and forthcoming when they make mistakes.

    2. The results of the three agencies is substantially the same. More importantly their results have been matched by several independent citizens, from many sides of the debate.

    3. No one who asserts there are fundamental issues with the GISS approach has seen fit to invest the few days of coding it would require to “prove” the case. Every attempt I am aware of has confirmed the general soundness of the approach.

    4. On the “correctness” of the answer. No answer is correct. There are two things we need to concern ourselves with. Error and bias. WRT Error the LLN is our friend, provided there is no BIAS. So lets talk about BIAS

    A. Sampling bias. The concern here is that we have a sample that is BIASED toward
    higher TRENDING stations and grid cells. Whats the evidence and whats the
    theory? I’ve resampled the sample of stations. That is, I randomly selected stations and recomputed the mean. no difference in trend. I’m going to resample
    on a cell basis ( pick 3×3 grid cells and resample) I already know the answer.
    the answer will be NO DIFFERENCE IN TREND. I know this because the distribution of trends for the cells is rather normal looking, a bit leptokurtic
    if I had to guess. Theory says that the trends at the Northern latitudes will
    be higher than those at lower latitudes. polar amplification. observation bears
    this out. The highest trending cell is a high latitude cell. That means, missing data
    NORTH of this cell, is likely to be at or above the trend observed at a lower latitudes. Adding more stations, covering more of the globe will not change the global average in any appreciable manner. If anything increased sample will raise
    the trend. I make this estimate based on theory( polar amplification) and observation of know data and its distrubutional characteristics.

    B. siting Bias. The Surface stations project will help us get a better handle on the
    QUANTIFICATION of the bias. From theory we can expect the bias to be in BOTH
    directions. Positive ( too hot) and negative (too cold). The latter fools people somewhat, but shading can and does cool sites. The issue with siting bias is:
    1. Magnitude.
    2. Mean direction (positive or negative)
    the CRN rating was created by Dr. leroy. His associate has done some field studies of the bias. Those preliminary estimates indicated a warm BIAS of .1C. That is,
    for CRN ,2,3,4 the mean BIAS was .1C warmer than a perfect site. CRN5 was not tested. This also confuses people. The bias, a mirco climate issue, is modulated by
    several factors. somedays you see it, other days NOT. The main difference between
    CRN 2,3,4 and CRN 1 is the VARIANCE in the measures. ‘2’ refers to the Spread of the data. with a CRN2 you can see temps as HIGH as 2C hotter or as Low as 2C cooler. The mean, is a +.1C. the SPREAD is 2C. Same for CRN3 and CRN4.
    Leroy, limited study will of course be supplemented by Anthony’s larger study.
    But, based on the limited field study, one should not EXPECT to see a large bias.
    one should expect and plan to find a small positive Bias.

    C. Adjustment and homogenization BIAS. Much of the data that goes into GHCN has
    already been homogenized and adjusted by the source data maintainers. GHCN adds QC on top of this. The source homogenization and adjustments are a BLACK BOX.
    One cannot conclude from this that the adjustments introduce a warm bias OR a cool bias. One cannot suppose either case. I would not expect 200 individual countries to systematically introduce a warm bias or a cool bias. Absent substantial evidence to the contrary, one can rationally assume a zero mean Bias. However, THIS is one issue that nobody has run down to the source data. This is an answerable question.
    Again, there is a difference between
    A. believing adjustments are biased warm. (the disbeliever position)
    B. believing the are Zero mean Biased (the warmist position)
    C. Being Open minded and asking people to look at the question ( lukewarmers)

    D UHI.
    UHI bias, like microsite bias can be both positive and negative. Although on an area basis, warm bias tends to be larger. In one particular study a cool bias on average of 2.6C was noted in areas where there was canopy(>64% canopy cover). sun cover. go figure.

    “Portland metropolitan area can experience a
    negative UHI during summertime afternoons. The
    regions that experience a negative UHI are those
    with greatest canopy coverage as determined by
    the first rule of daytime tree structured regression
    models, this is generally the area encompassing
    Forest Park. Forest Park is the coolest region of
    the city and is generally 2-4 ºC cooler than the
    rural surrounds and can be up to 10º C cooler than
    surrounding urban regions. During the afternoon
    (Figure 6a), the warmest areas of the city are just
    across the Willamette River (to the east) from the
    downtown area, an area characterized by
    commercial land-use and an industrial area just to
    the north of the downtown between Forest Park
    and the Willamette, these regions experience a
    predicted UHI intensity of up to 5 ºC. ”

    In areas where there was no cover you can get a warm bias. This bias is driven by Building size. Smaller buildings lead to a cool bias, larger building to a substantial warm bias.

    The Evidence shows the bias can be warm or Cool, depending on the location. This goes all the way back to Oke’s seminal work on UHI. So again, this observation suggests that we should look closely at the particulars before jumping to conclusions.

  44. jst:

    Greatly appreciate your reply. Your rejoinder was thought-provoking; some of the thoughts it prompted follow:

    We both agree on the proper place for science and taxes in our government. But there are powerful interests, I strongly suspect, that do not agree and have chosen to push a self-serving cap and trade tax and renewable energy policies by means of pseudo-science scare tactics and the considerable propaganda resources at their disposal. Sweet reason and good science are hardly likely to prevail in a contest with such interests, however much we might wish otherwise. Not that we shouldn’t be working on the science, as well.

    You say my logic “can turn on you.” As I see it, these guys already have their hands half-way in my pocket, I’m merely responding in kind.

    My proposal fights fire with fire. In so many words: If I must sacrifice my figurative SUV to save the planet then my betters must also sacrifice their palatial homes, private jet, yachts, etc. And since I’ve got a “little guy” distrust of plutocrats, I’d require that those fine fellows go first.

    My proposal would “smoke” out any hustle attached to AGW advocates and their cap and trade tax and alternative energy agenda items by insisting on “mutually assured destruction” of everyone’s lifestyle. That is, if AGW advocates aren’t willing to share with the helots the draconian life-style sacrifices entailed in their low-carbon brave-new-world, then almost certainly the whole AGW business is an alarmist crock and there isn’t any real peril to dear old Gaia. That’s the method behind the madness of my proposal.

    Not being totally naive, I estimate my fantasy “High Carbon Lifestyle Tax” would go nowhere in Congress for obvious reasons. But rather than satisfying ourselves with an occasional amusing story of Al Gore’s or John Kerry’s (others could be named) hypocrisy, I’d rather expose that hypocrisy (and the probable deceit it reveals) with a concrete legislative proposal. At a minimum, I think a legislative debate of my proposal, even if it leads to a likely rejection, has the best potential to derail climate legislation in its current form–namely, the big shots keep all their toys and even make money off the deal, while us kulaks loose it all.

  45. Thanks Mike. Good stuff here. I like the direct approach. That means getting their hand out of my pocket before trying to insert mine in theirs.
    Exposing hypocrisy requires a science before public policy approach. I think that gets us to where we both want to be. Nobody ever said citizenship was easy.

  46. Wow, this is good stuff!! Next time you plan on having an epic battle like this, might I suggest pay-per-view??

    Could help support the site!

  47. Why is DMI Arctic SAT running at or near freezing this El Nino? As melt accelerates after July, one would think this unusual cold snap will have an effect on the seasonal ice melt in the coming weeks. Why does DMI disagree with GISS so much?

    Also, on Arctic Polar Amplification, per Polyakov:
    Polar amplification of global warming

    Arctic variability is dominated by multi-decadal fluctuations. Incomplete sampling of these fluctuations results in highly variable arctic surface-air temperature trends. Modulated by multi-decadal variability, SAT trends are often amplified relative to northern-hemispheric trends, but over the 125-year record we identify periods when arctic SAT trends were smaller or of opposite sign than northern-hemispheric trends. Arctic and northern-hemispheric air-temperature trends during the 20th century (when multi-decadal variablity had little net effect on computed trends) are similar, and do not support the predicted polar amplification of global warming. The possible moderating role of sea ice cannot be conclusively identified with existing data. Observed long-term trends in arctic air temperature and ice cover are actually smaller than expected, and may be indicative of complex positive and negative feedbacks in the arctic climate system. In summary, if we accept that long-term SAT trends are a reasonable measure of climate change, then we conclude that the data do not support the hypothesized polar amplification of global warming.

    There are other more recent studies on the subject.

    Also, reading the RC content from your link, it is a bit odd that tropical tropospheric warming can be caused from “many sources”, yet in 2005 Gavin Schmidt coauthored a paper where it explicitly states:
    Amplification of Surface Temperature Trends and
    Variability in the Tropical Atmosphere

    Tropospheric warming is a robust feature of climate model simulations driven
    by historical increases in greenhouse gases (1–3). Maximum warming is predicted
    to occur in the middle and upper tropical troposphere.

    It seems no matter what happens, it is always “consistent with” AGW “theory”.

  48. Bob, you say: “You also need to assure that your base years and method of smoothing do not skew your results, and as far as I know you have elected not to do these things. Why? ”

    Bob, i use baseline 1951-80 on all data sets, and in the specific datasets used for this topic all are 5 yr averaged in the same way.
    I dont know why you think otherwise.

    Bob: “The problems arise when one assumes that GISS land surface data represents the same data as CRU and NCDC.”
    I do no such thing. On the contrary, i highlighted to you that these compares are complicated because it seems that GISS is more confident than the others, that coastal stations cover 1200 km radius of ocean. But the others use the same coastal stations diferently it seems. So im not saying that there is a high precision 1:1 relation with these “land” data estimates.

    Bob: “Are you saying that only land surface data inland 1200km from the coast should be evaluated? If so, why?”
    Due to the confusion with GISS claiming that coastal stations are 50% ocean area etc. well yes, if that where correct and you really wanted a LAND ONLY – would you not have to examine the stations more than 1200 km from coasts? This was just a thought, not at all essential to our discussion.

    Bob: “In this post I removed a known difference between GISS and the other two datasets by limiting the data to non-polar latitudes. I then subtracted the scaled SST anomalies from the combined land and sea surface temperature anomalies. The remainders from the three datasets (the contributions of the land surface data) are remarkably similar:
    http://i31.tinypic.com/33cp74w.jpg

    Bob this is your “good stuff”.
    Im NOT saying that you are plain wrong. The truth is, I find this very very complicated. Mostly because you do some assumtions i for now dont know if can be used.

    We attack this problem from to opposite angles.

    Here is a graphic that shows that the ocean area in the GISS “land” series was indeed reduced around 4% from 1980 to 1995:


    The red areas are areas that GISS stations covered in 1980 but no longer cover in 1995. The use of ocean area was reduced from 1980 to 1995.

    So in my tiny brain things doenst add up with you focus on this “fact” that im not considdering ocean coverage bu GISS.

    The ocean area used by GISS was reduced 4% from 1980 to 1995.
    BUT
    GISS over the same period INCREASE their use of the GISS “land” (incl this ocean area) from 40% to 73%.

    How on Earth can you explain the increased GISS “land” use from 40 to 73% with the GISS ocean part, when this part over the sae period according to GISS them selves where reduced 4%?

    There is a grotesk mismatch here, and then you show that if you reduce your focus to 60S to 60N then you make GISS, CRU and NCDC look alike.

    I understand your point, I understand why you look at 60S-60N, yes it is relevant to say that there are these curves then have similarities. But still, as a scietific minded person, I just dont think “we are there” yet. Its NOT bad will.

    In my approach we simply stick to analysing GISS data. According to GISS themselves their “land” temperature stations coverage was redued around 4% 1980-1995. At the same time they increase their use of this “land” component from 40% to 73%.

    Similarities with other datasets 60S – 60N are interesting, but we are still mising a real explanation. And im sorry to Niels A Nielsen and others that think im “irritating” but im just a irritating sceptic :-)

    Im truly aware that i could very well be wrong, but should i then say nothing?

  49. Bob, your approach that you want to compare GISS with other datasets forces you to chop of areas and only look at 60S-60N.

    When i only look at GISS data, A can look EXACLTY at the area GISS covers, no more no less. We are free of assumptions that i feel makes the whole issue cloudy and weaker.
    And doing so i find 3 severe problems mentioned here:

    http://hidethedecline.eu/pages/posts/follow-up-on-giss-rdquolandrdquo-temperatures-187.php

    Perhaps by using your 60S-60N the problem 3 i mention disappears… but if the issue of zero weight directly in the datasets used is still there when simply looking at GISS “land” / GISS total and the relevant SST´s, then the problem is stil there!!

    You cant say: “Ok, we can calculate from the GISS dataseries that they have not used their “land” component at all from 1900 to 1920, but ladies and gentlemen, when i look at 60S-60N I cant see the problem, and then voila, its gone”

    K.R. Frank

  50. One more thing, Bob. My topic was how much the GISS “land” was weighted in the total GISS. Thats the ptopic.
    Then you show that land data trends CRU, NCDC and GISS are similar.

    But I did not attack GISS land data.
    I attacked how the land (yes yes land incl ocean) was weighted by GISS in their total temperature graph.
    K.R. Frank

  51. – in fact I too got the result that these land data from 4 sources are rather similar- GISS in fact the one with least warming:

    But as i wrote: what does it help that GISS land appears less heat correcyed than the others if the GISS combining process in stead induces the heat in their total?

  52. Frank: “One more thing, Bob. My topic was how much the GISS “land” was weighted in the total GISS. Thats the ptopic.
    Then you show that land data trends CRU, NCDC and GISS are similar.
    But I did not attack GISS land data.
    I attacked how the land (yes yes land incl ocean) was weighted by GISS in their total temperature graph.”

    You don’t question that land data trends of CRU, NCDC and GISS are similar?
    But GISS “land” IS dissimilar. Bob has a description of it above. Zeke showed you that the GISS “land” trend differs considerably from a true land trend that does not include ocean and which trends considerably higher – much in line with CRU and NCDC as you can see from Bobs figure 8.

    You are really confused, Frank. In my opinion the odd GISS “land” index is not interesting at all. And you can’t use it the way you do. GISS doesn’t “use” or “weight” it in their combined land and ocean as you keep saying. Drop it.

  53. Frank Lansner: You replied, “Bob, i use baseline 1951-80 on all data sets, and in the specific datasets used for this topic all are 5 yr averaged in the same way.
    I dont know why you think otherwise.”

    I showed you why I think otherwise. Refer to:
    https://wattsupwiththat.com/2010/07/17/tipping-point-at-giss-land-and-sea-out-of-balance/#comment-435038

    And again, what type of smoothing are you using? Is it centered on the 1st, 30th, or 60th month? Why does it extend from 1880 to 2009?

    You replied, “I do no such thing. On the contrary, i highlighted to you that these compares are complicated because it seems that GISS is more confident than the others, that coastal stations cover 1200 km radius of ocean. But the others use the same coastal stations diferently it seems. So im not saying that there is a high precision 1:1 relation with these “land” data estimates.”

    The reason the GISS land data extends out over the oceans is because that dataset was developed for use WITHOUT SST data. When GISS introduces SST data to create a combined dataset, the land surface data does not extend out over the oceans by 1200km.

    You replied, “Due to the confusion with GISS claiming that coastal stations are 50% ocean area etc. well yes, if that where correct and you really wanted a LAND ONLY – would you not have to examine the stations more than 1200 km from coasts?”

    No. If you wanted to research the GISS land data that is used in its combined product, you would examine the data only where it appears on land, and in order to do this you would not examine it where it extends out over the oceans.

    And please provide a link to a GISS document “with GISS claiming that coastal stations are 50% ocean area etc.”

    You wrote, “Here is a graphic that shows that the ocean area in the GISS “land” series was indeed reduced around 4% from 1980 to 1995:
    http://hidethedecline.eu/media/GISSlandproblems/fig8.jpg”

    That is the wrong map to use. GISS does not use all of that data when it creates the combined land and sea surface data with 1200km radius smoothing.

    The following map shows (approximately) the areas GISS uses in its combined land plus sea surface temperature product with 1200km radius smoothing. The ocean data then fills in the rest. The exceptions to this are two small areas where there is no land or ocean data in the Arctic and Southern Oceans/Antarctica.

    You wrote, “The ocean area used by GISS was reduced 4% from 1980 to 1995…”

    Wrong. The amount on SST data GISS used in 1980 is the same as in 1995. Here’s a gif animation of the GISS (Hadley/Reynolds) SST anomaly maps for those two years. The same areas of the oceans are used.

    You wrote, “…GISS over the same period [1980 to 1995] INCREASE their use of the GISS “land” (incl this ocean area) from 40% to 73%.”

    Wrong. GISS used less land surface data in 1995 than it did in 1980.

    You wrote, “How on Earth can you explain the increased GISS ‘land’ use from 40 to 73% with the GISS ocean part, when this part over the sae period according to GISS them selves where reduced 4%?”

    And you wrote, “In my approach we simply stick to analysing GISS data. According to GISS themselves their ‘land’ temperature stations coverage was redued around 4% 1980-1995. At the same time they increase their use of this “land” component from 40% to 73%.”

    GISS did not increase “GISS ‘land’ use from 40 to 73%”. You are using the wrong GISS land dataset in your analysis.

  54. Bob wrote: “The divergence of land surface temperatures from sea surface temperatures is not only impacted by UHI. Land surface temperatures are also affected by land use changes, polar amplification, thermal inertia, etc. Does Frank account for these in his analysis?”

    I believe the main point to be that the land temperature records used for gloval trmperature calculations, and by GISS extrapolated far out in the oceans, despite the existence of SST data, are contaminated by a number of local factors such as those you mentioned, including also bad siting, measurement errors, reporting errors etc.
    How much faith should we place in the accuracy of the result derived from such dubious data and seemingly dubious calculation and extrapolation methods? The claimed allegedly “unprecedented” warming is hardly outside the error bars of the instruments used to gather the data. Just asking…

  55. I don’t think were anywhere close to building a modern Tower of Babel, but if and when that day comes, I just know something’s going to confuse our tongues and scatter us all to the four winds. Again!

  56. Frank Lansner wrote: “But I did not attack GISS land data.
    I attacked how the land (yes yes land incl ocean) was weighted by GISS in their total temperature graph.”

    But GISS does not use the “lan incl ocean” data to create its combined dataset. And this is why your analysis is wrong.

  57. Hi Bob. Pascvaks is correct, this is a stunning example of Babylonic problems…

    Anyway, you write:
    “I showed you why I think otherwise. Refer to:
    https://wattsupwiththat.com/2010/07/17/tipping-point-at-giss-land-and-sea-out-of-balance/#comment-435038

    And again, what type of smoothing are you using? Is it centered on the 1st, 30th, or 60th month? Why does it extend from 1880 to 2009?

    Yur link is not to a graph but to another comment which holds this graph:

    But data here goes from 1900 to 2007, i promise :-) The data foundation is from 1898 to 2009, so 5 yr avg for first point year 1900 is years 1898 to 1902 etc.

    K.R. Frank

  58. Bob, you wrote: “But GISS does not use the “lan incl ocean” data to create its combined dataset.”

    I used data source:

    1) http://data.giss.nasa.gov/gistemp/tabledata/GLB.Ts.txt
    “GLOBAL Temperature Anomalies in 0.01 degrees Celsius base period: 1951-1980
    sources: GHCN 1880-06/2010 (meteorological stations only)”

    2) http://data.giss.nasa.gov/gistemp/tabledata/GLB.Ts+dSST.txt
    “GLOBAL Land-Ocean Temperature Index in 0.01 degrees Celsius base period: 1951-1980
    sources: GHCN 1880-06/2010 + SST: 1880-11/1981 HadISST1 12/1981-06/2010 Reynolds v2”

    The issue is that 2) temperatures relatively comes closer and closer to 1) – and thus longer and longer away from SST.

    It doenst matter what GISS do when making 2) – as I see it – 2) ends up relatively more and more like 1). Thats the problem.
    So you must use still less SST compared to GISS 1) to get a result like 2).

    This is a problem, Bob :-)
    If you suddenly say they dont use SST, they dont use this or that, it doesnst matter. If you have to use more and more 1) in combination with SST to get the total 2) USED BY IPCC etc. we certainly have a problem! You dont get a sign of a data problem much better :-)

  59. Georgegr: You asked, “How much faith should we place in the accuracy of the result derived from such dubious data and seemingly dubious calculation and extrapolation methods? The claimed allegedly “unprecedented” warming is hardly outside the error bars of the instruments used to gather the data. Just asking…”

    Unfortunately, or fortunately, that’s not the subject of this post. Frank Lansner has written a post…
    https://wattsupwiththat.com/2010/07/17/tipping-point-at-giss-land-and-sea-out-of-balance/
    …and continues to believe, incorrectly, that GISS changes the ratio of Land to Sea Surface data over the 20th century. That’s the topic of this post and the ongoing discussion on this thread.

    Regards

  60. If GISS uses relativly less and less SST and their overall temperature product, this wil certainly give a resulting temperature curve with a warm errortrend simply originating from smaller and saller SST use after 1980!

    Obviously the Hansenizer organistaion has PILES of “methods” and blah blah, but you have to see the bug picture in all this:
    1) The SST has just a slow slow warming trend.
    2) GISS use SST less and less after 1980.
    3) GISS does NOT have a coverage of ocean growing strongly after 1980 to make excuse for this!

    Details are good, but dont miss the big picture :-)
    K.R. Frank

  61. What ever SST fraction is “correct” to use, you have to use the same fraction in 1980 as in 1995 if you want a usefeul temperature series?

    K.R. Frank

  62. if it suddenly takes much less SST and much more 1) to get the total GISS 2) after 1980 and this cannot be accounted for by strongly more ocean coverage in 1) – then there is a problem.

    Thanks for the debate :-) I will spare the moderators further coments for now :-)

    K.R. Frank

  63. Frank Lansner: You wrote. “Bob, its true that i cant use 1) to say exactly how much land or sea GISS has.
    But its just not the point here. Argh !! :-)”

    What is the point then? You are using that dataset to determine the ratio of land to ocean, and you cannot do that.

  64. Frank Lansner says: “if it suddenly takes much less SST and much more 1) to get the total GISS 2) after 1980 and this cannot be accounted for by strongly more ocean coverage in 1) – then there is a problem.”

    And again, the problems are you can’t use anomalies and you can’t use that land surface dataset the way you are using it.

    You can’t use anomalies because the percentages change depending on the base years YOU elect to use. And you can’t use that land surface dataset because it represents more than the continental land mass.

  65. Frank, pretend this GISS “land” dataset doesn’t exist. It is irrelevant. It is not “used” in any way by GISS to calculate their combined land-ocean index.

    Your patience is admirable, Bob.

  66. I say that GISS LST can be used, NOT to conclude something specific about the actual land area, but to conclude on the essential point of all these writings.

    Please then tell me where you think im wrong in the following logic:

    1) The GISS LST is the temperature for the area (both land and some ocean) covered as illustrated here in 1980 and in 1995:

    2) GISS fills in the remaining ocean area not covered by GISS LST with SST data from HADISST/Reynoldsv2.

    3) The GISS LST area has become around 4% smaller from 1980 to 1995, red areas shows where GISS LST area was bigger in 1980 than 1995:

    4) Therefore we should expect that GISS would need to fill in these missing ocean areas 1995 with more SST use from HADISST/ReynoldsV2

    5) Therefore, the SST fraction used in 1995 was likely to be slightly bigger than in 1980. Likewise, the GISS LST fraction used in 1995 is likely to be slightly smaller than in 1980 ( -since the GISS LST area was reduced)

    6) Fractions GISS LST and SST (HADISST/reynoldsV2) it takes to produce the combined GISS global:

    year/ GISS LST / SST (HADISST/ReynoldsV2
    1980/ 40% / 60%
    1995/ 73% / 27%

    7) So while the actual GISS LST got smaller 1980 – 1995, the fraction GISS LST it takes to produce GISS global increases massively from 40 to 73%.

    How can this be?

    8) Another problem: Since SST and Air temperatures are 2 significantly different sizes, it is wrong for GISS to use near 100% SST from 1900 to 1920, and then the much smaller fractions of SST in recent years.

    9) SST and Air temperatures has differend warming trends and therefore, the gradual method shift from 1900 to 2007 it self induces a temperature trend.

    10) If GISS clobal temperature graph should be useful, they should use same weight for all years (even though more GISS LST is available in some periods)

    Yes, i might be wrong, But then pinpoint exactly where it is.

    K.R. Frank

  67. ***
    Niels A Nielsen says:
    July 25, 2010 at 1:59 pm
    Frank, pretend this GISS “land” dataset doesn’t exist. It is irrelevant. It is not “used” in any way by GISS to calculate their combined land-ocean index.
    ***
    Niels,
    1) take a good look at the links:
    GISS global Land + SST
    http://data.giss.nasa.gov/gistemp/tabledata/GLB.Ts+dSST.txt

    GISS land temp
    http://data.giss.nasa.gov/gistemp/tabledata/GLB.Ts.txt

    You want me to pretend that the GLB.Ts is NOT used in the GLB.Ts+dSST ?????
    Where does this idea come from?

    K.R. Frank

  68. Niels, here is a tool where you combine the HADISST/reynoldsV2 SST in the areas NOT filled out with the GHCN land temperature station based GISS land fraction to get full values (and maps) globally:
    http://data.giss.nasa.gov/gistemp/maps/

    Thats how it works, you combine the GHCN temp station area data with the remaining SST area. And thats why its so odd that the SST use just explodes from 1980 – 1995 while the GISS land coverage actually is reduced 4% 1980 – 1995.

    So what temperature series is better to represent the GISS land than the updated
    http://data.giss.nasa.gov/gistemp/tabledata/GLB.Ts.txt
    ?
    Do you think that an alternative version of this will change anything? Tell me what dataset you believe is to be used then, and i will test it. Honestly i would be a little surpriced if an alternative GISS land component would change anything.

    K.R. Frank

  69. Correction, it shold say:
    And thats why its so odd that the GISS land component used just explodes from 1980 – 1995 (from 40 to73%) while the actual GISS land coverage actually is reduced 4% 1980 – 1995.

    Niels: Can you give ANY good reason for this? Just move on, nothing to see?
    – I think its a sceptic vertue to explore such curiosities.
    K.R. Frank

  70. Grank Lansner wrote: “The GISS LST is the temperature for the area (both land and some ocean) covered as illustrated here in 1980 and in 1995:
    http://hidethedecline.eu/media/GISSglobal/fig7.jpg

    This dataset…
    http://data.giss.nasa.gov/gistemp/tabledata/GLB.Ts.txt
    ..is NOT the dataset that GISS uses in its combined kand and sea surface temperature anomaly data:
    http://data.giss.nasa.gov/gistemp/tabledata/GLB.Ts+dSST.txt

    Therefore your analysis is wrong!

  71. Frank Lansner: Sorry about spelling your name wrong above.

    You replied to Niels A Nielsen: You want me to pretend that the GLB.Ts is NOT used in the GLB.Ts+dSST ?????
    Where does this idea come from? ”

    Because it is not the data used in the GLB.Ts+dSST data. That is what everyone has been trying to telling you. Here’s a link to my first post:
    http://bobtisdale.blogspot.com/2010/07/notes-on-gistemp-ratio-of-land-to-sea.html

    In it I wrote, “So in order to properly create a weighted average of GISTEMP land and sea surface temperature data with 1200km radius smoothing, the land surface data where it extends out over the oceans would first need to be masked.”

    And that means that YOU, Frank Lansner, would have to take the GISS land surface dataset that has a global coverage that looks like this…

    …and convert it to one that looks like this:

    After you have done that, then you have other problems to contend with before you can perform your analysis.

  72. Frank Lansner:You wrote, “Thats how it works, you combine the GHCN temp station area data with the remaining SST area.”

    Wrong. That is not how GISS creates its combined land plus sea surface temperature data,

  73. Frank Lansner: To confirm what I wrote to you above…

    ” Wrong. That is not how GISS creates its combined land plus sea surface temperature data”…

    refer to page 1 and 2 of Hansen et al (1996):
    http://pubs.giss.nasa.gov/docs/1996/1996_Hansen_etal_1.pdf

    In it they describe the early SST dataset they used, “Smith et al. (1996) developed an interpolation method using spatial patterns from empirical orthogonal functions (EOFs) to improve SST analyses for 1950-81, when satellite data were not available. The method produces spatial EOFs from OI analyses for 1982-1993. The dominant EOF modes are used as basis functions and fit, in least squares sense, to the in situ data to determine the time dependence of each mode. The SST field reconstructed from these spatial and temporal modes is confined to 59°N-45°S because of limited in situ data at higher latitudes.”

    That was an early version of the Reynolds SST data. The current version covers the oceans globally.

    Then Hansen et al (1996) describe the method they use to add the SST data. They write, “We have combined EOF SSTs for 1950-1981 and OI for 1982-1995 with meteorological station surface air temperature data to produce a surface temperature index for 1950-1995. The analysis method, as in HL87, divides the world into 8000 equal-area boxes. A box over land receives diminishing influence from stations up to 1200 km distant. A box over ocean between 59°N and 45°S uses the local SST, because of the complete coverage. A coastal box uses a meteorological station if one is located within 100 km of the box center.”

    Refer also to the post at Lucia’s:
    http://rankexploits.com/musings/2010/the-gistemp-land-fraction/

    Specifically, refer to Nick Barnes (Comment#49280) July 19th, 2010 at 10:24 am. In it he describes how GISS merges the land and sea surface data. He writes, “For each grid cell, if there are fewer than 240 valid monthly data points in the ocean series, or if the nearest land station is less than 100km from the centre of the cell, then the land series is used. Otherwise the ocean series is used for that grid cell.”

    And since both SST datasets are complete on a monthly basis, the portion of his description, “…if there are fewer than 240 valid monthly data points in the ocean series…” would no longer be a factor.

  74. Bob, you write:
    “This dataset…
    http://data.giss.nasa.gov/gistemp/tabledata/GLB.Ts.txt
    ..is NOT the dataset that GISS uses in its combined kand and sea surface temperature anomaly data:
    http://data.giss.nasa.gov/gistemp/tabledata/GLB.Ts+dSST.txt

    But this sounds odd…
    Assuming you are correct (?) And we cant use http://data.giss.nasa.gov/gistemp/tabledata/GLB.Ts.txt
    – GISS has not made available the dataset they then use? Why?
    And they only use parts of their http://data.giss.nasa.gov/gistemp/tabledata/GLB.Ts.txt data in the combined? Which parts?

    You dont have to answer all this, of course, it just sounds strange, and I will dig into this.

    K.R. Frank, and yes as niels says: you are a patient man, thankyou. I just have to know whats behind rather than just “be told” :-)

  75. Frank

    “But this sounds odd…
    Assuming you are correct (?) And we cant use http://data.giss.nasa.gov/gistemp/tabledata/GLB.Ts.txt
    – GISS has not made available the dataset they then use? Why?
    And they only use parts of their http://data.giss.nasa.gov/gistemp/tabledata/GLB.Ts.txt data in the combined? Which parts?

    You dont have to answer all this, of course, it just sounds strange, and I will dig into this.”

    get the code and read how it works. Its about 10K lines. doesnt take very long to figure it out. sheesh.

  76. Frank Lansner: Sorry, my July 26, 2010 at 12:33 am reply above was wrong. It should have read, Your assumed problem 3 also appears to be based on your use of the wrong land surface dataset.

    With respect to “problem 2” in your post, the marine air temperature data is different from the other datasets in a number of ways. Marine air temperature datasets are typically known as NMAT and the “N” stands for night or nighttime, because the NMAT datasets avoid “solar” biases by using only readings taken at night. On the other hand, SST and the island station and “ship station” used by GISS include daytime and nighttime readings.

    The coverage of NMAT datasets are also very sparse. There is no data in the Northern Hemisphere north of 60N, and very little data in the Southern Hemisphere:

    So to compare the NMAT and SST datasets correctly, you would have to match the latitudes of the SST data so that they are the same as the NMAT data.

  77. I know what to do now :-) – I will post an update on the subject within 2-3days.
    Thanks for valuable input all, its highly appreciated.

    K.R. Frank

  78. Bob i would very much if you could anser my quetion above:
    ***
    Is not
    http://data.giss.nasa.gov/gistemp/tabledata/GLB.Ts.txt

    the temperatures for the entire area covered by the GHCN stations used bhy GISS, land + ocean as illustrated here:

    ***
    I think yes, but if NO: What is then this
    http://data.giss.nasa.gov/gistemp/tabledata/GLB.Ts.txt
    ?
    And Bob, I think you are overlooking a severe problem mentioned in my “problem 2”, that has nothing to do with missing masking from my part or the like, but let it be, i will explain illustrated soon, i wont take more of yours or the moderators time.

    K.R. Frank

  79. Frank Lansner: You asked, “Is not
    http://data.giss.nasa.gov/gistemp/tabledata/GLB.Ts.txt
    the temperatures for the entire area covered by the GHCN stations used bhy GISS, land + ocean as illustrated here:
    http://hidethedecline.eu/media/GISSglobal/fig7.jpg”

    GISS uses only parts of it. The GISS land+ocean data uses SST over the oceans and land surface data over land. They will extend land surface data over the oceans by 100km if a surface station is along the coast.

    But, of course, the Arctic and the Southern Ocean around Antarctica are different. In these oceans, SST data is deleted where it is permanent sea ice or seasonal sea ice. GISS then extends land surface data out over the oceans. Refer to:
    http://bobtisdale.blogspot.com/2010/05/giss-deletes-arctic-and-southern-ocean.html

    You asked, “What is then this
    http://data.giss.nasa.gov/gistemp/tabledata/GLB.Ts.txt?”

    The “GISS GLB.Ts” dataset is an updated version of the data used in Hansen and Lebedeff (1987):
    http://pubs.giss.nasa.gov/docs/1987/1987_Hansen_Lebedeff.pdf

    You wrote, “And Bob, I think you are overlooking a severe problem mentioned in my ‘problem 2’, that has nothing to do with missing masking from my part or the like…”

    I discussed your “problem 2” in my July 26, 2010 at 1:31 am reply.

    Regards

  80. Frank Lansner: Here is something to consider: GISS released the source code for GISTEMP in September 2007. Refer to:
    http://climateaudit.org/2007/09/08/hansen-frees-the-code/

    Steve McIntyre and many other persons who understand programming have thoroughly investigated the methods GISS uses since then. If there had been an unusual change in the ratio of land to sea surface temperatures in GISTEMP, I believe it would have been noticed already.

    While I don’t believe you should stop investigating data, always keep in mind that there are other people who have looked at it before you. And if they haven’t found what you believe to be an error, maybe it’s not an error.

  81. Bob this is really really fine information, thankyou very much.
    As I said we are working on a follow up where the GISS temperatures are tested from another angle, and all you write will be taken into account.

    But I have to say some of the info i receive from here and there is not perfectly matching :-) , So perhaps Joanne nova or me will ask you a few questions in the coming time, perhaps on mail.

    K.R. Frank

  82. tallbloke: You wrote, “I’m interested in the 1910 and 1940 down and up spikes. These coincide with generally negative and positive phases of the PDO, and I was wondering how much they are a feature of all basin temps of that era.”

    Sounds like an idea for a future post. As noted in the post above, the severity of the dip from 1880 to 1910 depends on the SST dataset. It’s strong in the ERSST.v3b and HADSST2 and weak in the HADISST.

    You have to keep in mind how sparse the data is back then. A while back I investigated the period of 1870 to 1940 by comparing ERSST.v3b, HADISST, and HADSST2 to the source ICOADS data. I limited my investigations to shipping lanes in the individual ocean basins since that’s where the majority of the data was located. The differences between the ICOADS data and the other datasets varied between ocean basins for single dataset and they varied from one dataset to the next. There were no constants. For example, I could not find what anyone might consider a constant “Folland adjustment”, even in the HADSST2, which receives the least amount of tinkering. And since I could not identify or explain what was going on, I shelved the post. My conclusion was, the data is so sparse in the early 20th century and the ERSST.v3b and HADISST depend so strongly on the infilling using optimum interpolation, EOF, and EOT (stands for empirical orthogonal teleconnections and it’s used in ERSST) that I was wasting my time trying to identify and explain the differences.

    In other words, was there a major dip and rebound from 1880 to 1940? Maybe.

  83. Frank Lansner: You wrote, “So perhaps Joanne nova or me will ask you a few questions in the coming time, perhaps on mail.”

    Or leave me a link to your new post.

    Regards

  84. Hi Bob,

    Yes, how much of a dip and rebound is the question. The characteristic of the SST’s relationship to the solar cycles on your 37 month smoothed plots is revealing. During the modern period, from 1960 the cycles are clearly visible, but much less so in the 1880-1940 period. No doubt a lot of that is down to poor data collation and calibration across datasets, but if my thinking is right, then some of it (1880-1900) is due to biggish el nino’s firing off as the sea lost heat while solar activity declined. Some of it is due to the sst’s not following the sun closely during the very low solar cycles (1900-1920) and then the steep climb to 1940 would have to be down to low cloud albedo in the roaring 20’s and the depression 30’s. I wonder how best to mine 20’s/30’s weather reports for clues without having to sit in library stacks….

  85. Here’s one clue:
    Warming Arctic Climate Melting Glaciers Faster, Raising Ocean Level, Scientist Says – “A mysterious warming of the climate is slowly manifesting itself in the Arctic, engendering a “serious international problem,” Dr. Hans Ahlmann, noted Swedish geophysicist, said today. – New York Times, May 30, 1937

  86. Bob you wrote:
    “While I don’t believe you should stop investigating data, always keep in mind that there are other people who have looked at it before you. And if they haven’t found what you believe to be an error, maybe it’s not an error.”

    I agree, but one have to balance things, and its a little tough. Before this writing i have been rather wrong in one or two times – out of many many similar writings.

    In this case, im pretty sure that even though on the paper i did had a mistake, the bigger picture is hardly changed much due to this. But sadly not many appears aware.

    You see: If the trend i used for land should be significantly wrong, then the coastal temperature stations all over the world should have a significant (in fact opposite) trend than the land data all over the world!

    How likely is it, that so many stations spread out drop wise over the whole Earth should have a rather consistent trend that is near opposite of all the the other stations also spread out over the world?

    It is just not likely and thus there is reasonable indication for now that to some degree (perhaps not as much as I first suggested) GISS is indeed weighting their “Land” data more specifically in recent years. My furhter analysis supports this – not quite as huge differences as first suggested, but still.

    Obviously I was not sure of any of this at all to begin with due to the reasons you mention. Thats why Joanne and I first send out to several peoble to look at it. And then we posted at Joannes site ASKING if anyone could explain this. And Bob, then came one after the other FAULTY pinpointing of errors in the comment.

    First after this, I published WUWT in a more confident style – and PANG, first then came a lot of heavy critics from the Blackboard and you.

    K.R. Frank

Comments are closed.