Version 6.0 of the UAH Temperature Dataset Released: New LT Trend = +0.11 C/decade

April 28th, 2015 by Roy W. Spencer, Ph. D.

by Roy W. Spencer, John R. Christy, and William D. Braswell

(a PDF version of this post is available here. Monthly updates will use Version 6 starting with the April update.)


Version 6 of the UAH MSU/AMSU global satellite temperature dataset is by far the most extensive revision of the procedures and computer code we have ever produced in over 25 years of global temperature monitoring. The two most significant changes from an end-user perspective are (1) a decrease in the global-average lower tropospheric (LT) temperature trend from +0.140 C/decade to +0.114 C/decade (Dec. ’78 through Mar. ’15); and (2) the geographic distribution of the LT trends, including higher spatial resolution. We describe the major changes in processing strategy, including a new method for monthly gridpoint averaging; a new multi-channel (rather than multi-angle) method for computing the lower tropospheric (LT) temperature product; and a new empirical method for diurnal drift correction. We also show results for the mid-troposphere (“MT”, from MSU2/AMSU5), tropopause (“TP”, from MSU3/AMSU7), and lower stratosphere (“LS”, from MSU4/AMSU9). The 0.026 C/decade reduction in the global LT trend is due to lesser sensitivity of the new LT to land surface skin temperature (est. 0.010 C/decade), with the remainder of the reduction (0.016 C/decade) due to the new diurnal drift adjustment, the more robust method of LT calculation, and other changes in processing procedures.

1. Introduction & Some Results

After three years of work, we have (hopefully) finished our Version 6.0 reanalysis of the global MSU/AMSU data. Many procedures have been modified or entirely reworked, and most of the software has been rewritten from scratch. (Please, before you ask a question, read the following to see if your question has already been answered.)

The MSU and AMSU instruments measure the thermal microwave emission from atmospheric oxygen in the 50-60 GHz oxygen absorption complex, and the resulting calibrated brightness temperatures (Tb) are nearly equivalent to thermometric temperature, specifically a vertically-weighted average of atmospheric temperature with the vertical weighting represented by “weighting functions”.

One might ask, Why do the satellite data have to be adjusted at all? If we had satellite instruments that (1) had rock-stable calibration, (2) lasted for many decades without any channel failures, and (3) were carried on satellites whose orbits did not change over time, then the satellite data could be processed without adjustment. But none of these things are true. Since 1979 we have had 15 satellites that lasted various lengths of time, having slightly different calibration (requiring intercalibration between satellites), some of which drifted in their calibration, slightly different channel frequencies (and thus weighting functions), and generally on satellite platforms whose orbits drift and thus observe at somewhat different local times of day in different years. All data adjustments required to correct for these changes involve decisions regarding methodology, and different methodologies will lead to somewhat different results. This is the unavoidable situation when dealing with less than perfect data.

After 25 years of producing the UAH datasets, the reasons for reprocessing are many. For example, years ago we could use certain AMSU-carrying satellites which minimized the effect of diurnal drift, which we did not explicitly correct for. That is no longer possible, and an explicit correction for diurnal drift is now necessary. The correction for diurnal drift is difficult to do well, and we have been committed to it being empirically–based, partly to provide an alternative to the RSS satellite dataset which uses a climate model for the diurnal drift adjustment.

The following plot (Fig. 1) shows the variety of satellites making up the satellite temperature record and their local solar time of observation as the satellites pass northbound across the Equator (ascending node).

Fig. 1.  Local ascending node times for all satellites in our archive carrying MSU or AMSU temperature monitoring instruments.  We do not use NOAA-17, Metop (failed AMSU7), NOAA-16 (excessive calibration drifts), NOAA-14 after July, 2001 (excessive calibration drift), or NOAA-9 after Feb. 1987 (failed MSU2).

Fig. 1. Local ascending node times for all satellites in our archive carrying MSU or AMSU temperature monitoring instruments. We do not use NOAA-17, Metop (failed AMSU7), NOAA-16 (excessive calibration drifts), NOAA-14 after July, 2001 (excessive calibration drift), or NOAA-9 after Feb. 1987 (failed MSU2).

Also, while the traditional methodology for the calculation of the lower tropospheric temperature product (LT) has been sufficient for global and hemispheric average calculation, it is not well suited to gridpoint trend calculations in an era when regional — rather than just global — climate change is becoming of more interest. We have devised a new method for computing LT involving a multi-channel retrieval, rather than a multi-angle retrieval.

The MSU instrument scan geometry in Fig. 2 illustrates how the old LT calculation required data from different scan positions, each of which has a different weighting function (see Fig. 2 inset). Thus, only one LT “retrieval” was possible from a scan line of data. The new method uses multiple channels to allow computation of LT from a single geographic location.

Fig. 2. MSU scan geometry, MSU2 weighting functions at different footprint positions and the basis for the old LT and new LT computation.

Fig. 2. MSU scan geometry, MSU2 weighting functions at different footprint positions and the basis for the old LT and new LT computation.

The LT retrieval must be done in a harmonious way with the diurnal drift adjustment, necessitating a new way of sampling and averaging the satellite data. To meet that need, we have developed a new method for computing monthly gridpoint averages from the satellite data which involves computing averages of all view angles separately as a pre-processing step. Then, quadratic functions are statistically fit to these averages as a function of Earth-incidence angle, and all further processing is based upon the functional fits rather than the raw angle-dependent averages.

Finally, much of the previous software has been a hodgepodge of code snippets written by different scientists, run in stepwise fashion during every monthly update, some of it over 25 years old, and we wanted a single programmer to write a unified, streamlined code (approx. 9,000 lines of FORTRAN) that could be run in one execution if possible.

Before addressing details of how the new Version 6 processing is different from the old (Version 5.6) processing, let’s examine some results. First let’s look at time series (Fig. 3) of the global average lower tropospheric temperature (LT), and how it compares to the old (Version 5.6) LT:

Fig. 3. Monthly global-average temperature anomalies for the lower troposphere from Jan. 1979 through March, 2015 for both the old and new versions of LT (top), and their difference (bottom).

Fig. 3. Monthly global-average temperature anomalies for the lower troposphere from Jan. 1979 through March, 2015 for both the old and new versions of LT (top), and their difference (bottom).

Note that in the early part of the record, Version 6 has somewhat faster warming than in Version 5.6, but then the latter part of the record has reduced (or even eliminated) warming, producing results closer to the behavior of the RSS satellite dataset. This is partly due to our new diurnal drift adjustment, especially for the NOAA-15 satellite. Even though our approach to that adjustment (described later) is empirical, it is interesting to see that it gives similar results to the RSS approach, which is based upon climate model calculations of the diurnal cycle in temperature.

The next plot we will examine (Fig. 4) is the gridpoint LT trends during 1979-2015. Version 6 has inherently higher spatial resolution than the Version 5 product, which had strong spatial smoothing as part of the data processing and through the nature of how LT was calculated:

Fig. 4. New LT gridpoint temperature trends, Dec. 1978 through March 2015.

Fig. 4. New LT gridpoint temperature trends, Dec. 1978 through March 2015.

The gridpoint trend map above shows how the land areas, in general, have warmed faster than the ocean areas. We obtain land and ocean trends of +0.19 and +0.08 C/decade, respectively. These are weaker than thermometer-based warming trends, e.g. +0.26 for land (from CRUTem4, 1979-2014) and +0.12 C/decade for ocean (from HadSST3, 1979-2014).

The gridpoint trends for LT in Fig. 4 are very difficult to measure accurately over land, primarily due to (1) the diurnal drift effect, which can be at least as large as any real temperature trends, and (2) how LT is computed, which in the old LT methodology required data from different view angles, and thus different geographic locations which can be from different air masses and over different surfaces (land and ocean).

As a result, users can expect that there will be differences between old and new LT trends on a regional basis. Differences are also attributable to our use of a new, more accurate land mask in Version 6. For example, going from Version 5.6 to 6.0 the Australia trend increased from +0.17 to +0.24 C/decade, but the USA48 trend decreased from +0.23 to +0.17 C/decade. The Arctic region changed from +0.43 to +0.23 C/decade. Note that trends are noisy over Greenland, Antarctica, and the Tibetan Plateau, likely due to greater sensitivity of the satellite measurements to surface emission and thus to emissivity changes over high altitude terrain; trends in these high-altitude areas are much less reliable than in other areas. Future changes, probably minor, can be expected as we refine the gridpoint diurnal drift adjustments and other aspects of our new processing strategy.

Fig. 5 illustrates the changes from v5.6 to v6.0 for a variety of regions of interest:

Fig. 5. Regional lower tropospheric (LT) temperature trends in Versions 6.0 and 5.6. “L” and “O” represent land and ocean, respectively.

Fig. 5. Regional lower tropospheric (LT) temperature trends in Versions 6.0 and 5.6. “L” and “O” represent land and ocean, respectively.

Notice the trends decreased the most over the Northern Hemisphere extratropics, especially the Arctic, while tropical warming trends increased somewhat, especially over land. Near-zero trends exist in the region around Antarctica.

We want to emphasize that the land vs. ocean trends are very sensitive to how the difference in atmospheric weighting function height is handled between MSU channel 2 early in the record, and AMSU channel 5 later in the record (starting August, 1998). In brief, the lower in altitude the weighting function senses, the greater the brightness temperature difference between land and ocean, mostly because land microwave emissivity is approximately 0.90-0.95, while the ocean emissivity is only about 0.50. As a result, if the AMSU channel 5 view angle chosen to match MSU channel 2 is too low in altitude, the net effect after satellite intercalibration will be a spurious warming of land areas and spurious cooling of ocean areas (at least when intercalibration is performed with land and ocean data combined). We were careful to match the MSU and AMSU weighting function altitudes based upon radiative transfer theory, and are reasonably confident that the remaining land-vs-ocean effects in the above map are real, that is, the land areas have warmed faster than the ocean regions. This is consistent with thermometer datasets of surface temperature, although our warming trends are weaker. Given the importance of the microwave oxygen absorption theory to the land-versus-ocean trends, we hope to update that portion of our processing for a future version update.

2. Major Changes in Processing Procedures with Version 6

The following is meant to provide a general introduction to the new processing steps in Version 6, emphasizing departures from past practices, and not to provide exhaustive detail. It will likely be close to two years before a peer reviewed paper with greater detail gets published in a scientific journal.

2.1 LT Calculation
We have fundamentally changed the calculation of the lower tropospheric temperature product, LT, from a multi-angle method to a multi-channel method. The main reason we changed methods for LT calculation is the old view angle method had unacceptably large errors at the gridpoint level. While the errors cancel for global averages on a monthly time scale, on a regional or gridpoint basis they can be large. The errors arise because the different view angles necessary to calculate a single LT “retrieval” sample different geographic locations, for instance radiometrically colder ocean and warmer land (see Fig. 2, above).

This would not present as big a problem if the data from the different regions were simply averaged together, but instead they are differenced. The problem is further magnified (literally) because the old LT required a weighted difference between view angles (and thus regions) with large weights (+4, -3 for the MSUs), which amplified any regional Tb differences. Compounded with the need to do diurnal drift adjustments, which can vary substantially from land to ocean, the problems with the old LT were deemed to be too large to continue the old LT calculation methodology.

So, instead of the past method of calculating LT as a weighted difference between different view angles of MSU2 (or AMSU5), we are now calculating it as a weighted difference between MSU channels 2, 3, and 4 (or AMSU channels 5, 7, and 9) at a constant Earth incidence angle. This has the very important advantage that all satellite data necessary for the LT retrieval come from the same location.

This required a correction for calibration drifts in MSU channel 3, especially during 1980-81, which was the original stated reason why a multi-channel retrieval method was not implemented over 20 years ago. That correction is made based upon regression of global monthly anomalies of MSU3/AMSU7 data against MSU2/AMSU5 and MSU4/AMSU9 during 1982 through 1993 (a 12 year period exhibiting two large volcanic eruptions with differential responses in the different altitude channels). We then apply the resulting regression relationship to the entire 1979-2015 period to estimate MSU3 (AMSU7) from MSU2,4 (AMSU5,9), and compare it to the raw intercalibrated global MSU3/AMSU7 time series. A difference time series of the regression estimated and the observed MSU3/AMSU7 time series is fitted with a piecewise linear estimator to give a time series of adjustment which are then applied to the MSU3/AMSU7 monthly anomaly fields. The resulting corrections cause a few hundredths of a degree per decade increase in the MSU3/AMSU7 trend (1979-2014), which ends up being very close to zero.

The following graph (Fig. 6) shows the resulting time series of LT, MT (mid-troposphere, from MSU2/AMSU5), TP (our new “tropopause level” product, from MSU3/AMSU7) and LS (lower stratosphere, from MSU4/AMSU9):

Fig. 6. Monthly global-average temperature variations for the lower troposphere, mid-troposphere, tropopause level, and lower stratosphere, 1979 through March 2015.

Fig. 6. Monthly global-average temperature variations for the lower troposphere, mid-troposphere, tropopause level, and lower stratosphere, 1979 through March 2015.

The LT computation is a linear combination of MSU2,3,4 or AMSU5,7,9 (aka MT,TP, LS):

LT = 1.538*MT -0.548*TP +0.01*LS

As seen in Fig. 7, the new multi-channel LT weighting function is located somewhat higher in altitude than the old LT weighting function. But if global radiosonde trend profile shapes (dashed line in Fig. 7) are to be believed, the net difference between old and new LT trends should be small, less than 0.01 C/decade. This is because slightly greater sensitivity of the new LT to stratospheric cooling is cancelled by even greater sensitivity to enhanced upper tropospheric warming.

Fig. 7. MSU/AMSU weighting functions which define the sensitivity of the various channels to temperature at different altitudes.  Also shown is the vertical profile of the average trends from two radiosonde datasets during 1979-2014, and the weighting function-sampled trends that would result from hypothetical satellite measurements of those radiosonde trends.

Fig. 7. MSU/AMSU weighting functions which define the sensitivity of the various channels to temperature at different altitudes. Also shown is the vertical profile of the average trends from two radiosonde datasets during 1979-2014, and the weighting function-sampled trends that would result from hypothetical satellite measurements of those radiosonde trends.

Specifically, we see from Fig. 7 that application of the old and new LT weighting functions to the radiosonde trend profiles (average of the RAOBCORE and RATPAC trend profiles, 1979-2014) leads to almost identical trends (+0.11 C/decade) between the new and old LT. These trends are a good match to our new satellite-based LT trend, +0.114 C/decade.

The new LT weighting function is less sensitive to direct thermal emission by the land surface (17% for the new LT versus 27% for the old LT), and we calculate that a portion (0.01 C/decade) of the reduction in the global LT trend is due to less sensitivity to the enhanced warming of global average land areas. The same effect does not occur over the ocean because all of these channels’ microwave frequencies are not directly sensitive to changes in SST since ocean microwave emissivity decreases with increasing SST in such a way that the two effects cancel. This effect likely also causes a slight enhancement of the land-vs-ocean trend differences. Thus, over ocean the satellite measures a true atmosphere-only temperature trend, but over land it is mostly atmospheric with a small (17%, on average) direct influence from the surface. One might argue that a resulting advantage of the new LT is lesser sensitivity to long-term changes in land surface microwave emissivity, which are largely unknown.

The rest of the reduction in the LT trend between Versions 6.0 and 5.6 (-0.016 C/decade) is believed to be partly due to a more robust method of LT calculation, and the new diurnal drift adjustment procedure, described later. It is well within our previously stated estimated error bars on the global temperature trend (+/- 0.040 C/decade).

2.2 Monthly Averaging Methodology
In order to compute gridpoint values of LT, we must first compute gridpoint averages of the three channels used to compute LT. We have a new methodology for computing monthly gridpoint averages from MSU channels 2, 3, 4 (AMSU ch. 5, 7, 9) which is based upon initially computing monthly gridpoint averages from all channels’ view angles separately: 6 view angles from 11 footprints of MSU, or 15 view angles from 30 footprints of AMSU, which are separately averaged in 2.5 deg. lat/lon bins during the month.

The resulting monthly Tb gridpoint averages for each of the three channels are then fitted as a function of Earth incidence angle with a second order polynomial. The Tb for any desired Earth incidence angle is then estimated from the fitted curve, rather than from the raw view-angle averages.

An example of this fit is shown in Fig. 8, for AMSU channel 5 for a single gridpoint for a single month from a single satellite (NOAA-15):

Fig. 8. Example of how monthly gridpoint averages of AMSU ch. 5 Tb from separate footprints are fitted as a function of Earth incidence angle so Tb can be estimated from the smooth functional fit to the data.

Fig. 8. Example of how monthly gridpoint averages of AMSU ch. 5 Tb from separate footprints are fitted as a function of Earth incidence angle so Tb can be estimated from the smooth functional fit to the data.

This new averaging procedure has the following advantages:

1) All of the different view angle Tb measurements are included in the optimum estimation of the Tb at the desired Earth incidence angle, reducing sampling noise.

2) The resulting average calculation for a gridpoint location is based only upon data from that location, a new feature that avoids sampling noise inherent in the old calculation of LT from geographically different areas.

3) The orbit altitude decay effect (which has been large only for calculation of the old LT), as well as different satellites’ altitudes, is automatically handled since we use routine satellite ephemeris updates to calculate Earth incidence angles, which are the new basis for Tb estimation, not footprint positions per se.

4) Working from monthly grids of separate view angle averages allows rapid reprocessing of all of the data from 1979 forward, allowing us to efficiently test the use of different nominal view angles for the products, matching of the MSU and AMSU view angles, changes in diurnal drift estimation, etc.

The nominal footprint position we use for all MSU channels (see Fig. 2) is footprint position 4 and 8 (Earth incidence angle 21.59 deg), rather than nadir position 6; and nominal footprint position 6.33 and 24.66 for AMSU channel 5 (Earth incidence angle 34.99 deg.); and Earth incidence angles 13.18 deg. for AMSU7 and 36.31 deg for AMSU9. The choice of MSU prints 4,8 is because the resulting sampling at those footprint positions gives approximately 28 measurements evenly distributed in longitude around the Earth twice a day, rather than only 14 samples if the nadir position (MSU print #6, or AMSU print #15,16) was used as the reference. We find this greatly reduces sampling noise in the middle latitudes caused by coincidental phasing of moving weather systems with the satellite orbital sampling patterns.

Nevertheless, a few months in the record still exhibit mid-latitude striping patterns (especially over the southern oceans) when the precession of satellite orbits combined with warm and cold air mass movements happen to lead to non-random sampling patterns, even with as many as three satellites operating. So, we apply a +/- 2 gridpoint smoother in the east-west direction for the monthly gridded anomaly grid fields, which is applied over land and ocean separately to prevent “bleeding” of signals between land and ocean.

2.3 Diurnal Drift Calculation
As the 1:30 satellites drift to later local observation times (an indirect result of orbit decay), the MSU2 (AMSU5) Tb tend to cool, especially over land in certain seasons, due to the day-night cycle in temperature. As the 7:30 satellites drift to earlier observation times, the Tb tend to warm for the same reason. These average relationships change at very high latitudes because the ascending and descending satellite passage times converge — while they are ~12 hours apart at the equator, they approach the same local time at high latitudes.

These diurnal drift effects are empirically quantified at the gridpoint level by comparing NOAA-15 (a drifting 7:30 satellite) to Aqua (a non-drifting satellite), and by comparing NOAA-19 against NOAA-18 during 2009-2014, when NOAA-18 was drifting rapidly and NOAA-19 had no net drift. The resulting estimates of change in Tb as a function of local observation time are quite noisy at the gridpoint level, and so require some form of spatial smoothing. Since they also depend upon terrain altitude and the dryness of the region (deserts have stronger diurnal cycles in temperature than do rain forests), a regression is performed within each 2.5 deg. latitude band between the gridpoint diurnal drift coefficients and terrain altitude as well as average rainfall (1981-2010) for that calendar month, then that relationship is applied back onto the gridpoint average rainfall and terrain elevation within the latitude band. Over ocean, where diurnal drift effects are small, the gridpoint drift coefficients are replaced with the corresponding ocean zonal band averages of those gridpoint drift coefficients.

Fig. 9 shows an example of the diurnal drift coefficients (in deg. C per hour of ascending node time drift) used for MSU ch. 2 at nominal footprint 4 (and for AMSU ch. 5, a nominal footprint position between #6 and #7) for the month of June:

Fig. 9.  Example diurnal drift coefficients (deg. C/hr) for MSU2/AMSU5 for the month of June for adjustment of the afternoon (“1:30”) satellites.

Fig. 9. Example diurnal drift coefficients (deg. C/hr) for MSU2/AMSU5 for the month of June for adjustment of the afternoon (“1:30”) satellites.

The reason why the drift coefficients change sign at high northern latitudes is a combination of early sunrise time in June, late sunset time, and the fact that the ascending and descending orbit satellite observations at high latitudes approach the same time, instead of being 12 hours apart as they are at the equator.

We also compute and apply diurnal drift coefficients for MSU channels 3 and 4 (AMSU channels 7 and 9), but the drifts and resulting adjustments are very small.

3. Final Comments

This should be considered a “beta” release of Version 6.0, and we await users’ comments to see whether there are any obvious remaining problems in the dataset. In any event, we are confident that the new Version 6.0 dataset as it currently stands is more accurate and useful than the Version 5.6 dataset.

The new LT trend of +0.114 C/decade (1979-2014) is 0.026 C/decade lower than the previous trend of +0.140 C/decade, but about 0.010 C/decade of that difference is due to lesser sensitivity of the new LT weighting function to direct surface emission by the land surface, which surface thermometer data suggests is warming more rapidly than the deep troposphere. The remaining 0.016 C/decade difference between the old and new LT product trends is mostly due to the new diurnal drift adjustment procedure and is well within our previously stated range of uncertainty for this product’s trend calculation (+/- 0.040 C/decade).

We have performed some calculations of the sensitivity of the final product to various assumptions in the processing, and find it to be fairly robust. Most importantly, through sensitivity experiments we find it is difficult to obtain a global LT trend substantially greater than +0.114 C/decade without making assumptions that cannot be easily justified.

The new Version 6 files are located here:

Lower Troposphere:
Lower Stratosphere:

214 Responses to “Version 6.0 of the UAH Temperature Dataset Released: New LT Trend = +0.11 C/decade”

Toggle Trackbacks

  1. RW says:

    Wow, thanks for the detailed info and update, though I don’t think I can figure it out…

  2. Kristian says:

    Thanks, Roy 🙂 Much anticipated!

  3. FAH says:

    Dr. Spencer,

    I did not see in the post a mention of when the new Version 6 data set will be available for the general public. Do you expect that next month’s update will have a link to the new data instead of the data from the previous version? Is an updated data set available for download now?


  4. FAH says:

    Oops,sorry I just saw the links at the bottom and answered my own question.

    Thanks and nevermind!

  5. pochas says:

    A Herculean effort. The three authors deserve our thanks and appreciation for their dedication to this important task.

  6. FAH says:

    Dr. Spencer,

    Is it my imagination or does the Version 6 data generate much lower trends for the past (18, 15, 10 years, etc. you pick) than the previous version. It looks to me like UAH and RSS have converged closer in that time period, both departing from the GISS/HADCRUT world. Based on what you had said the changes were expected to be I was not expecting that.

    Regards. Appreciate your work.

  7. David Johnson says:

    Many thanks, great work

  8. dave says:

    A lot of work, indeed;


    agreeing with recent RSS picture.

  9. Werner Brozek says:

    Do you have a new ranking of all years along with their average anomalies? Thank you!

    • Rank, Year, Deg. C (vs 1981-2010 avg.)
      1st 1998 0.463
      2nd 2010 0.333
      3rd 2002 0.195
      4th 2005 0.181
      5th 2003 0.166
      6th 2014 0.151
      7th 2007 0.144
      8th 2013 0.113
      9th 2006 0.098
      10th 2001 0.095
      11th 2009 0.087
      12th 2004 0.060
      13th 2012 0.049
      14th 1995 0.048
      15th 1987 0.026
      16th 2011 0.021
      17th 1988 0.014
      18th 1991 -0.004
      19th 1990 -0.012
      20th 1997 -0.031
      21st 1996 -0.034
      22nd 1999 -0.036
      23rd 2000 -0.041
      24th 1983 -0.065
      25th 1980 -0.067
      26th 1994 -0.088
      27th 2008 -0.114
      28th 1981 -0.136
      29th 1993 -0.225
      30th 1989 -0.232
      31st 1979 -0.236
      32nd 1986 -0.243
      33rd 1984 -0.260
      34th 1992 -0.306
      35th 1982 -0.320
      36th 1985 -0.385

  10. Grant Ceffalo says:


    Thank you for your hard work! I appreciate that you publish not only your conclusions, but also your data and your methodology, like a good scientest should.

    I appreciate your data files. Is there any way to get them in comma separated variable (.csv) or tab separated? I have to manipulate them to put them into a format I can use.

    Again, many thanks!

    • what program are you using? I can’t imagine it doesn’t allow space-separated values.

      • Grant Ceffalo says:


        It was my error in the path I was importing the data into excel. I tried another way and it works fine.

        My first method (through Word) was somehow putting extra spaces such that when I was importing to excel and parsing using spaces, I was getting extra cells.

        When I imported after saving to a text file, the parsing worked without error.

        Thanks again!

  11. Figure 7 seems to indicate that the lowest 200 meters or so of the troposphere has had a warming trend .02-.03 degree/decade faster than the lower troposphere as a whole, according to radiosonde data.

    Is anyone here aware of a map showing worldwide distribution of the lowest troposphere warming faster than the lower troposphere as a whole? I think there could be hotspots of this phenomenon where albedo has decreased but convection usually does not occur. Otherwise, I wonder about nighttime cooling of the surface being decreased by change of radiation transfers due to increase of greenhouse gases.

  12. Dr. Spencer, thanks for this update.

  13. Steve C says:

    Fascinating stuff – thanks for the post, Dr. S. As a rank amateur who occasionally manages to get a half-decent picure from the NOAA birds on 137MHz, using completely unsuitable equipment, I’m probably even more impressed by what the professionals are doing with their proper kit than most. Very nice work.

  14. Svend Ferdinandsen says:

    Good post, even if i have not consumed it all.
    This dataset really differs from all the others, that allways increase the trend every time they make a new version or just a new compilation of the global temperature.

  15. Werner Brozek says:

    According to RSS, 2014 was the 6th warmest and after 3 months in 2015, RSS is on track for 2015 to be 5th warmest.

    Version 5.6 had 2014 as third warmest and after 3 months in 2015, 5.6 was on track to be 3rd warmest.

    However version 6 now has 2014 to be 6th warmest and after 3 months, version 6 is on track for 2015 to be 5th warmest, exactly like RSS.

    As soon as I find out how long the period of pause is on the new data set, I will let you know. RSS is 18 years and 4 months and version 5.6 was at an even 6 years to the end of March.

    • oh, we already know the answer to that. But we let others calculate it, too, just to spread the fun around. 😉

      • cunningham says:

        Dr. Spencer, you’ve stated in the past that uah was a better fit with the balloon data than rss. In layman’s english (please…), what do you now say about the uah fit with the balloon data?

      • fonzarelli says:

        Dr. S., i’m surprised you’re not out here answering more questions like the pope with his faithful! (after all, this is HUGE!!!) Congrats and thanx to you on this your big day and many more thanx for enriching our lives every day…

  16. myrightpenguin says:

    Congratulations, obviously a lot of painstaking efforts went into this.

  17. Eli Rabett says:

    When will the v6.0 code be posted on

    where you are required to make it available so that it may be audited?

    • after the datasets have gone through beta review by users to make sure there are no obvious problems.

      • It would be a good idea to release the code now. So that people might read it, and spot errors that way.

        Code review is generally reckoned to be an excellent idea.

        • geran says:

          When are you scheduled for your next “review”?

        • Rob says:

          The V6.0 is still in beta (stated a number of times) and as such the beta reviewers are testing it before a more general code release. I believe that previous versions of the UAH methodology are already published so of course Dr Spencer can be trusted to do the same here.

          I just wonder if certain commentors would have been so quick to ask for the code if the adjustments had resulted in changes in the other direction….

        • DZ says:

          Code review is an excellent idea. I just wish that pesky William Connolley would get his friends to release code..they won’t even release it when courts ask for it, I am glad there is so much more transparency here 🙂

          • Eric says:

            Volunteering for code review. Being professional sw developer I think I can contribute a lot to code quality, readability, efficiency, commenting and more.

            Crowsourcing open software distributed development, buzzwords of the century.

        • Gordon Robertson says:

          @William Connolley…things getting a bit sticky over at realclimate, Bill? You guys starting to sweat it with 17 years of no warming trend?

          Need to check the sat code in the faint hope you will find something wrong?

          How’s things on wiki, Bill? Still taking shots at Fred Singer?

          Has Gavin figured out what positive feedback is?

          • Gordon Robertson,

            “You guys starting to sweat it with 17 years of no warming trend?”

            No, because, apparently in contrast to you, we actually know what the scientific method is, which includes to evaluate all the observational evidence, instead of cherry picking data (i.e., selecting subsets of data or datasets in a way that a preconceived view is confirmed), and to not ignore arguments based on statistical analyses.

        • Gordon Robertson says:

          @William Connelly “It would be a good idea to release the code now. So that people might read it, and spot errors that way”.

          You mean errors in the code or errors in the science?

          My understanding is that you have a degree in computer science. How would that help you spot errors in the science used in satellite telemetry?

          Based on the theories at realclimate I can’t see how you’d be able to spot errors in real science. I don’t equate climate modeling with hard science, it’s more like social science.

      • …also, we are not “required” to provide the code. NOAA didn’t pay for Version 6. But we will, since we’re such nice guys. 🙂

        • Eli Rabett says:

          Now some, not Eli to be sure, might point out that data sets are not code. something that needs to be stated up front. The version 1-5 data sets were released, the code, not until about 2013.

          Still, allow Eli to remind you of NASA’s policy on open source software

          NASA Open Source Software

          NASA conducts research and development in software and software technology as an essential response to the needs of NASA missions. Under the NASA Software Release policy, NASA has several options for the release of NASA developed software technologies. These options now include Open Source software release. This option is under the NASA Open Source Agreement “NOSA”.

          The motivations for NASA to distribute software codes Open Source are:

          To increase NASA software quality via community peer review

          It’s a moving train Roy, you better jump on.

          • Since Dr. Roy is a NASA instrument scientist, he doesn’t need to be lectured about NASA policy. (but this is not a NASA-funded project).

            Allow Dr. Roy to suggest Eli stop acting like a poser.

          • Eli Rabett says:

            Since NASA requires that only the results of peer reviewed science can be used, that leaves v6.0 a couple of ears out there, no?

          • Gordon Robertson says:

            @EliRabbett…”NASA requires that only the results of peer reviewed science can be used, that leaves v6.0 a couple of ears out there, no?”

            Are you talking about NASA proper or the climate modeling division at GISS? If it’s GISS, why would they require peer review when they run unvalidated models that do not meet the requirements of the scientific method?

            Peer review is not a requirement of the scientific method and the way peer review is used today, it’s anti-science. It is used to prevent people like Roy, John Christy, and Richard Lindzen from publishing papers.

            Enough of your crap. If you don’t have the guts to post under your real name as a scientist then don’t berate people for not going through a corrupted peer review system.

          • Ric Werme says:

            Odd, if NASA says “that only the results of peer reviewed science can be used,” why are William and Eli so interested in using the V6 product?

            If a phrase was accidentally dropped from after “used,” that describes who is saddled with that restriction, I’d be interested in seeing it.

            I am heartened that you respect community peer review. I wish more climatologists did.

    • DZ says:

      Well Eli, likely can’t help you there. He never helped his other friends review their math, or release anything. Good thing your transparency allows for people to simply see the equation adjustments here, and apply them themselves in order (except for those that have difficulty with math like the rabbit).

      Thank you guys very much for your open and transparent approach. It is interesting to see this correspond even more greatly with the RSS dataset, I look forward to further review.

    • Gordon Robertson says:

      Eli Rabett…why don’t you use your real name, Josh, when consulting a fellow scientist who uses his real name?

      What exactly are you trying to hide?

    • Gordon Robertson says:

      @Eli Rabett…hay Josh, have you recovered from the butt-kicking you took from Gerlich & Tscheuschner on their rebuttal to your paper?

      Even I spotted the flaw in your argument. You can’t tell the difference between heat and infrared.

  18. geran says:

    My prediction for the next UAH update (for March 2015) is +0.36º.

    I’m curious if this latest version of the software will impact my prediction!


  19. ossqss says:

    Thanks Doc!

    How refreshing! Transparency in climate science is a rare and welcome sight!

  20. Slipstick says:

    I am shocked…shocked!…to see a climate scientist manipulating the data! Clearly you are trying to obfuscate the correlation between TSI and PDO-ACO-M-O-U-S-E and the 1.21 jiggowatts/day (no uncertainty) of adiabatic cooling that has been occurring since the Medieval Warm Period, the hottest period in the history of the planet, that is causing the global subsidence that makes it appear that the Mean Sea Level is rising. Alert the blogosphere!

    Seriously, though, congratulations on the reprocessing and many thanks for the overview of the changes, an interesting read.

  21. Andrew says:

    From the charts it looks like measurements at the poles are not taken

    How much of the poles are not measured ( i.e. degrees/ square miles )?


  22. Slipstick says:

    I was curious about the Dec ’84 – Jan ’85 blip in the V5.6 v. V6.0 anomaly. I’m guessing it is related to the failure of NOAA-8, but I thought NOAA-8 died earlier in ’84, although the earlier chart shows some data from -8 in mid-’85. Can you clarify?

  23. dave says:

    Measurements are taken from 85 N to 85 S.

  24. Walt Allensworth says:

    Dr. Spencer,
    As a techno-weenie I really enjoyed that write-up.
    So, interestingly, this will bring UAH more into alignment with RSS, which will not please the CAGW faithful.

    • lewis says:


      The true believers are not concerned with what science says. They are far beyond the science and into mythology. From that shall come political and economic control and power.

      The naive Dr. Spencer will be found guilty of blasphemy and fed to the blogsharks.

      Dr. Spencer,

      Interesting work you carry on. Much math and checking behind oneself. It obviously requires great perseverance and focus.

      But the results seem to be that it is 20% colder than we had been told. Is that what is causing all the snow and ice?

      Best wishes and, as always, thank you for taking the time to keep this blog current and interesting.

      Lewis Guignard
      Crouse, NC

  25. tom0mason says:

    Dr. Spencer thank-you for this update, and all your hard work and dedication over the years.

    No doubt as things ramp up to the Paris UN-IPCC meeting and evermore ridiculous claims are heard across the media, I shall be comforted that at least here, at your site, there is a knowledgeable, logical and truly scientific source of honest unbiased opinion.

    Again I say a big thank-you.

  26. Walter Dnes says:

    Dr. Spencer, can you please confirm that is the upgraded version of ? I follow it for personal interest at home, and plot it on a Google spreadsheet, with various other temperature indices, at

  27. Jim says:

    Can’t thank you enough for the hard work and dedication you have put into this work as well as keeping us well informed and up to date.

  28. JohnM says:

    A full explanation of reasoning and data. What a rare sight in the climate science world!

    • William Connolley says:

      But no code. You don’t care about the code, yes?

      • Sun Spot says:

        blah blah blah Billy

      • David Johnson says:

        Are you being wilfully obtuse Connelley or did you not see Spencer’s earlier reply?

      • DZ says:

        If only William Connolley and his friends cared about any measure of transparency, courts come and they still hide behind each other.

        It is refreshing to see this, so much more open than those Connolley is paid by, I mean friends with, I mean colleagues with, I mean the group of friends that review each other’s papers..I mean the people who refuse to even show courts their code when ordered..I mean the people who hide behind sophistry and show virtually no work and have no open-ness besides saying my friends looked at it..I mean..oh you know what I mean.

  29. Transport by Zeppelin says:

    Dr Spencer;

    The new version 6.0 UAH produces a warming trend in the mid troposphere of 0.069 C/decade & 0.114 C/decade for the lower troposphere .

    The RSS data produces a warming trend in the mid troposphere of 0.077 C/decade & 0.122 C/decade for the lower troposphere .

    A very slight difference between the two data sets figures is shown above, but, what is interesting is, both data sets give equal results when comparing the difference between the rates of warming in the mid troposphere & the lower troposphere. Both produce a faster rate of warming in the lower troposphere of 0.045 C/decade compared to the mid troposphere.

    This is positive lapse rate feed back.

    Dr Spencer; considering the IPCC report stated that –

    “models with a larger (negative) lapse rate feedback also have a larger (positive) water vapour feedback”

    does your data & the data of RSS prove water vapor feedback is in fact negative.

    • Ken Gregory says:

      It is indeed very interesting that both the RSS data and the new version 6.0 of UAH show exactly the same difference of the LT warming trend minus the MT warming trend of 0.045 C/decade.

      Climate models predicted that the MT should warm faster than the LT, but both satellite data sets show the opposite.

      The models show a positive water vapor feedback by calculating that relative humidity in the upper troposphere should stay approximately constant with warming, so specific humidity would increase. But higher humidity also decreased the lapse rate, which is the the rate at which temperatures decline with increasing altitude. The lapse rate is a negative feedback (in the models), which only partially offsets the positive water vapor feedback. Warmer oceans cause more evaporation which releases more latent heat in the upper atmosphere, decreasing the lapse rate.

      The satellite data shows that the lapse rate does not become less with warming, but increases slightly as the LT warms faster than the MT, implying that the lapse rate feedback is positive. However, changes in water vapor profile is the major driver of both the water vapor feedback and the lapse rate feedback, which is the reason that there is much less variation between models of the sum of the two feedbacks compared to the variations of each of the feedbacks individually.

      This also suggests that specific humidity in the upper atmosphere declines with increasing temperatures. The dry lapse rate is 9.8 C/km, the saturated lapse rate is abut 5 C/km, and the actual lapse rate is about 6.5 C/km. So, I would say the satellite data strongly suggests that the water vapor feedback is negative, and the sum of the negative water vapor plus the positive lapse rate feedback is also negative, but just slightly negative.

      On the other hand, perhaps the higher LT warming than the MT warming is due to contamination of the LT warming by the direct surface emissions, rather than just the emissions from the near surface air.

      • Increasing lapse rate as a result of warming and/or increase of greenhouse gases appears to me as a sign that the lapse rate feedback is a negative one. The increasing overall tropospheric lapse rate, in combination with cooling at the tropopause level (by increase of greenhouse gases), would increase convection of heat from the surface to the tropopause level.

      • the layer trends from satellite are very close to those computed from radiosonde data (see Figure 7). So, the satellite data can be said to be “consistent” with the radiosonde trend profile, which shows weak negative lapse rate feedback. But an infinite number of radiosonde trend profiles would also be consistent with the satellite measurements.

        In models, negative lapse rate feedback is at least counterbalanced by positive water vapor feedback. They tend to occur together. One possibility is that lapse rate feedback is nearly zero, and that water vapor feedback is also nearly zero.

        As I have addressed ad nauseum, water vapor feedback is dominated by free-tropospheric vapor, not boundary layer vapor, and is controlled by microphysical processes in precipitation systems, the temperature dependence of which (in a climate change sense) we don’t even understand, let alone have accurately included in climate models.

        • Gordon Robertson says:

          @Roy ….”In models, negative lapse rate feedback is at least counterbalanced by positive water vapor feedback”.

          Correct me if I am wrong but did you not once describe atmospheric positive feedback as a not-so-negative negative feedback?I understood what you said above, that models are using a positive feedback but do they mean it in the same was as I think you once described it?

          If not, they are wrong. Positive feedback in physics requires an amplifier. Some people think positive feedback itself can cause amplification but it can’t. PF is part of an amplifier circuit, the gain is caused by an amplifier like a transistor, which depends on an external power supply.

          Gavin Schmidt of GISS once laid out a description of positive feedback in an equation. Engineer Jeffrey Glassman debunked his equation claiming that it lacked a gain factor.

          See section titled “Gavin Schmidt on positive feedback”

          I think Glassman is right, and since Schmidt is programming climate models it is plain that modelers are all wet about positive feedback.

      • Will Kernkamp says:

        Something that struck my eye is that LS cooling has stalled during the pause. This cooling should happen in direct proportion to the CO2 concentration which continued to increase. This is so because the CO2 is the emitter that balances the O3 absorption of incoming SW. This LS pause suggests that the ozone layer traps more SW radiation as the temperature reduces. If true, this effect would reduce ECS.

  30. Kasuha says:

    Thank you very much for the update and extensive description.
    It seems to me that closes a lot of the gap between RSS and UAH, does anyone have a comparison? There goes the reasoning that ‘RSS is flawed’ by people who didn’t want to see their trendlines pointing down over recent periods, I guess. It’s going to make many ‘skeptical’ ‘pause watchers’ very happy 🙂

    • dave says:


      The peak of the satellites’ “brightness temperatures”, for the global Lower Troposphere, was attained in April, 1998. According to RSS the present (March 2015) figure is 0.6019 C lower, and according to UAH it is 0.61 C lower.

      • nigel says:

        “…There goes the reasoning that ‘RSS is flawed’…”

        Immediately to be replaced by ‘Both RSS and UAH are flawed’.

        I do not know why anyone bothers. Every form of data shows a lengthening pause. It is merely a matter of ‘face’; whether one can more credibly shout it is an ‘uppy’ pause or a ‘downy’ pause. In stock-market parlance, a ‘flying pennant’ which is probably just an interruption in a continuing move, or a ‘distribution pattern’ before a bear market. Reading the charts is hardly a science, in any case.

        Comparing LT April 1998 to March 2015 in UAH 6.0., one sees that the Tropics had a positive anomaly in 1998 of 1.16 Degrees C but at present there is no anomaly at all (0.01 C).

        I suppose it is the difference between a proper El Nino and a feeble one – as at present – when the Trade Winds do not weaken and the local mixing of the Pacific Ocean is little affected.

  31. dave says:

    I said peak in temperatures. I meant peak in anomalies of temperature, of course. Everybody knows that the “brightness temperatures” themselves fluctuate in a completely regular cycle, by some 3 Degrees C during a year; and the actual global “brightness temperature” is higher in July than April, because the land-rich Northern Hemisphere is tilted towards the sun. This is despite the fact that the earth is further from the sun at, that time. Nobody said this climatology was simple. Oh, wait – yes, they did.

  32. nigel says:

    “…in July…”

    During the “Climatic Optimum” of nine thousand years ago, the Northern Summer coincided with the earth being closest to the sun. This is consistent with estimates that summer temperatures in Europe and the Middle East were a couple of degrees higher, during the spread of the first high cultures, than now.

  33. Euan Mearns says:

    Roy, many thanks for this update and much hard work. My initial reaction was “oh no not another adjustment to the temperature record”. But why should it be OK for you to adjust the satellite record while we may protest about GHCN adjusting their homogenisation algorithms? I think I have an answer to that but I’m interested to see what others may think first.

    Myself and Roger Andrews have been making compilations of GHCN V2 “good” records. This comparison of my summary of 174 S Hemisphere records and UAH.

    And this Roger Andrew’s summary of over 300 S Hem records:


    • dave says:


      Because I believe Spencer et al, and RSS, probably to be honest, and the other providers of data to be less obviously so.

      Do you know what the Chair of one of the bodies entrusted with coming up with variation of the homogenization algorithms calls people like you, who try to think for themselves, using the Internet among other resources? “Click whores!” Nice, eh?

      • Mike M. says:


        “Because I believe Spencer et al, and RSS, probably to be honest, and the other providers of data to be less obviously so.”

        In other words, you like Spencer’s results better than the others? That is an unscientific basis for a judgement.

        In a very important sense science is always wrong. What makes science special is that it has a method of becoming less wrong. That process requires constant adjustments, as we see here, and in the surface data sets. Hopefully, such adjustments lead to a convergence of results, as we now have between UAH and RSS.

        Eventually, the normal scientific process should lead to a resolution of the differences between the satellite and surface temperatures. Perhaps one or the other will end up being right, or the end result will be something in between, or they are both right and we will come to understand exactly how it is that they are measuring subtly different things.

        The proper scientific approach is to suspend judgement and/or to work to figure out what causes the difference.

        Accusations of fraud (from which you have correctly refrained) interfere with the free debate and will likely slow down the process. Sad.

        • An Inquirer says:

          Although the word “adjustment” is used for both adjustment to the satellite record and adjustment to the surface record, the concept and procedure of adjustments are starkly different. No doubt others can explain it better that I will, but I will make a stab at it:

          In the surface record, you have actual observations of thermometers. These observations have been recorded, and we cannot go back and re-observe them. For the satellite record, you have a process to estimate temperatures based on an information technology procedure. We still have the observations fed into the algorithm to produce that estimate. If we can improve the algorithm, then we can improve the estimates.
          Also, in the surface record, there are ten of thousands of observation points, and we cannot be certain what happened at each point in terms of station moves, UHI, and site compromises. Therefore, computer programs are written to estimate what happened. We can go back and spot check whether those computer programs are giving us accurate answers, and we have found that the programs many times have been wrong. For the satellite record, we have a small number of observation activities, and we have excellent documentation on what has happened (drift, etc.) that can affect the estimate.
          Furthermore, the results for the surface record are not consistent with actual observations. (A couple of examples: Great Lakes having record ice, while the adjusted surface temperatures show normal temperatures in Great Lake states. Also, adjusted temperatures showing record heat, but highest actual temperatures recorded per state have not increased.) Meanwhile, The adjusted satellite record produces results that are consistent with real-life phenomena.

          • Euan Mearns says:

            This is pretty well aligned with my view. The satellite record is remote sensing. What is measured gets processed to convert it to a temperature. If you come up with a better way of processing then you are entitled to use it. But we always need to be aware of introducing bias. The fact that V6 converges with RSS is good – we all think so any way. But need to be aware of confirmation bias.

            The thermometer record is different. Thermometers are supposed to measure air temperature “directly”. There is no doubt that there are a myriad of things that can go wrong, and since you can’t go back and re-meaure, what to do? There is little doubt in my mind that bot driven homogenisation is folly. I have a strong preference for a few hundred hand curated records where physical changes to the measuring environment are “quantitatively” built into the measured record. Latitude, altitude, TOD, shade and urban heat are I believe the main variables. Maybe we should give up on the thermometer record?

          • William Connolley says:

            This is embarrassingly naive. Our host would explain your mistakes, if he could be bothered.

          • mpainter says:

            I’m sure that everyone is thankful that we have the satellite temperature data as a balance against over manipulation of the temperature data sets. Everyone, that is, except for the manipulators of data (and also of the documents, records, manuscripts, etc&etc&etc….&etc…

          • Gordon Robertson says:

            @Euan Mearns “The satellite record is remote sensing. What is measured gets processed to convert it to a temperature….. The thermometer record is different. Thermometers are supposed to measure air temperature “directly””.

            They are both proxies. Thermometers do not measure temperatures directly. They measure an expansion of mercury, or similar substance, in a narrow column. Sat telemetry measures the microwave radiation of oxygen molecules with the frequency of the emission corresponding to the ambient temperature.

            Actually, what is being measured is the kinetic energy of the oxygen molecules, which corresponds to heat. Thermometers are responding to the same average kinetic energy of air molecules. The difference between the two methods is where the real problem lies.

            Thermometers record a daily high and low and the two are averaged. Sat telemetry samples billions and billions of data points (O2 molecules) in one position of their scanners. Therefore they are giving a far more accurate temperature record in a space at one time.

            The sat scanners are far more comprehensive, covering 95% of the lower atmosphere globally as opposed to about 30% for thermometers. It is a small price to pay for the adjustments of orbits and differences between satellites to get the vast amount of temperature data available to the sats.

            I have worked in electronics for decades and such adjustments are not foreign to me. The system can be made precise.

          • An Inquirer says:

            When Dr. Spencer says “what Mike M. said,” I understand that Dr. Spencer is agreeing with these two statements from Mike M:

            “The proper scientific approach is to suspend judgement and/or to work to figure out what causes the difference.
            Accusations of fraud (from which you have correctly refrained) interfere with the free debate and will likely slow down the process.”

            I hope that we all agree with Mike M. for those two statements.

            My problems with Mike M’s post is the assertion that we believe Dr. Spencer to be honest because we like the results. It is not the results which are scientifically attractive. Rather, it is easier to square Dr. Spencer’s adjustments with the scientific process than adjustments to the surface temperature. A priori, I did rule out the possibility of legitimacy of surface adjustments are legitimate, but the task is much more difficult for them and the results are not consistent with known phenomena.

          • MarkB says:

            A priori, I did rule out the possibility of legitimacy of surface adjustments are legitimate,
            “. . . did rule out the possibility . . .” – Freudian slip? 🙂

            but the task is much more difficult for them
            Maybe this is simplistic, but the diurnal drift problem for satellite observations seems completely analogous to the time of observation problem in ground measurements. Also inter-satellite calibration is analogous correcting for station moves in the ground data. Both have to deal with station calibration/drift problems. At a conceptual level, I don’t see much difference.

            In a nutshell, it seems like the satellite guys have better records of changes but a very small sample set on which to derive statistically based corrections. The surface measurement guys have orders of magnitude more observations, but typically less insight into the performance and history of any specific site.

            and the results are not consistent with known phenomena.

          • An Inquirer says:

            Mark B.,
            My apologies. Not a Freudian slip, but way too often I leave off the word “not” when I am typing fast. My mistake.

            Your post does sound logical and insightful — until you ruin it with your final comment of “balderdash.” If you are not aware of the numerous conflicts between the results of surface temperature adjustment results vs. known phenomena, then your credibility in other statements is undermined.

            Again, I would not claim that the satellite measurements or adjustments are perfect, but I am not aware of their results being in conflict with known phenomena.

  34. Jeff Id says:

    Doc, Doc, Doc,

    You’ve got it all wrong. Corrections to temperature trends must always be warmer. It’s in the IPCC handbook for climate science somewhere.

    Thanks for the hard work and the update. I wonder how people will react to the increased disparity between sat and land.

    • Mike M. says:

      Jeff Id:

      “Corrections to temperature trends must always be warmer. It’s in the IPCC handbook for climate science somewhere.”

      Corrections to land T have generally made the trend larger, but correct to sea surface T have made the trend smaller. Overall, the corrections have little effect on the global trend in the surface T.

      The scientists doing this are by-and-large, like Dr. Spencer: honest people working hard to get things right.

      • Jeff Id says:

        Mike M,

        I’ve been around this information for quite a while now and am aware of the adjustments and why they exist. Hell, I’ve even been misquoted in two psychology papers for the fact.

        Here is a link to a global temperature calculation, done myself with the adjusted ground data some years ago.

      • Jeff Id says:

        The sea surface temp history is a serious mess and it is dominant in the data. You might want to take a more skeptical look at it.

  35. Doug Cotton says:

    Roy, the yellow line in this plot shows how you should calculate the natural long-term trend which will reach a maximum by about the year 2058 and then head downwards for nearly 500 years.

    About 120 years ago the trend was 0.06°/decade, whereas now it is about 0.05°/decade, which is to be expected seeing that the maximum could be 40 to 50 years from now.

    • dave says:

      Up and down, on various scales. Quite plausible.

      It has been mentioned before, but…

      In Britain during the 1920’s and 1930’s there was a surge of suburban house building. As the climate seemed to be steadily improving (in those days, warming was a good thing), all the plumbing was put on outside walls, with the approval of local authorities. In the colder winters of the 1940’s and 1950’s the pipes and lavatories regularly froze, therefore. Many, like my parents, cursed those government “experts” who did not realize that matters naturally go in cycles.

      • James Strom says:

        Dave, thanks for that great story. I spent some fretful nights this past cold winter wondering about water pipes routed just inside my walls.

        But it could have been worse. Think of Fukushima.

  36. Werner Brozek says:

    Bob Tisdale has the pause for version 6 at 18 years and 3 months here, which is basically the same as RSS at 18 years and 4 months. See:

    • David A says:

      At what level is that statistically significant?

      • Werner Brozek says:

        With respect to statistically significant warming, that is obviously going to increase greatly.
        With version 5.6, Dr. McKitrick had it at 16 years while Nick Stokes had it at 18 years and 8 months.
        For RSS, Dr. McKitrick had it at 26 years while Nick Stokes had it at 22 years and 3 months.
        But with the pause being 18 years and 3 months on version 6, I would say Nick’s time would increase by 4 years to over 22 years.
        As for Dr. McKitrick, notice that most of the points for RSS since last April when he made his calculation for RSS are below the trend line. So if he were to calculate RSS today, he could get 27 years. I have no clue about UAH6, but I would not be surprised if it would be similar.

        • David A says:

          Werner, you’re the one spreading these numbers around. That makes it your responsibility to indicate their statistical significance, including as much autocorrelation as you find is relevant. Do the calculations.

          If it takes 26 years for some other dataset to be statistically significant, then your numbers likely aren’t. Right?

    • Philip Shehan says:

      According to this trend calculator by Kevin Cowtan,

      The trend since 1979 for satellite data with 95% confidence limits are

      UAH Trend: 0.139 ±0.065 °C/decade (2σ)

      RSS Trend: 0.122 ±0.065 °C/decade (2σ)

      There is clearly “statistically significant” warming.

      The trends are in line with those given in the text (the trend calculator data may not be as quite as up to the month as that here) but no confidence intervals are given here.

      Anthony Watts says that version 6.
      shows “no warming” for 219 months (beginning January 1997) with a linear trend of 0.00 °C/decade, and for 220 months (since November 1996) for RSS data. No confidence limits are given.

      The trend calculator gives these results for UAH version 5.6 and RSS:

      UAH Trend: 0.101 ±0.180 °C/decade (2σ)

      RSS Trend: -0.001 ±0.175 °C/decade (2σ)

      Again the headline trend for RSS matches that given by Watts, but the confidence limits mean that actually there is a 95% chance that the trend is somewhere between warming of 0.174 °C/decade and cooling of 0.176 °C/decade. I suggest this is not within a bulls roar of statistical significance as far as a declaration of a “pause” or a “hiatus” or no warming is concerned.

      Indeed the data supports the null hypothesis, that the difference between the shorter period and statistically significant warming period cannot be excluded as being due to chance.

      The same goes for UAH version 5.6 data:

      Trend: 0.101 ±0.180 °C/decade (2σ)

      I would like to see the 95% confidence limits for version 6.0, from 1979 and 1997, but I would be very surprised if that changed the situation.

      In short Watts is making claims for the new version (and indeed the old) that are invalid.

      I would like to hear what Werner, David A and Dr Spencer have to say on this.

      • Gordon Robertson says:

        @Philip Shehan “The trend since 1979 for satellite data with 95% confidence limits are
        UAH Trend: 0.139 ±0.065 °C/decade (2σ)
        RSS Trend: 0.122 ±0.065 °C/decade (2σ)”

        Philip…that was explained in the UAH 33 year report.

        Many people analyzing the data do not take into account that it is based on the 1980 – 2010 average and that the first 17 years of the record were under the baseline (cooling)due to volcanic aerosols and such. That is, the fist 17 years show a slight cooling, therefore that warming was not true warming and included a recovery from cooling.

        It was not till the large El Nino in late 1997 that the record showed true warming and rose above the baseline.

        see bottom of page 2 and top of page 3:

        “Part of the upward trend is due to low temperatures early in the satellite record caused by a pair of major volcanic eruptions,” Christy said. “Because those eruptions pull temperatures down in the first part of the record, they tilt the trend upward later in the record.”

        “Christy and other UAHuntsville scientists have calculated the cooling effect caused by the eruptions of Mexico’s El Chichon volcano in 1982 and the Mt. Pinatubo volcano in the Philippines in 1991. When that cooling is subtracted, the long-term warming effect is reduced to 0.09 C (0.16° F) per decade, well below computer model estimates of how much global warming should have occurred.”

        One of the tenets of statistics is to define context. That’s why it is dangerous to take numbers over a range and jump to conclusions. If you look at the 33 year UAH record it has discontinuities that change the context over the record.

        One major discontinuity is the 1997/98 El Nino, which seems to have produced a residual warming following it. Immediately following the El Nino, there is a dip below the baseline then the warming abruptly rises in 2001. There is a similar discontinuity in 2010.

        You simply cannot enter the UAH data into Excel and expect to get an objective trend over the 33 year range that means a lot.

        There’s no argument that a few degrees C warming has occurred but it has not been a smooth warming over the range, rather most of it occurred in one year, 2001, following a major El Nino spike. That’s not the signature of anthropogenic warming and the trends you quote above certainly don’t represent AGW.

        • Philip Shehan says:

          Thanks for the reply Gordon but I was really interested in what the 95% confidence limits for the version 6 are.

  37. Mike M. says:

    Dr. Spencer:

    Is there some way I can get the numbers, or formulas, for the weighting functions in Figure 7?

    • MarkB says:

      Mike M,
      For what it’s worth, I plotted the nominal weighting functions for the nadir view as described in the link (MSU CH 2,3,4 data) and got something similar, but not identical to Figure 7.

      I haven’t looked hard for representative RAOB trend data, but what I found indicates that uncertainties are large.

      It’s clear that the V5.6 and V6.0 algorithms are measuring somewhat different regions of the atmosphere, but it’s not clear (to me at least) how much this matters. From similar charts on the REMSS site, it looks like their TLT region is virtually the same as V5.6 but, again, it’s not clear how much this matters for comparison of the two.

  38. Charlie says:

    Is that paper published anywhere in the scientific literature? Peer-review?

  39. mpainter says:

    Congradulations to you, John Christy, and Wm. Braswell on this comprehensive achievement – three years worth!

    Most interesting is the gridcell trend map, fig. 4. I wonder what certain temporal segments of that would show, say, from 2002 to present, or other segments, say, 1993 to ’98. That would be of great interest.

  40. Thanks for the explanation, Dr. Spencer.
    I’ll set to work on how to present the new V6.0 in my pages.
    This is great, we now have better agreement between satellite-based LT temperature data bases.

    • dave says:

      “…better agreement…”

      There was never much of a disagreement between RSS and UAH as regards the mid-troposphere. RSS has a trend of +0.077 C per decade and UAH V. 5.6 a trend of +0.05 C per decade and UAH V. 6.0 a trend of +0.069 C per decade. The mid-troposphere is barely affected by the diurnal cycle and so that pesky adjustment for drift is less crtical.

  41. Climanrecon says:

    Is there a simple relationship between what is being measured here and what gets measured by surface thermometers? I know there is a big difference between the surface and the troposphere, I’m looking more for what is the closest surface data quantity, for example is it the simple average of the surface MAX and MIN temperatures, or a more complicated function of MAX and MIN?

    • Mike M. says:


      Since high daytime T promotes convective mixing, surface max tends to be representative of the entire boundary layer, allowing for adiabatic lapse rate. Low nighttime T promotes stratification, so nighttime min tends to be representative of only a very thin layer near the surface.

      So I think surface max would tend to follow the satellite data better than the average of max and min. That might be part of the reason the satellite trend is smaller than the surface trend.

  42. Roy
    You still calculate trends linearly over the full data set. This really misrepresents what has happened and more importantly for climate and energy policy what is likely to happen.
    It is of interest that the trends in the UAH v6 are much closer to the RSS data,. In particular they confirm the RSS global cooling trend since 2003 when the natural millennial solar activity driven temperature cycle peaked.
    It is the satellite data sets which should be used in climate discussions because the land and sea based data sets have been altered and manipulated so much over the years in order to make them conform better with the model based CAGW agenda.
    The IPCC climate models are built without regard to the natural 60 and more importantly this 1000 year periodicity so obvious in the temperature record. This approach is simply a scientific disaster and lacks even average commonsense .It is exactly like taking the temperature trend from say Feb – July and projecting it ahead linearly for 20 years or so. The models are back tuned for less than 100 years when the relevant time scale is millennial. This is scientific malfeasance on a grand scale.
    The temperature projections of the IPCC – Met office models and all the impact studies which derive from them have no solid foundation in empirical science being derived from inherently useless and specifically structurally flawed models. They provide no basis for the discussion of future climate trends and represent an enormous waste of time and money. As a foundation for Governmental climate and energy policy their forecasts are already seen to be grossly in error and are therefore worse than useless.
    A new forecasting paradigm urgently needs to be adopted and publicized ahead of the Paris meeting.
    For forecasts of the timing and extent of the coming cooling based on the natural solar activity cycles – most importantly the millennial cycle – and using the neutron count and 10Be record as the most useful proxy for solar activity check my blog-post at
    The most important factor in climate forecasting is where earth is in regard to this quasi- millennial natural solar activity cycle which has a period in the 960 – 1020 year range. For evidence of this cycle see Figs 5-9. From Fig 9 it is obvious that the earth is just approaching ,just at or just past a peak in the millennial cycle. I suggest that more likely than not the general trends from 1000- 2000 seen in Fig 9 will likely repeat from 2000-3000 with the depths of the next LIA at about 2650. The best proxy for solar activity is the neutron monitor the count and 10 Be data. My view ,based on the Oulu neutron count – Fig 14 is that the solar activity millennial maximum peaked in Cycle 22 in about 1991. There is a varying lag between the change in the in solar activity and the change in the different temperature metrics. There is a 12 year delay between the neutron peak and the probable millennial cyclic temperature peak seen in the RSS data in 2003.
    There has been a declining temperature trend since then (Usually mis-interpreted as a “pause”) There is likely to be a steepening of the cooling trend in 2017- 2018 corresponding to the very important Ap index break below all recent base values in 2005-6. Fig 13. The Polar excursions of the last few winters are harbingers of even more extreme winters to come more frequently in the near future.

    • David A says:

      “The IPCC climate models are built without regard to the natural 60 and more importantly this 1000 year periodicity so obvious in the temperature record.”

      What are the physical causes of these cycles? A 60-yr cycle is plausible, from the cyclic AMO and PDO.

      A 1000-year cycle? Based on what physics?

      • For evidence of the 1000 year +/- cycle see figs 5-9 at

        It is not necessary to understand the physics to make perfectly useful forecasts. However the post says

        “The solar/astronomical basis of this persistent millennial temperature and climate cycle is discussed in detail and illustrated in several of Scafettas publications e.g. Fig 12 at:

        Fig7.” see the Fig 7

        As to the mechanisms. The post says

        NOTE!! The connection between solar “activity” and climate is poorly understood and highly controversial. Solar “activity” encompasses changes in solar magnetic field strength, IMF, CRF, TSI, EUV, solar wind density and velocity, CMEs, proton events etc. The idea of using the neutron count and the 10Be record as the most useful proxy for changing solar activity and temperature forecasting is agnostic as to the physical mechanisms involved.

        Having said that, however, it is reasonable to suggest that the three main solar activity related climate drivers are:
        a) the changing GCR flux – via the changes in cloud cover and natural aerosols (optical depth)

        b) the changing EUV radiation – top down effects via the Ozone layer
        c) the changing TSI – especially on millennial and centennial scales.

        The effect on climate of the combination of these solar drivers will vary non-linearly depending on the particular phases of the eccentricity, obliquity and precession orbital cycles at any particular time.

        Of particular interest is whether the perihelion of the precession falls in the northern or southern summer at times of higher or lower obliquity.”

        • David A says:

          It’s a very long blog post (tl;dr) with a million figures. What am I supposed to look at?

          BTW, the MWP wasn’t global. So that’s one less data point right there.

          “Continental-scale temperature variability during the past two millennia,” PAGES 2k Consortium, Nature Geosciences, April 21, 2013

          • David Read section 2.3 and look at the Figures. For the timing and extent of the MWP and LIA see Fig 9.
            For the relative amplitude of the millennial cycle in the NH and globally see

          • David A says:

            Neither the MWP/MCA or LIA were global (PAGES 2k). So how can they say anything about global cycles or global warmign?

          • David A says:

            Your cited paper by Shindell et al is about regional climate change, not global climate change.

          • David
            Here is the abstract of the Shindell paper,
            “We examine the climate response to solar irradiance changes between the late 17th-century Maunder Minimum and the late 18th century. Global average temperature changes are small (about 0.3 degrees to 0.4 degrees C) in both a climate model and empirical reconstructions. However, regional temperature changes are quite large. In the model, these occur primarily through a forced shift toward the low index state of the Arctic Oscillation/North Atlantic Oscillation as solar irradiance decreases. This leads to colder temperatures over the Northern Hemisphere continents, especially in winter (1 degrees to 2 degrees C), in agreement with historical records and proxy data for surface temperatures.”

            The “region” they refer to is actually the NH. This contains most of the earths landmass, population and food production. The amplitude here is 1- 2 dgrees. They do not say that there is no glogal effect only that it is smaller. This is because of the thermal inertia of the vast southern oceans.

          • Sun Spot says:

            @David A says: , Dave please give us a theory on the meter logical structures that would have to exist to have a local warm spot (about 3 degrees warmer that today) hang over Europe for about 100 years and not effect the rest of the planet. The MWP was global, as many studies from other parts of the world have confirmed.

          • All those events were local because that is the alarmist talking point.

          • Sun Spot,

            “Dave please give us a theory on the meter logical structures that would have to exist to have a local warm spot (about 3 degrees warmer that today) hang over Europe for about 100 years and not effect the rest of the planet.”

            The surface area of all Europa makes only about 2% of the surface area of the whole planet. Accordingly, a temperature anomaly of 3 Kelvin in all Europe would mathematically increase the globally averaged temperature by merely 0.06 Kelvin. The effect stays the same for the time for which this European anomaly lasts. The effect doesn’t increase with time.

            For comparison, the globally averaged surface temperature anomaly relative to the base period 1951-1980 amounts to about 0.7 Kelvin nowadays, i.e., it has become already more than 10 times as large, compared to the effect, which the European anomaly of 3 K would have had on the global average.

            “The MWP was global, as many studies from other parts of the world have confirmed.”

            You need data that sufficiently cover the whole planet to draw such a conclusion. Please name some of the alleged many studies from other parts of the world which have provided that information and which show that there was a MWP, during which it was supposedly substantially warmer than today as a global phenomenon. Those studies would be in contradiction to the study cited by David A.

        • CoRev says:

          David A, repeating the same erroneous argument does not improve the error. “BTW, the MWP wasn’t global.” Are you claiming they had no effect on global climate?

          Let’s try this counter argument. The glaciations were not global! THEY DID EFFECT GLOBAL CLIMATE.

          • CoRev,

            The changes between glacials and interglacials were global climate changes. The globally averaged surface temperature between the two different climate states changed by 4-7 K. There isn’t any logic in the conclusion that MWP and LIA must have been global phenomena, because the change from interglacials to glacials and back had been such. It’s a non-sequitur.

  43. David A says:

    Am I right that there are some huge changes in the new dataset, compared to v5.6?

    For the LT:

    A change of +0.83 C for USA48 in 1/1979
    A change of +0.85 C for USA49 in 2/1979
    A change of -0.76 C for AUST in 12/1978
    A change of -1.31 C for SoPol_Land in 9/1982

    Even some recent changes are big:

    A change of -0.31 C for NH_Ocean in 12/2014
    A change of -1.40 C for NoPol_Ocean in 3/2010, and -1.25 C for 2/2010
    A change of -0.71 C for USA48 in 2/2009


    • David A says:

      No responses?? If these kind of changes were made to GISS or HadCRUT4, every single person here would be fuming for months and calling for scalps. Don’t even try to deny it.

      Science needs skeptics, not sycophants. No matter who’s behind the numbers.

    • Philip Shehan says:

      As a resident of Australia it is indeed interesting that local skeptics are cheering this developement and contrasting it to the Australian Bureau of Meteorology adjustments which they claim falsely reduce historical temperatures based on the large changes at a few named sites.

      Of course they ignore the map of the entire continent which shows that show that the warming trend across the whole of Australia looks bigger when you don’t homogenise the data than when you do.

      I wonder what the BOM’s critics would say if they actually read Dr Spencer’s article, rather than the interpretation of it presented on Watts’ and Andrew Bolt’s blogs.

      “For example, going from Version 5.6 to 6.0 the Australia trend increased from +0.17 to +0.24 C/decade”

  44. Ken Gregory says:

    The UAH trend from January 2002 to March 2015 decreased from +0.057 C/decade (ver.5.6) to -0.029 C/decade (ver.6.0), or a change of -0.086 C/decade. The RSS trend for the same period is -0.049 C/decade. The average of the two trends is -0.039 C/decade.

    2002 is the about 10 years after the peak of solar magnetic flux, which is when the temperature response to the century long increase of magnetic flux peaks. The temperature response is expected to be delayed about 10 years due to the large heat capacity of the oceans.

    • Ken
      See my 4/29/10:18 AM post above.
      12 years lag from neutron count low (activity high)1991 to RSS temperature millennial peak seen at about 2003.
      All downhill from here to depths of next LIA about 2650.

      • Ken Gregory says:

        I did read your excellent post, and I went to your website from the link you provided. You say 12 year lag, I used a 10 year lag, close enough. I agree that 1000 year cycles are very important, however, the “all downhill from here” may be offset by some AGW, which is good news! I think Transient Climate Response is about 0.6 C to doubling CO2, but reasonable people can have greatly different opinions on this. Dr. Spencer thinks that long term ocean cycles can be caused by nothing except chaos, where as I think they are linked to solar changes, with chaos messing thing up ….

    • Philip Shehan says:

      Although as I write elsewhere that the current slowdown in temperature increase utterly fails the statistical significance test, I am persuaded that the recent studies providing evidence for the 60 year decadal oscillations adequately accounts for the small difference between the current models and the data.

  45. David A says:

    What error bars or confidence limits come with the new monthly anomaly numbers?

    And, sometimes you use 2 significant places for your data and results, and sometimes you use 3. Which is scientifically accurate???

  46. Go Whitecaps!! says:

    It appears that the anomaly for USA48 is approximately 1.75C/decade. This is close to Anthony Watts’s 1.55C/decade.

  47. Lance Wallace says:


    Following the link provided by Bob Tisdale in WUWT

    I calculated the value for the 1981-2010 reference period and it was between -0.020 and -0.022 for all entries (global, tropics, etc.) Doing the same for v. 5.6 results in 0 to within 10^-5. Can you check and correct if need be?

    I left this message for Bob at WUWT.

    • David A says:

      Yes, I get the same results.

      In v6.0beta, the 1981-2010 averages all come out to about -0.021 C, for every region.

    • yes, we are fixing this now, thanks for the heads up. It’s because our MSU3/AMSU7 calibration adjustments were done AFTER the anomaly computations, and we then neglected to the remove the residual annual cycle that resulted.

      …thus the reason wht this was a “beta” release. 😉

  48. MikeN says:

    Didn’t you previously say that RSS is running cool? I’ve seen a reference to ‘spurious cooling’. Do you now think they are doing a fine job, or perhaps that UAH is running cool as well?

    • fonzarelli says:

      Mike, glad to see your “reality check” comment here… We all seemed to have such great faith in 5.6 which turned out to be quite wrong. Should we have a greater faith in 6.0? (it’s kind of like lucy holding the football for charlie brown all over again) One thing we don’t have is the uah corroboration with balloon data any more. THAT argument was oft the basis for dr. spencer’s claim of uah superiority over rss. (i might also add that uah no longer tracks so well with carbon growth) So it looks, now, as though we could be flying blind with both of these satellite data sets. And if they both happen to be wrong? Let me end by quoting the t.v. fonzie’s grandma nussbaum: “two wrongs a right don’t make (honey…)”

      • David A says:

        Is UAH v6.0 going to be submitted for peer review?

        • crakar24 says:

          Yep it sure will, however there is a bit of a backlog ATM with all those GCM peer reviews getting stuck.

          • David A says:

            I doubt you speak for UAH…. It’s a legitimate question — I think we deserve to know the answer, considering the issues with UAH results in the past.

          • mpainter says:

            That issue has been satisfactorily resolved. Feeling a wee bit strident this evening?

          • David A says:

            How has peer review been “resolved?”

        • MarkB says:

          David A,
          In the first paragraph of section 2 of the article, it says that a paper will be submitted to a journal, though it will likely be some time before it makes its way through to publication.

        • sigh….

          “Please, before you ask a question, read the following to see if your question has already been answered.”

          “It will likely be close to two years before a peer reviewed paper with greater detail gets published in a scientific journal.”

      • crakar24 says:

        I look at this another way……..Dr Spencer has stated his methodology is completely different to RSS (RSS use models to calculate adjustments etc)whereas UAH dont.

        Therefore we have two completely approaches to calculating the temps and yet we get a similar result, this if anything should give up more confidence.

      • dave says:

        fonzarelli says:

        “The one thing we don’t have is the uah corroboration with balloon data anymore.”

        It is the dotted line on Figure 7.

        • fonzarelli says:

          Here you go Dave,

          This is from a post earlier this month. If uah is now trending like rss, then we no longer have the corroboration of uah with balloon data…

          • DZ says:

            That is a non-sequitur. Just because some trends changed, does not mean the very specific data points do not match with balloon data. Not to mention the changes are not as great as you are making them out to be. If you had taken the time to look into things before making spurious statements you would realize that still yet UAH even with 6.0 revision matches balloon data better than RSS dataset, which highlights still the differences between the models, in specificity.

          • fonzarelli says:

            “if you had taken the time…”

            DZ, you got that right! This was the quickest thing i could find as dr spencer posted it just a few weeks ago. Admittedly this is not an ideal graph to use in response to dave. (AND, i’m still looking!) Much of the change actually would seem to improve the corroboration with the balloon data. Toward the end though, the least few years, it would seem that uah has a divergence with the balloon data. How that shows up on plotted graphs of the two data sets is what i’d like to see. It could be as much as .1 degrees and if that grows with time it makes one pause…

            This probably would have been a better graph to give in response to crakar24. As even with the adjustments the two trendlines of uah and rss would yet look very different. At any rate, thanx for the “stinging rebuke”. I actually feared a much worse response from some one. (i regret the comment…) Your terse and objective reply was spot on…

      • Kristian says:


        This whole fixation on ‘carbon growth’ vs. satellite temp anomalies appears to be yours alone. Could you please show us why (old) UAH was so much ‘better’ at tracking the ‘carbon growth’ than RSS? And why there needs to be such a tight fit – trendwise – in the first place?

        Also, it was very clear in the end that UAH ran way too high post 2005 because of a significant and spurious rise in its land portion relative to all other comparable datasets, and notably, relative to its own oceanic portion. RSS was much more internally consistent in this regard.

        RSS also matches the ‘outgoing LW’ curve of the CERES EBAF-ToA product in quite impressive fashion:

        UAH didn’t. But now it does. A definite improvement.

        As you can see, UAH v6.0 still in general tracks ever so slightly higher than RSS after 2005 (~0.03K):

        That final correction is most likely RSS’s to make.

        • fonzarelli says:

          Kristian, firstly i’d like to say kudos to both you and phi for pointing this out to dr spencer a while back (that the land data ran high post 2005…). It has not been clear whether you both actually influenced dr spencer or if he was already aware that there was a problem. So, you may have made history here! (shouldn’t you and phi get PAID if that’s the case?!) I wanted to tell you this, kristian, up top (in reply to your ‘thanks’ to dr spencer) but realized that i’d said an ugly thing or two to you way back when and you might not want my praise. I hope you accept my praise and apology as well… (some times i take my ‘tee shirt and leather jacket’ schtick too far)

          As far as the carbon growth argument goes, it’s a fairly well known argument and i’m just parroting that argument. By doing so, i’m hoping to undermine the notion that nature is consistently taking out 50% of anthropogenic carbon. The 50% paradigm has been a stubbornly pervasive one notably (to me) here at this blog. Not only with the commenters, but also with dr spencer himself. After a year or so of dogging anybody who made mention of it, i’ve finally managed to wipe it out. It only comes up rarely…

          The ‘old’ uah was a better fit simply because it WAS a better fit… Carbon growth has been high in resent years and so also was uah. There was also corroboration with balloon data AND (hadley) southern hemisphere land data. So that made four very different data sets running lock step with one another. You ask “why there needs to be such a tight fit- trend wise- in the first place?”. Perhaps it doesn’t… But, it didn’t hurt to have four different data sets telling us the same thing. At any rate, carbon growth still tracks with temps rather than human emissions. That’s the main point i’ve been making. Truth be told, i found your (and phi’s) revelation disturbing enough (along with dr. spencer showing some doubt as well) that i’d pretty much dropped using uah and have been opting for hadley southern hemisphere data to make my point of late. It has the added benefit of going all the way back to 1959…

  49. Jason says:

    Roy me boy, didn’t you get the memo? Recent temps are only meant to be revised upwards.

    • fonzarelli says:

      Jason, back in the day they bashed Dr. Roy for having data set that ran too cool that WAS revised upwards. Do you suppose they’ll be any more charitable to him this time round?

      • nigel says:

        “…this time round…”

        It is less variable; that is the main thing.

        Apart from that, the global LT figures have been moved a couple of tenths of a degree lower, for the last five years. Other global data series are hardly affected.

        A new Edition, not a new book.

        • dave says:

          “A new Edition, not a new book.”

          Better regional resolution – which, however, does not affect the global picture.

          Discounting of land surface emissivity effect – which is a change of interpretation rather than new data.

          A better adjustment for diurnal drift – which is the main thing bringing results closer to RSS over the last few years.

          Dr Spencer always gave the error bars from V.5.6, for the ‘global warming trend,’ as “+/- 0.04 C per decade”. As he points out, the adjustments made by V 6.0 lie within that interval. He has not given new bars. After the ‘beta’ period, that will be adressed, I hope.

        • Carrick says:

          I agree with dave, the improved regional-scale resolution is a “big thing”.

          I think the diurnal drift correction is going to need scrutinizing.

    • What a stupid comment.

      • mpainter says:

        I believe that Jason was being sarcastic. As I read it, his comment was directed at the keepers of the thermometer datasets who notoriously keep adjusting their data upward but never (almost) downward.

        • I understood that it was sarcastically meant. And it is a stupid comment.

          Your assertion is a lie.

          • mpainter says:

            This is science; do try to keep it a civil, if you can. But perhaps you can’t.
            How so is the assertion as you claim, pray tell?

          • You are confused. There wasn’t any science in the “sarcastic” comment. Nor was any in your assertion about what the scientists were allegedly doing who are the “keepers of the thermometer datasets”. It was an attempt of smearing these scientists, instead. You don’t have any moral ground for demanding “civility”.

          • mpainter says:

            There is plenty of reason to assert that there is manipulation in the thermometer data sets.
            You unconscionably have fouled the thread with your rudeness.
            That aside, do not you agree that it is a good thing to have satellite temperature data to act as a balance against over manipulation of thermometer datasets?
            Also, are you not glad that UAH 6 confirms the duration of the “pause”? And thus it underscores the hollowness of the climate alarmism, does it not?
            Think Happy Thoughts.

          • mpainter,

            “There is plenty of reason to assert that there is manipulation in the thermometer data sets.”

            What reasons would that be? Your own biased perception, according to which results from analyses of these datasets must be wrong, when they don’t confirm the premise in your thinking of a “hollowness of the climate alarmism”, for whatever non-scientific reasons (economical? political? ideological? religious?) you uphold this premise? Projection of your own approach to these things onto the scientists who do their work?

            “You unconscionably have fouled the thread with your rudeness.”

            I suppose then, smearing scientists who publish results from research, which you don’t like doesn’t constitute “rudeness” in your circles, whereas being called out on it does.

            “That aside, do not you agree that it is a good thing to have satellite temperature data to act as a balance against over manipulation of thermometer datasets?”

            Loaded question. If I answered it I would accept the premise of the question, which is just a repetition of the smearing.

            The usefulness of the satellite based data doesn’t arise from “balancing” against alleged “over manipulation” of the surface temperature data. Instead, they help to get a more complete picture. The satellite based temperature datasets don’t measure the same thing as the surface temperature datasets. And there isn’t any contradiction between what the surface and satellite data say.

            “Also, are you not glad that UAH 6 confirms the duration of the “pause”? And thus it underscores the hollowness of the climate alarmism, does it not?”

            How do the satellite data confirm an alleged “pause”? Something for which there isn’t any proper scientific basis? It starts with the lack of scientific definition, what this “pause” is supposed to be. A “pause” from what? It continues with the lack of any metric, based on which it is in a statistically testable way established that there actually was a “pause”. Nor is it explained why the satellite data are supposedly the ones that were relevant, compared to other data, to draw scientific conclusions about this.

          • mpainter says:

            Much better. I note that you deny the “pause”, so called and also known as the “hiatus” and by other appellations. Many of your persuasion would disagree, there being scores of papers referring to or otherwise discussing the termination of the warming trend of the lower troposphere; written by scientists that you would never call a skeptic, yourself. Even your idol Hansen has written one.You are extreme in your views that there has been no pause.

            Of course, there are those who, like yourself, cannot bear to acknowledge that the warming has stopped. They use terms like “slowing of the warming”. This amuses. My view is that this sort of denial reveals the cult aspects of this climate farce.

            Why should satellite data be preferred, you ask? Because it gives a vastly more complete coverage of the planet, missing only from 85 to 90° at both poles. There are other reasons for preferring satellite data. It can be perfected, at least theoretically, while the temperature data sets are polluted with dubious data and can never be improved.

            Happy thought: that Roy and his team have taken a big step toward achieving this perfection.

          • mpainter,

            “Much better. I note that you deny the ‘pause’, so called and also known as the “hiatus” and by other appellations. Many of your persuasion would disagree, there being scores of papers referring to or otherwise discussing the termination of the warming trend of the lower troposphere;”

            So, it shouldn’t be a problem for you to provide some references as examples from the alleged “scores of papers” that come to the conclusion that there was a “termination of the warming trend” x years ago or that global warming “has stopped” for x years. Although it is true that there are papers, where the term “hiatus”/”pause” is used as a work term, often in quotation marks, it doesn’t mean that they actually support your assertions/conclusions that there was a “stop” of global warming.

            “Even your idol Hansen has written one.”

            Are you claiming Jim Hansen had stated somewhere global warming had stopped? If you do you are making things up.

            “Of course, there are those who, like yourself, cannot bear to acknowledge that the warming has stopped.”

            So tell me, purely from the diagnostic side, not even talking about causalities and physics, on what statistical evidence is your claim based, according to which global warming had stopped? All you have delivered so far is argumentum ad populum and argumentum ad hominem targeting my person. Is this your “science” which you had mentioned before?

            “Why should satellite data be preferred, you ask? Because it gives a vastly more complete coverage of the planet, missing only from 85 to 90° at both poles.”

            Again, they do not measure the same thing. They do not have a “more complete coverage” than surface temperatures. The satellite temperature datasets are for the free atmosphere. They don’t tell you about surface warming. They also don’t measure what is happening with the ice at both pole caps, or what is happening with the oceans. You are not arguing for more complete information, you are arguing for ignoring information. You present the view that only satellite temperature data should be considered to draw conclusions. It seems to me that you only want to consider data that you perceive as supportive for your preconceived views (which is a misperception based on cherry picking and ignorance of statistics, anyway).

            With respect to physics, what is your argument for the view that the tropospheric temperature trend is the most important metric for conclusions whether global warming has stopped or not, but not other metrics, like the ocean warming trend?

            Regarding heat accumulation, the atmosphere is only an energetic appendix, compared to the oceans, because of the very small heat capacity of the atmospheric mass, compared to the one of the oceans. There isn’t any substantial argument from physics for claims that global warming “had stopped” as long as the oceans are heating up. And now, please give me some names and papers from the “scores” of scientists, about which you claim that they supported your assertions, who allegedly say something else than I do. Since you display it as if my views were a minority view on the “extreme” side. They aren’t.

          • mpainter says:

            Jan Perlwitz.
            Scores if papers have been written to try to account for the pause, or hiatus, using those or similar terms. Where have you been?

            I do not pay much attention to them. One of the most recent accounted for it through volcanic aerosols for last 15 years. Ben Santer was a co-author. There many more such papers. I am astounded that you are unaware of these. Search the internet.
            These authors might not agree with me on the specifics of the “pause”, but they do not deny it but rather attempt to explain it.
            Again, I refer you to the internet. It is all there.

            I’ll reiterate: I believe that satellite data is more reliable than thermometer data for the purpose of devising temperature indices, for the reasons I gave above.
            Regarding your inventions concerning my acts, thoughts, and intentions, which inventions you have attributed to me, no thanks, I will pass. Tar babies have no appeal for me. Best wishes.

          • Let’s summarize mpainter’s scientific/statistical arguments in support for his assertions about the alleged “hiatus”/”pause” and what he has delivered to back these arguments up as allegedly supported by “scores” of scientific papers: Zilch

          • mpainter says:

            See my comments above.

  50. Robert says:

    When you say an increased spatial resolution does this mean that the product will be provided at that higher resolution or will it be aggregated to the same resolution as the prior version?

    Also is there a netcdf ‘style’ gridded dataset that is downloadable?

    • The latter…still reported at 2.5 deg lat/lon resolution, but the new product is much less smoothed than the old product.

      We don’t do NetCDF. 😉 But the grids are available as ASCII files.

      • JRChristy says:

        Actually, for v5.6 we do provide NetCDF files to NOAA/NCEI (formerly NCDC). Whenever v6.0 is finalized, we will be required to complete all of the ATDB NOAA documentation (a major chore) provide the code, and then we will generate NedCDF files. In truth, the community prefers ASCII (txt) because what you see is what you get.

  51. nigel says:

    There is one new product – the tropopause data series. That is flat over the last thirty-six years (-0.02 C per decade).

  52. Rob JM says:

    Thank you and your associates for all your hard work!

    By the way the lower stratosphere appears to show multidecadal effect from volcanic eruptions, with a flat trend outside the boom and bust effect of the large SO2 injections for el chichon and pinatubo. The injected SO2 that is eventually rained out should result in a loss of H20 from the stratosphere that presumably cannot be replenished due to the inversion of the stratosphere.
    The long term stratospheric cooling due to volcanic SO2 injection has not even been considered for its effect on global climate.

    It should be noted that the volcanic SO2 injection directly proceeds the large changes in cloud cover shown in the international satellite cloud climatology project what directly proceed the changes seen in global temp.

    Food for thought!

    • Mike M. says:


      “The injected SO2 that is eventually rained out should result in a loss of H20 from the stratosphere”

      Have you tried to run any numbers? Aren’t the SO2 mixing ratios a few ppb whereas water vapor is a few ppm? And the particles are something like 70% H2SO4 by weight, so there are only a few moles of H2O removed for each mole of S.

      “that presumably cannot be replenished due to the inversion of the stratosphere”

      There is some water vapor transport across the tropopause. And methane oxidation also provides a source of water vapor to the stratosphere.

  53. ehak says:

    Different weighting functions means version 6 is not the same product as 5.6 and before. More stratospheric cooling and less groundlevel warming in this new dataset. You must give this 6’er a new name. Not TLT. Perhaps TMLT.

    • did you read the analysis of what effect this has on the results?

      We are hopeful that the old LT methodology will eventually be abandoned by others. It has outlived its usefulness. (And I’m the one who invented it.)

      • ehak says:

        Seriously Spencer, is 6beta the same product ac 5.6? The same atmospheric layers.

        Not. Call it something else. And tell people not to use the 6beta to say anything about the validity of groundlevel measurements.

  54. Jeff Id says:

    Mike M,

    I’ve been around this information for quite a while now and am aware of the adjustments and why they exist. Hell, I’ve even been misquoted in two psychology papers for the fact.

    Here is a link to a global temperature calculation, done myself with the adjusted ground data some years ago.

  55. Scott Scarborough says:

    Won’t this make the comparison between the average of the two satellite data sets and the balloon data sets show a greater difference? The plot of the average of the two satellite data sets and the ballon data sets (I don’t remember how many there were) looked very impressive and close. It seeems to me that this will make them further apart.

    • from what radiosonde comparisons I’ve seen from John Christy, the results changed very little. Remember, those comparisons are affected by variations on ALL time scales…we think we now have better month-to-month variability than before, which improves the agreement with sondes, but now that we are closer to RSS, the low-frequency agreement with sondes decreased slightly. The two effects seem to largely cancel.

      As I’ve told John C., don’t assume sondes are truth…they have had their own changes in instrumentation and software over the years.

  56. Roy,

    I find it encouraging, if the satellite based temperature datasets are in better agreement with each other now.

    Having said this, I take from your post that you have done some major changes to the methodology how you derive the temperature data from the satellite measurements, compared to previous versions. And your announcement in your post that it will likely take close two years before a paper about this will be published, is vague. Is this your current plan? How far are you from actually submitting the manuscript? I find it a bit problematic to release a dataset that is based on methodology whose details and scientific rational hasn’t gone through any peer review and isn’t documented anywhere, particularly if the data like the ones provided by you are of such importance for other scientists for analyzing tropospheric and stratospheric temperature variability and for climate model evaluation. How can we use the new data with any confidence in their reliability as long as that is the case?

    For instance, a vague statement like this,

    “We have performed some calculations of the sensitivity of the final product to various assumptions in the processing, and find it to be fairly robust. Most importantly, through sensitivity experiments we find it is difficult to obtain a global LT trend substantially greater than +0.114 C/decade without making assumptions that cannot be easily justified.”

    doesn’t really provide any usable information about the range of the uncertainty in the trend of +0.114 C/decade, which comes from uncertainties in your methodology. Can you provide any quantification of the uncertainty range? Thanks.

    • I understand your concerns. The new dataset can be taken with a grain of salt until it is published. Just keep in mind we have spent more time examining different ways to analyze the satellite data than anyone on the planet. It’s easy to have an idea like, “Oh, I think the data should be analyzed this way…”, and you WILL get an answer. But determining whether you are just fooling yourself takes a lot more effort…effort most who have tried this have not gone through. It’s way easier to be wrong in this business than to be right.

      • I agree about the effort that is needed to do this kind of work, and I can very much relate to your sentiment about people telling you how you should do things differently, particularly if it comes from people who haven’t invested as much effort, sometimes not even a small fraction of it. This is true not only for the work you do.

        As for the uncertainty. RSS have done 400 Monte-Carlo simulations do provide an estimate about the uncertainty due to measurement errors and decisions about the methodology (not to be confused with the uncertainty from natural variability in the temperature data sets). They provide the gridded data for all the realizations on their website. For the global average, the 5-95% range of all the realizations for the retrieved TLT trend (up to Dec 2012) is 0.075 to 0.19 C/decade.
        (, based on
        Mears, C. A., F. J. Wentz, P. Thorne and D. Bernie, (2011) Assessing Uncertainty in Estimates of Atmospheric Temperature Changes From MSU and AMSU Using a Monte-Carlo Estimation Technique, J. Geophys. Res., 116, D08112, doi:10.1029/2010JD014954)

        This 5-95% range is quite substantial. The uncertainty range will be even larger for regional data or shorter time intervals, then. Some people draw conclusions from the RSS trends over short time intervals, anyhow, w/o taking into consideration any of this.

        Are you planning to assess the uncertainty in your retrievals in a similar way? It would be great if you provided this kind of information to the users, too. It would be very useful, e.g. for comparing climate model output to the satellite data, to have data on the uncertainty in your satellite retrievals.

  57. Carrick says:

    Roy Spencer, could you show a version of Fig.4 where you subtract the version 5.6 trends from the version 6.0, so we can see the geographical distribution of the corrections to LT?

    Thank you by the way for taking to time to write up this detailed description of your methodology. It’s great to have descriptions of the main components of your analysis framework all in one place.

  58. Bob Matulis says:

    Very thorough and very impressive. It is amazing how something as seemingly simple as tracking sensible temperatures can be so complicated! (whether with surface sensors or satellite)

  59. Doug Cotton says:

    I can’t stress enough Roy that your estimate of the rate of increase is about double reality. The world has not warmed at that rate ever since the Little Ice Age, and nor did it between the Dark Ages and the Medieval Warming Period. The long-term rate is about 0.05 degree/decade.

    To calculate the long-term trend you have to eliminate the effect of the superimposed 60 year cycle, as was done in the graph linked in my first comment on this thread.

    All climate change is entirely natural and has nothing to do with carbon dioxide. The news is out that about half of the youth in the US are skeptical. That’s progress!

    To save me writing it all again, I’ll just copy an email still in my clipboard that I just sent to the Leader of the Opposition in Australia, Bill Shorten, and about 150 politicians and a dozen or so international scientists …

    Dear Bill

    I write as one who has published two peer-reviewed physics papers and a book all on the climate debate which I have studied extensively.

    I can prove beyond a shadow of a doubt (using correct physics) that all that the 0.04% of carbon dioxide in Earth’s atmosphere can do is to cool (not warm) the surface, but by less than 0.1 degree.

    I have personally offered a $5,000 REWARD for anyone who can prove the physics I present to be substantially wrong and to produce a study similar to mine but showing that the most prolific “greenhouse gas” water vapour warms, rather than cools as I have shown. Rain forests are not 40 degrees or more hotter than dry deserts.

    My question is …

    Are you, Bill Shorten, prepared to give me a personal hearing (somewhere in Sydney where I live, or Canberra) in the presence of other persons suitably qualified in physics but not having a pecuniary interest in maintaining the world’s greatest ever fraudulent hoax which is all based on fictitious fiddled physics?


    Doug Cotton

  60. Doug Cotton says:

    Roy (and others)

    You may well have got an “A” in some climatology course on thermodynamics, but you do not understand the Second Law of Thermodynamics, Roy, whereas I have written two papers and a book about it and its relevance to climate.

    Until you, or any reader, understands precisely how physicists explain (using the Second Law and Kinetic Theory) precisely why the density gradient forms and is stable because it is the state of thermodynamic equilibrium then you will never understand how the same process forms a stable temperature gradient when entropy is maximized. The importance of this is HUGE in respect of the climate debate, because it demolishes the greenhouse.

    And now we have, in this 21st Century, empirical proof in centrifugal force experiments that what I am saying here is correct, Roy.

  61. Doug Cotton says:

    Roy and others:

    Please read Comment #1500 on that previous thread here.

  62. Ric Werme says:

    Bookkeeping oops:

    At the bottom of the directory listing shows:

    uahncdc_ls_5.6 06-Mar-2015 11:47 73K
    uahncdc_ls_5.6.txt 06-Mar-2015 11:47 73K
    uahndcd_ls_5.6 10-Apr-2015 06:32 74K
    uahndcd_ls_5.6.txt 10-Apr-2015 06:32 74K

    It appears to me the files with “ndcd” in their names should have overwritten the files with “ncdc” in their names.

    Also, I’m a little confused about URLs for the LT files, is the right file for V5.6 data and the right file for V6.0 data? the t2lt and tlt directories seem to be odd places for those files.

  63. StuartL says:

    Dr Roy. some people on Yahoo answers are saying that UAH 6.0 uses NOAA-15 data. can you confirm or deny please.

  64. JamesG says:

    Have you noticed that you show the Antarctic peninsula as cooling yet that is the one part of Antarctica we absolutely know is warming? Is this an error or does surface warming lead to troposphere cooling, or what?

  65. dave says:

    “…show [wrongly] the Antarctic peninsular as cooling…”

    Perhaps the picture of regional trends can not resolve this skinny finger of land, as a stand-out from the Southern Ocean and its sea-ice.

    Also, being essentially a long mountain range, most of it is so high that the air at local ground level is not even in the Lower Troposphere. Meteorologists routinely “adjust out” orography (when giving pressures for aviation, for instance), but that would not seem appropriate when the idea is to distinguish between differing, small, tendencies in different layers of the atmosphere.

  66. dave says:

    Actually, this has raised a question for Dr Spencer. What IS the definition of Lower Tropospheric? Is it “air close to the ground” (which in the case of the South Pole would be at an altitude of 7,000 feet) or is it “air up to (say) 500 KP level”?