Australia Surface Temperatures Compared to UAH Satellite Data Over the Last 40 Years

Reposted from Dr. Roy Spencer’s Blog

April 3rd, 2019 by Roy W. Spencer, Ph. D.

Summary: The monthly anomalies in Australia-average surface versus satellite deep-layer lower-tropospheric temperatures correlate at 0.70 (with a 0.57 deg. C standard deviation of their difference), increasing to 0.80 correlation (with a 0.48 deg. C standard deviation of their difference) after accounting for precipitation effects on the relationship. The 40-year trends (1979-2019) are similar for the raw anomalies (+0.21 C/decade for Tsfc, +0.18 deg. C for satellite), but if the satellite and rainfall data are used to estimate Tsfc through a regression relationship, the adjusted satellite data then has a reduced trend of +0.15 C/decade. Thus, those who compare the UAH monthly anomalies to the BOM surface temperature anomalies should expect routine disagreements of 0.5 deg. C or more, due to the inherently different nature of surface versus tropospheric temperature measurements.

I often receive questions from Australians about the UAH LT (lower troposphere) temperature anomalies over Australia, as they sometimes differ substantially from the surface temperature data compiled by BOM. As a result, I decided to do a quantitative comparison.

While we expect that the tropospheric and surface temperature variations should be somewhat correlated, there are reasons to expect the correlation to not be high. The surface-troposphere system is not regionally isolated over Australia, as the troposphere can be affected by distant processes. For example, subsidence warming over the continent can be caused by vigorous precipitation systems hundreds or thousands of miles away.

I use our monthly UAH LT anomalies for Australia (available here), and monthly anomalies in average (day+night) surface temperature and rainfall (available from BOM here). All monthly anomalies from BOM have been recomputed to be relative to the 1981-2010 base period to make them comparable to the UAH LT anomalies. The period analyzed here is January 1979 through March 2019.

Results Before Adjustments

A time series comparison between monthly Tsfc and LT anomalies shows warming in both, with a Tsfc warming trend of +0.21 C/decade, and and a satellite LT trend of +0.18 C/decade:

Fig. 1. Australia average surface temperature (red) and satellite lower tropospheric temperature (LT, blue) anomalies from January 1979 through March 2019.

The correlation between the two time series is 0.70, indicating considerable — but not close — agreement between the two measures of temperature. The standard deviation of their difference is 0.57 deg. C, which means that people doing a comparison of UAH and BOM anomalies each month should not be surprised to see 0.6 deg. C differences (or more).

Part of the disagreement comes from rainfall conditions, which can affect the temperature lapse rate in the troposphere. For reference, the following plot shows Australian precipitation anomalies for the same period:

Fig. 2. Australia precipitation anomalies from January 1979 through March 2019.

If we take the data in Fig. 1 and create a scatter plot, but show the months with the 25% highest precipitation anomalies in green and the lowest 25% precipitation in red, we see that drought periods tend to have higher surface temperatures compared to tropospheric temperatures, while the wettest periods tend to have lower surface temperatures compared to the troposphere:

Fig. 3. Scatterplot of the data in Fig. 1, but with color coding of those months with the 25% highest (green) and lowest (red) precipitation departures from average.

A More Apples-to-Apples Comparison

Comparing tropospheric and surface temperatures is a little like comparing apples and oranges. But one interesting thing we can do is to regress the surface temperature data against the tropospheric temperatures plus rainfall data to get equations that provide a “best estimate” of the surface temperatures from tropospheric temperatures and rainfall.

I did this for each of the 12 calendar months separately because it turned out that the precipitation relationship evident in Fig. 3 was only a warm season phenomenon. During the winter months of June, July, and August, the relationship to precipitation had the opposite sign, with excessive precipitation being associated with warmer surface temperature versus the troposphere, and drought conditions associated with cooler surface temperatures than the troposphere (on average).

So, using a different regression relationship for each calendar month (each month having either 40 or 41 years represented), I computed a satellite+rainfall estimate of surface temperature. The resulting “satellite” time series then changes somewhat, and the correlation between them increases from 0.70 to 0.80:

Fig. 4. As in Fig. 1, but now the satellite data are used along with precipitation data to provide a regression estimate of surface temperature.

Now the “satellite-based” trend is lowered to +0.15 C/decade, compared to the observed Tsfc trend of +0.21 C/decade. I will leave it to the reader to decide whether this is a significant difference or not.

To make the differences in Fig. 4 a little easier to see, we can plot the difference time series between the two temperature measures:

Fig. 5. Difference between the two time series shown in Fig. 4.

Now we can see evidence of an enhanced warming trend in the Tsfc data versus the satellite over the most recent 20 years, which amounts to 0.40 deg. C during April 1999 – March 2019. I have no opinion on whether this is some natural fluctuation in the relationship between surface and tropospheric temperatures, problems in the surface data, problems in the satellite data, or some combination of all three.

Conclusions

The UAH tropospheric temperatures and BOM surface temperatures in Australia are correlated, with similar variability (0.70 correlation).
Accounting for anomalous rainfall conditions increases the correlation to 0.80. The Tsfc trends have a slightly greater warming trend than the tropospheric temperatures, but the reasons for this are unclear. Users of the UAH data should expect monthly differences between the UAH and BOM data of 0.6 deg. C or so on a rather routine basis (after correcting for their different 30-year baselines used for anomalies: BOM uses 1961-1990 and UAH uses 1981-2010).

Advertisements

34 thoughts on “Australia Surface Temperatures Compared to UAH Satellite Data Over the Last 40 Years

  1. Is the correlation before or after the BOM adjustments or is it raw temp data?

    because the correlation error might simply be the error due to the adjustments they make and they have made some colossal stuff ups in the last decade or so

    • There are problems with Dr Spencer’s scatter plot regression which may not be doing the best rainfall corrections.

      Dr Spencer. I would once again draw your attention to the problem of regression dilution when dealing with such noisy scatter plots. Both your OLS slopes are visibly too low. This is probably affecting the effectiveness of your attempted corrections.

      I invite you to do the regression with the axes swapped and compare the results. The true best fit probably lies between the two.

      There are various bisectional methods for guessing where between the two results you should take the estimated slope. Geometric mean of the two slopes is one solution. See whether that improved correlation of the adjusted time series.

      More detail here:
      https://climategrog.wordpress.com/2014/03/08/on-inappropriate-use-of-ols/

      here is an example of two slopes using some synthetic data : linear slope + noise.

      https://climategrog.wordpress.com/?attachment_id=853

      The true slope is 3.0, the regression slope comes out a 0.58 ! This is not a pedantic detail.

    • They are really good at messing with the historical records for sure. They went to the same school of data manipulation as NASA/NOAA. The Net Net here is that there is warming. It is detectable using land based and satellite measurements. That said, it is absolutely normal, not unprecedented, and consistent with natural variability. 40 years with the “highest CO2” in past X thousands of years, but not the warmest and not the fastest rate of warming. Its as if they never heard of glacial cycles of our planet.

  2. How do the correlations change if you chart only the surface daily max temperature anomalies instead of the average? What about the daily minimum anomalies.

    My wild guess is that using surface Tmax will give a better correlation to UAH, and Tmin will give a worse correlation.

    • Precisely. I found this in 2015. In fact tropical (northern) rain from the summer monsoon season is by far the major determinant of the difference between satellite anomalies and surface maxima anomalies for all of Australia.

    • “My wild guess is that using surface Tmax will give a better correlation to UAH”
      It’s pretty wild. The thing is, UAH data doesn’t have nearly as much diurnal variation as surface, so comparing with a diurnal surface peak is not apples to oranges, but more like apples to onions.

      • Sorry Nick, he’s right. During daylight hours there is turbulent convectional mixing high into the atmosphere, at night this reduces to almost none. Higher rainfall leads to greater amounts of moisture condensing high up thus warming the atmosphere more relative to surface temperatures. In drier times less moisture condenses high up, so the atmosphere is cooler relative to the surface. The daytime convection is what causes this, thus the difference between Tmax and UAH correlates with rainfall.

  3. I note that the diagonal (-2,-2 to +2,+2), which would represent a perfect correlation, is much steeper than the plotted lines. Is that significant?

    • Mostly regression dilution due to large errors in x axis. A common misuse of OLS. See my reply to the first post.

  4. Is you for real cobber? The idea of measuring the lower troposphere temperature is to get a idea of the temperature as it relates to that area on the globe at that particular temporal point. Not a ‘vigorous’ weather occurence 100’s of thousands of miles away. Wtaf?

    Is it just me, or is this a totally unnecessary variable? If this is justified, i would suggest ALL weather data is nonsense and needs to be replaced with a measuring device that CANNOT be, or need be bastardised.

    • Please forgive the atrocious grammatical errors. My phone is playing ‘lets piss off the human’ 😖

    • 100’s of thousands of miles away.

      ??? Well, Columbus didn’t know how big the planet was, either. :>)

      • Scotty: “I find it hard to believe I’ve traveled millions of miles…”
        McCoy: “…thousands…”
        Scotty: “…thousands of miles for an invited tour…”

        ^¿^

  5. Thank you for the post. I was just mulling over the global temp map and noting my part of southern Australia showing zero anomaly while recently seeing the media declaring it the hottest EVAH! or at least on record. The veracity and usefulness of the BOM “record” seems to be in steady decline.

    • Indeed!
      The tropospheric temperature increase is, per CAGW theory, suppose to be about 33 percent above the surface increase.

      Therefore, only about 2/3rds of the total tropospheric increase ( call that 2/3rds about 1.2 degrees) can be ascribed to CO2 caused surface warming, no matter the actual surface T increase.

      Any recorded surface warming above that is due to something other then
      CO2. It could be UHI, or poor adjustments or… but it is not CO2.

      So the observed CO2 caused surface warming is estimated at 1.2 degrees per century ( two thirds of the observed tropospheric warming)

  6. The UAH spectrum shows strong periodicity around 44.5 months, but where is this periodicity coming from?
    The Earth’s LOD spectrum (apparently related to the outer liquid core internal circulation where the Earth’s magnetic field is generated) has strong 7.5 years periodicity, and its second harmonic just under 4 years which is coincidental with the UAH spectrum peak of about 44.5 months as shown in the two graphs illustrated here
    http://www.vukcevic.co.uk/UAH-spectrum.htm
    Prof. Dr. Andy Jackson’s lecture
    Geomagnetic Secular Variation as a Window on the Dynamics of Earth’s Core
    to the AGU can be found here : https://youtu.be/1SOSmHPTods
    (p.s. there is a view that the 7.5 periodicity is related to the lunar tides)

  7. “The correlation between the two time series is 0.70, indicating considerable — but not close — agreement between the two measures of temperature.”

    To put it another way, a correlation coefficient of 0.70 means that less than half of the variance in the dependent variable can be explained or predicted by the variance in the independent variable.

  8. So, since the LT is warming faster than the surface (according to the theory), can we estimate that the real surface warming is actually in the region of 0.10 to 0.15 C/decade as opposed to the BOM’s imperfectly-sampled 0.21?

  9. If this is showing plots AFTER the data is “normalized”, then I don’t understand what it is supposed to be proving (or demonstrating). My argument is the adjusted data is tampered with, not that it doesn’t match satellite data. All you have to do is wait for them to tamper with the satellite data to get it in line.

    You CANNOT make a valid scientific point until the data is trusted. This requires all adjustments be explained and pass through a spectrum of criticism and refinement before being approved. Until someone convinces me that the UHIE has correctly been accounted for, and that all these adjustments to the past measurements are valid, there is NO POINT in trying to use the tampered data to convince me of anything.

    I am an old experienced computer nerd (all the way back to paper tape) – and if there is one thing I understand it’s “Garbage In – Garbage Out”. Fix the garbage, then we can compare plots.

    • “You CANNOT make a valid scientific point until the data is trusted. This requires all adjustments be explained and pass through a spectrum of criticism and refinement before being approved. ”

      sure ask roy to post his
      A) raw data
      B) all the regressions he does to adjust his data. there are several undocumented( before and after) adjustments.

      • So Mosher, Are you suggesting that UAH is purposefully fudging the temperature data too low so that they want to hide the fact that we are all in danger of a hellhole earth like the alarmists like you want us to believe? That would mean that you are accusing Christy and Spencer of purposefully wanting to commit suicide along with the other 1.4 billion skeptics. It doesnt work the other way because if and when the alarmists are proved wrong, nothing happens except a waste of trillions of $.
        On the other hand, if the alarmist are right the world does come to an end and people like Christy and Spencer will have knowingly caused the world to commit suicide. Anybody that believes that Christy and Spencer would do that should be committed to an insane asylum. Mosher, your logic is beyond belief and just as bad as your belief in CAGW.

        On a side note why stop at the comparison to Australia? Christy and Spencer should publish the correlations with dozens of other regional surface temperatures.

        • Well people want to know the technique, because it is devilishly hard to correct for all the problems. The fact that it appears to have a growing cool bias relative to radiosondes makes people curious, especially given all the orbital drift stuff etc.

          Nothing personal; skeptics prefer transparency and independent reproduction.

    • It’s even worse than you think!™

      I’ve just spent a week in conversation on my blog with four well-educated people regarding significant digits and error and uncertainty propagation. All four were unanimous in stating that significant digit rules were only guidelines and not firm rules, and that it was perfectly OK to average measurements with one decimal point to three decimal point results. They all seemed to miss the point that increasing N reduces the standard error, but it does not improve the mean itself.

      We could have the best data money can buy (I think that’s the USCRN, right?), but if it’s believed that significant figure and error propagation rules are just “guidelines,”

      Here are a few highlights:

      We keep being told that climate folk don’t follow the rules of significant digits. So OK, what are those rules? Where do we find them written down?

      Consider the measurement 193.2475. Assume we make 100 measurements, and get the following results:

      All 100 have the digits 1,9,3.
      Approximately 99 (say, 98 or 99 or 100) of them have the 2.
      The 100 measurements disagree on the digit listed here as “4”, with a majority of them being 3, 4, 5, 6, or 7, but a few being 2 or 1, or 8 or 9.

      Thus we would say that “2” is the least significant digit at alpha = 0.01, and that “4” is not significant.

      In the best-case scenario, where the individual measurement errors are iid and gaussian, the error for the mean would be 0.05/sqrt(3) = 0.03. So the mean would be reported as “18.4 +- 0.03” … slightly more accurate than the individual measurements, but not much.
      But with enough additional measurements, you could get that down to 18.433 … if it was worth the cost of thousands of measurements.

      I’m going to claim that the sig fig rules are a convenient heuristic that’s great for classwork, but that when it comes to cutting edge work, scientists care about detectable signals. These scientists do care deeply about uncertainty, and noise, and limits, but in my experience, they don’t use sig fig rules at all (as far as it is relevant to “my experience”, I received bachelor and masters degrees in a scientific discipline, and a PhD in an interdisciplinary science/engineering program).

      It would be a waste of information for BEST, NOAA, NASA etc. to constrain their reported estimates of global mean temperature to the relatively coarse level of precision provided by the original measurements. They are correct to report these data to a higher level of precision.

  10. About 80% of Aussies live in the capital cities of our 6 states. All are close to the sea. The heat wave history of those capitals is quite different to the overall story here. In most capitals for the last 100 years, heat waves have NOT become longer, hotter or more frequent.
    If you planned for fire services, electrical cooling demand and so on, you would be stupid to use these general figures.It is unforgivable that Australian officials appear not to know this. Geoff

    • My comment above failed to acknowledge the skill and work that went into the realisation of microwaves to measure atmospheric temperatures, data gathering, quality control etc of the UAH figures in the main post. Thank you again, Dr Spencer & Dr Christy.
      I am not so complimentary about the BOM. They have not published anything I have seen about the city/outback temperature differences. I sent them an essay with spreadsheet a few years ago, noting differences from their official position that heat waves WERE getting longer, hotter and more frequent, with no caveat or qualification that this might be of no more than academic interest so far as most Aussies were concerned. Relevant to some of our now small population of farmers, of course, but hardly impacting city folk. The BOM replied that it could not consider submissions that were nor published after peer review. I do not think that science literature needs yet another paper no more elegant than adding up and taking away numbers.

      This is another example of large potential economic harm to the nation when a government body becomes consumed by dogma, regards objectors as stupid (see Climategate) and is not held accountable for the cost of huge mistakes. This is the same syndrome that has, in 15 years, made our national electricity supply expensive, unreliable, complex and a discouragement for investors in new industry. Geoff
      Please see these articles that need updating for the last few years.
      http://www.geoffstuff.com/capitalheatwaves.pdf
      http://www.geoffstuff.com/graphs_sydmelb_heatwaves.pdf

  11. Guys get used to the idea that a major mistake is being made here A force called surface tension is being ignored. The existence of surface tension means that physical heat can not penetrate the surface of water. The radiated energy from the sun can penetrate the surface of water but physical heat no.

  12. “All four were unanimous in stating that significant digit rules were only guidelines and not firm rules, and that it was perfectly OK to average measurements with one decimal point to three decimal point results. They all seemed to miss the point that increasing N reduces the standard error, but it does not improve the mean itself.”

    Aren’t you just falling into the old trap of thinking that the measurement uncertainty in the average of thousands of instruments is the same as that from each individual one? It’s not, because for the average to be off by the full uncertainty range of one instrument, all the thousands of instruments would have to be off in this same direction by the same amount (of course, if the calibration of the accuracy of the original measurement device is off, i.e. they just all read high, you could get results like that).

    Typically this is useless for criticisms of climate science because what the critic wants to do is say the warming trend might not be happening or is significantly unknown. But even the individual measurements have a bias, it is hard to explain that the average is rising over time via measurement error. This would mean that all the instruments are biasing more and more to the high end over time.

    This is an area of stock mathematical errors/fallacies pretty commonly made on climate science criticism web sites. That said I may be misinterpreting your comment (apologies if so).

Comments are closed.