By Andy May
It is common for the news media and consensus scientists to report global average surface temperatures without mentioning that both the warming rate and average temperatures vary a lot around the Earth over one year. For example, Earth’s global average surface temperature varies about 7° F every year. The rate also varies by hemisphere. Phil Jones and colleagues[1] show that the global average monthly temperature, from 1961-1990, was about 54° F in January and 61° F in July. In the Northern Hemisphere it varied from 46° F in January to 70° F in July, the respective average temperatures in the Southern Hemisphere are 61° F and 51° F.
As shown in Figure 1, the Hadley Climatic Research Unit HadCRUT5 temperature dataset suggests the Northern Hemisphere warmed 0.74°C (1.3° F) more than the Southern Hemisphere from 1979 through 2021. The UAH satellite temperature lower troposphere record is directionally similar but the difference between the hemispheres is much less, as shown in Figure 2. Both figures have identical vertical scales of 1.4°C. The UAH global satellite warming difference, from 1979 to 2021, between the hemispheres is about 0.2°C (0.4° F), less by more than a factor of 3.
The UAH and HadCRUT5 Southern Hemisphere warming rates are nearly identical, both are roughly 0.2 F per decade. But the Northern Hemisphere rates are very different, the HadCRUT5 rate is about 0.5 degrees F per decade and the UAH rate is 0.29 degrees F per decade. The HadCRUT5 Northern Hemisphere rate is a whopping 72% more than the UAH satellite rate. In the Southern Hemisphere the UAH warming rate is slightly larger than the HadCRUT5 rate, but the difference is very small, only 5%, and probably not significant. But 72% is certainly significant.
A difference in hemispheric warming rates is characteristic of solar and orbital forcing due to the tilt of Earth’s axis relative to the orbital plane combined with Earth’s rotation. It is not characteristic of forcing due to a well-mixed greenhouse gas, like CO2. The large Northern Hemisphere HadCRUT5 warming rate is suspicious.
About 68% of Earth’s land area is in the Northern Hemisphere, and the HadCRUT5 surface temperature record is a combination of two very different datasets. The HadSST4 sea surface temperature dataset measures the temperature at an ocean depth of about one foot and the CRUTEM5 land-surface dataset measures air temperature at an altitude of about six feet. The UAH satellite dataset is a lower troposphere dataset, and measures air temperature at about 10,000 feet, the measurements are all made using the same satellite data, as explained in my previous post.
It is possible that the urban heat island effect,[2] combined with the UK Hadley Centre “homogenization” algorithm has contaminated the land-only CRUTEM5 record and distorted it.[3] Because land surface temperatures are often highly variable over short distances there is a need to “homogenize” or smooth them to get mappable values. One of the large differences is between cities and rural weather stations, since cities are warmer than the surrounding countryside due to pavement, cars, air conditioners, etc. The difference is the urban heat island effect, and as Nicola Scafetta explains[3] the homogenization of the surface temperatures has the effect of spreading the extra warming in cities over large areas. He calls this “urban blending.”
If we look at the whole HadCRUT5 hemispheric record, the difference in rates shown in Figure 1 looks even more suspicious, as shown in Figure 3. The very large difference between the hemispheres is not seen in the earlier years of the HadCRUT5 record.
Conclusions
The UAH and HadCRUT5 records both show the Northern Hemisphere warming more rapidly than the Southern, but the amount of extra warming in the HadCRUT5 record is anomalous. While the Southern Hemisphere satellite and surface rates are similar, the Northern Hemisphere rates are very different.
The global UAH lower troposphere and HadSST4 warming rates are both 0.14°C/decade from 1979 to 2020, the same to two decimals, as shown in my previous post. The oceans cover 70% of Earth’s surface and the HadSST4 dataset is used in combination with the CRUTEM5 land-surface temperature record to produce the curves in Figure 1. Given this, the large difference in Northern Hemisphere warming rates between Figures 1 and 2 is very hard to explain. Frankly, it suggests there is a problem with CRUTEM5. Figure 3 suggests the problem is getting worse in recent years, not better. This topic is discussed in more detail in my latest book: The Great Climate Change Debate: Karoly v Happer.
The bibliography can be downloaded here.
This post was originally published at the CO2 Coalition.
-
(Jones, New, Parker, Martin, & Rigor, 1999) ↑
-
The Urban Heat Island effect is the result of humans paving over land and replacing transpiring trees and plants that have a natural cooling effect. Humans also have heat generating air conditioners, factories, and vehicles. Cities can be up to 7 degrees warmer than the surrounding countryside. ↑
-
(Scafetta, 2021) explains how the urban heat island corrupts the global surface temperature record. ↑
Andy, Temperature records of the two hemispheres allows analysis of seasonal fluxes and positive or negative accumulations. Some time ago, Chiefio made an intriguing statement:
“Solstice is the time when the sun stops the apparent drift in the sky toward one pole, reverses, and heads toward the other. For about 2 more months, temperatures lag this change of trend. That is the total heat storage capacity of the planet. Heat is not stored beyond that point and there can not be any persistent warming as long as winter brings a return to cold.
I’d actually assert that there are only two measurements needed to show the existence or absence of global warming. Highs in the hottest month must get hotter and lows in the coldest month must get warmer. BOTH must happen, and no other months matter as they are just transitional.”
Testing that assertion involves using BEST land only records, whereby Tmin and Tmax fluxes between summer and winter can be seen separately between NH and SH.
That analysis confirms that:
NH summer months are cooler than average overall and since 1950. Warming does appear since 1998 with a large anomaly in June and also warming in August. SH shows no strong pattern of Tmax warming in summer months. A hot December trend since 1998 is offset by a cold February. Overall SH summers are just above average, and since 1950 have been slightly cooler.
NH winter Tmin warming is stronger than SH Tmin trends, but shows quite strong cooling since 1998. An anomalously warm February is the exception in the period 1979-1998.Both NH and SH show higher Tmin warming in winter months, with some irregularities. Most of the SH Tmin warming was before 1979, with strong cooling since 1998. June was anomalously warming in the period 1979 to 1998.
Conclusion: I find no convincing pattern of summer Tmax warming carrying over into winter Tmin warming. In other words, summers are not adding warming more than other seasons. There is no support for concerns over summer heat waves increasing as a pattern.
https://rclutz.com/2015/06/22/when-is-it-warming-the-real-reason-for-the-pause/
Glad to someone dealing with “seasons”. Splitting NH winter and SH winter across years makes no climate science sense. It is more what computer programmers/mathematicians would do so calendar dates can be easily used?
Thanks Ron and thanks for the link. Sounds valid.
Ron. Congrats. I have always said that Tmax and Tmin are more important than Tmean to figure out the warming. Tmax is going down from 1997….
Tmax is a proxy for energy coming in.
So where does the extra heat from UAH come from?
There are 2 other sources….
“I have always said that Tmax and Tmin are more important than Tmean to figure out the warming.”
Me, too. If you want to know how hot it is, now, and in the past, you go with the Tmax. That’s what should be compared if we are talking about human-caused global warming/climate change.
I don’t know why there are no complete data sets on minima and maxima. Obviously it is because they reveal so much…
You can look at mine,
Who or what turned up the heat? | Bread on the water
One thing happening here is multidecadal oscillations. When they favor warming, they generally favor warming more in the northern hemisphere than in the southern hemisphere (that happened in the 1940s warm period) and more at the surface than in the satellite-measured lower troposphere. The converse is also true, with multidecadal oscillation favorability to cooling (or to bucking the trend of warming by increase of greenhouse gases) also happening more in the northern hemisphere than in the southern hemisphere, and I expect the same will happen for difference between the surface and the satellite-measured lower troposphere once we get satellite monitoring of a multidecadal oscillation downturn.
As for HadCRUT5: I see part of the great post-1979 (or post-1976) NH warming in HadCRUT5 being erroneous, from HadCRUT5 overconsidering the extra warming the Arctic has been having after 1976-1979. HadCRUT5 indicates more warming after 1979 than ERA5 and JRA55 do. Previous versions of HadCRUT err in the opposite direction, although I consider HadCRUT4 doing that only a little.
It would be helpful to compare the unadjusted or unhomogenized surface temperature data versus UAH. If the unadjusted, unhomogenized surface temperature averages more closely match the UAH data, it would be a good indicator of those surface temperature adjustment processes are illegitimate, something many of us have suspected for a number of years.
As Tony Heller’s comparison’s of US adjusted versus unadjusted data have shown, the adjustments suspiciously decrease temperatures prior to 1960 and increase temperatures after 1990, artificially creating a hockey stick that isn’t in the raw data. The “But it’s only US data” argument has some validity, except that data from the rest of the world is much sparser pre-1960 and the US has excellent records nationwide going back to 1900 if not earlier. And post 1960, there are still huge gaps in the surface temperature data from parts of the world which seems to be rather arbitrarily assigned values using questionable homogenization methods, extending their methodological error and increasing its impact to create a rising temperature narrative.
Ironically, even with this surface temperature manipulation, temperatures fall short nearly all of the IPCC’s models predictions.
Russia and the Artic region dominate the Northern Hemisphere. I was surprised to learn recently that Russia collects the data from the arctic weather stations. It would be interesting to see an analysis of Russian/Arctic surface warming rate vs the rest of the Northern Hemisphere. If someone wanted to manipulate public opinion among the political left in the west, no one would have a greater motive, means and opportunity than Putin.
Solar activity also affects circulation in the equatorial Pacific. Why might a triple peak of La Niña occur? If the solar wind had been increasing strongly since the beginning of the solar cycle a strong La Niña would likely occur, followed by an El Niño. However, the solar wind has been weak since the beginning of the cycle, hence a strong La Niña could not form. At the same time, as the strength of the solar wind increases the easterly winds in the equatorial Pacific will persist.
http://www.bom.gov.au/cgi-bin/oceanography/wrap_ocean_analysis.pl?id=IDYOC007&year=2022&month=03
It is sooo trivial! The warming is linked to the WHERE and WHEN of air traffic. It became a quantitative factor since the 1970s and stayed mainly in the NH.
https://www.flightradar24.com
There is a beautiful video of a GoPro sent into the stratosphere (over Germany). Pay attention to the 10 minutes when it crosses flight level from 01:05:00 to 01:15:00. At 01:11:26 you can even see an aircraft passing by.
https://www.youtube.com/watch?v=KnxvS9XFJnE
There is a dense haze layer just there, doubtlessly caused by air traffic. And what it does is to reflect only little sun light, as these ice crystals are far less reflective than water dropplets. On the other side this layer is extremely cold (~215K), only emitting ~120W/m2. So for every percentage of the sky it effectively covers, you get a forcing of 1.2W/m2, which is huge.
You are observing haze microdrops. The approximately 0.2 micron dust clay particle micronuclei accumulate in layers of low turbulence over USA and other eroding landscapes.
Absorbing about 60 watts m-2 incoming solar, this increasing humid haze is associated with global dimming.
The aridifying landscape below, now devoid of trees and wetlands, allows vast quantities of this clay dust to be eroded.
Water vapour collecting on these dust microdrops is not available to adequately condense into reflective cloud or precipitate. Very little latent heat is delivered aloft.
C’mon, man! Under either solar or infrared forcing, land will warm faster than ocean. The seasonal cycle of temperature demonstrates this clearly.
This is the case when the pavement is not covered with snow and the wind in winter does not blow from around the polar circle.
Roy,
Who is this meant for? If it is me, I agree with you. All the datasets show land warming faster than ocean, Plus, it is logical. I never said otherwise, did I?
The questions I pose are:
You are all missing the obvious.
Note that the arctic ocean is or was warming faster than the surrounding land there….
The extra CO2 is making the earth greener. This changes albedo. There is also heat coming free at night from the growing process of the vegetation.
Click on my name to read my essay.
Roy,
Below is an exchange with Nick Stokes from earlier in the comments on the same subject. I did not discuss land versus ocean in the post, except to point out 68% of the land is in the NH. My main point is that the total surface is warming 70% faster, according to HadCRUT5, than the LT, according to UAH. This is not consistent with the models and seems too large to be explained by a difference in land/ocean warming rates. This may clarify my thinking and the issue for you and any readers still reading the comments.
Nick Stokes
“Given this, the large difference in Northern Hemisphere warming rates between Figures 1 and 2 is very hard to explain.”
It is very easy to explain, and this post dances around it. The land is warming much faster than the ocean, and the NH has much more land. The reason is clear – the ocean is a heat sink, and SST is right on top of it, while solid land can absorb heat only very slowly. So, the extra heat must be transported down a temperature gradient to the sea.
UAH TLT is measuring at an altitude where there has already been a lot of mixing.
Andy May
Hi Nick,
The land is warming much faster than the ocean, and the NH has much more land. The reason is clear – the ocean is a heat sink, and SST is right on top of it, while solid land can absorb heat only very slowly.
This is an obvious statement, both the UAH and HadCRUT5 records agree that the land surface and lower troposphere(LT) over land are warming faster than the ocean surface and LT. No news here.
But it cleverly does not address my point. I said the UAH and HadCRUT5 NH warming rates were different. Do you admit that the land and ocean surface is warming much faster than the lower troposphere? You realize that this means the IPCC climate models are invalid.
Comparing AR5 to AR6 – Andy May Petrophysicist
Nick Stokes
“Do you admit that the land and ocean surface is warming much faster than the lower troposphere?”
Well, UAH tells me that TLT is warming at 0.134C/decade since 1980. RSS says 0.213. I can only think that satellite measures of TLT are uncertain, as the big version changes reinforce. Hadcrut 5 says 0.19; GISS says 0.188.
Andy May
Hmmm. We have TLT 0.14 to 0.21 and surface 0.19 to 0.19. Probably no extra warming in the TLT. Not looking good for the models.