2GW Blog - How We Measure and Report
Global Temperatures
June 5, 2021
I thought the
temperatures would just be what they are, as the scientists say they are, and
we can go from there.
But a few folks don’t want to accept the temperatures as
real. I reckon these folks don’t
like what the temperature records are indicating, and rather than wrestling
with all the considerable available information, it’s easiest to just deny it
all as a conspiracy of deliberate wrong doing.
When you start to look at where the temperature
information comes from, in any depth, you soon go down a rabbit hole. For anyone interested in the excruciating details,
I am attaching 20-page Appendix 1 – World Temperature Measurement.
What Scientists Do with the Measured
Temperatures
First, we look at how
temperatures are measured (and adjusted, where appropriate).
Look on the
internet: Explainer: How do Scientists Measure Global Temperature? From
CarbonBrief.org.
To get a complete
picture of the earth’s temperature, scientists combine measurements from the
air above the land with measurements of the ocean’s surface collected by ships,
buoys and sometimes satellites also. The temperature at each land and ocean
station is compared daily to what is ‘normal’ for that location and time of
day, normal typically being the long-term average over a 30-year period. The
differences are called ‘anomalies’ and they help to scientists to evaluate how
temperature is changing over time. A positive anomaly means the temperature is
warmer than the long-term average; a ‘negative’ anomaly means its cooler. Daily
anomalies are averaged together over a whole month. These, in turn, are used to
work out temperature anomalies from season-to season and year-to-year.
See a Climate &
Environment at Imperial post by the Grantham Institute, 2015, Taking the
Planet’s Temperature: How Are Global Temperatures Calculated?
Granthaminstitute.com.
The World
Meteorological Organization recommends defining the temperature of a location
for a 24-hour period as the average of the maximum and minimum temperatures
recorded during that period. This is practiced in many countries. Although not the best calculation now
available, it is the easiest to apply consistently for the calculation of
temperature anomalies. For a better method, see Ma and Guttorp, University of
Washington, Seattle, WA, USA and Norwegian Computing Centre, Oslo, Norway. Some
countries use a linear combination of measurements taken at different times
during the day.
That same Grantham
Institute website indicates that land and sea surface temperature data are
quality-checked and adjusted to remove biases from each different measurement
process. On land, these adjustments include changes in the time of day of
observations and moves or changes to measurement locations. Observations from
modern, well-sited, automated equipment are treated as accurate and historical
data are adjusted to use the baseline set by these modern observations. For sea
surface temperatures from ships, one of the checks is that consecutive readings
recreate a sensible ship’s course, allowing time or location errors to be
spotted. Generally, land surface temperature adjustments increase the global
land temperature trend slightly (discussed in detail in my big document). Sea
surface temperature adjustments decrease the sea temperature trend considerably
(discussed in detail in my big document).
Overall, the surface
temperature adjustments cause a significant reduction in trends over a century
or more, while making little difference to the conclusion that global warming
is real. The surface
temperature adjustments make the calculated extent of global warming less, not
more.
The Temperature
Data Sets
There are four
main data sets available for global temperatures, discussed in more detail
below. The NASA GISTEMP record is the most detailed of the four data sets, with
each grid box two degrees longitude by two degrees latitude. The other three
data sets have grid boxes which are each five by five degrees.
The four data sets
differ in the number of land stations they have around the world:
·
HadCRUT4
has about 5500 stations
·
GISTEMP
about 6300
·
MLOST
has 7000 stations
·
The
number of land stations for JMA was not given.
HadCRUT4 stretches
back the furthest in history, to 1850. GISTEMP and MLOST both begin in 1880.
JAL starts in 1891.
While there are some minor differences between the four
sets of data, they are all quite consistent.
The four main groups
listed above all keep track of tropospheric temperature and all four show a
warming trend over the last 30 years.
Satellites are used
as a quality check. As well as measuring the temperature of the earth’s
surface, satellites can collect data from the bottom 10 kilometres of the
earth’s atmosphere, the lower troposphere. Unlike the surface temperature
record, tropospheric temperatures only extend back to the start of the
satellite era in 1979. Lower troposphere temperature is different from the
temperature at the surface of the earth but not much. The influence of the El
Nino weather phenomenon is much larger, for example. Scientists can use lower
troposphere measurements as a further evidence of a changing climate.
Early satellite data
were incorrect, because the scientist that programmed the satellite temperature
sensors got it wrong.
Working Up the Data into Temperature
Anomalies
After working out
the average annual temperature anomalies for each land and ocean station, the
next step is to divide the earth’s surface into grid boxes. Scientists work out
the average temperature for each grid box by combining the data from all
available stations in that grid box. The smaller the grid box, the better the
determined temperature of the box will reflect the actual temperature at any
given point, leading to a more accurate estimate of the global temperature when
you add them all together. The greater the number of temperature measurements
within a grid box, the better the determined temperature of the box will reflect
the actual average temperature for that grid box.
By combining the
results for all the grid boxes, scientists calculate the average temperatures
for the northern and southern hemispheres. The contribution of each grid box to
the global average temperature is adjusted to account for the fact that a degree
of longitude is bigger at the equator than at the poles. Taken together, the
two hemispherical values provide an estimate of the global average temperature.
It’s not as simple as adding the two hemispheres together, however. To avoid
the better sampled northern hemisphere dominating the temperature record,
scientists take the average of the two hemispheric values.
The Global Temperature Record
The most detailed
temperature information exists since 1850, when methodical (mercurial) thermometer-based
records began.
The web post Global
Temperature Record on en.m.wikipedia.org (last edited Nov 3, 2019) indicates
that proxy methods can be used to reconstruct the temperature record for the
historical period, before recent times. Quantities such as tree ring widths,
coral growth, glacial length variations, borehole temperatures, and isotope
variations in ice cores, ocean and lake sediments, cave deposits, fossils and
ice cores are correlated with climate fluctuations.
But hey, a proxy is
like kissing a picture of your sister.
The website
indicates that proxy reconstructions extending back 2000 years have been
performed, but reconstructions for the last 1000 years are supported by more
and higher quality independent data sets. The reconstructions indicate:
·
Global
mean surface temperatures over the last 25 years have been higher than any
comparable period since AD 1600, probably since AD 900.
·
There
was a Little Ice Age centred in AD 1700.
·
There
was a medieval Warm Period centered on AD 1000; the exact timing and magnitude
are uncertain and have regional variation.
Some Hic-cups With Satellite Derived
Temperatures
Christy and Spenser
from the University of Alabama were pioneers on using satellites to measure
temperatures of the surface of the earth and throughout the atmosphere.
Satellites do not measure temperature, per se. They measure radiance and an
algorithm is used to convert radiance to temperature. Initially, Christy and
Spenser had some mistakes in their algorithm, which lead to false temperatures.
They used the (unknowingly) false temperatures to make some wrong conclusions;
they claimed that the stratosphere was warming, as well as the troposphere.
Other scientists investigated and found errors in the methods used by Christy
and Spenser to adjust the data.
It took 13 years
after the original papers that that adjustments that Christy and Spenser
applied were found to be incorrect. See Mears et al. (2003) and Mears et al.
(2005). When the correct adjustments to the measurements were applied, the data
matched much more closely the trends expected by climate models. The corrected
data were also more consistent with the historical record of troposphere
measurements obtained from weather balloons. Once corrected, the differences
between the tropospheric and surface temperatures diminished – and a warming
trend was then clear for the troposphere. The corrected data show a cooling
trend in the stratosphere, consistent with the concept of global warming by an
enhanced greenhouse effect (it’s not more sun).
Despite all this, Spenser continues to be a big climate
change denier. He continues to think he is so right, even when he made some big
time mistakes in his pioneering satellite work, which lead to fundamentally
wrong conclusions.
My Comments
Deniers often
disdain the concept of adjusting the data. But the global warming game is so
complex that adjustments are indeed valid and needed from time to time.
Some people want the temperatures taken as
they are, without any adjustments. As discussed above, when this was done with
the four data sets, the extent of global warming looked greater, rather than
the less they were expecting.
Christy and Spenser made
some mistakes, likely unintentionally. Nonetheless, this experience is an
example of deniers latching onto something that fits their mindset, despite
their disdain for data adjustments. And then we find the very thing the deniers
ranted about, people (Spenser) adjusting data, had inherent mistakes that
created a false impression of the truth.
Just in this little story, we have two examples of deniers who should have been careful what they wished for.
Appendix 1 – World Temperature
Measurement
June 2000
What Scientists Do with the Measured
Temperatures
We are going to
cover this first because it will give you a better perspective when we look at
how temperatures are measured (and adjusted).
Look on the
internet: Explainer: How do Scientists Measure Global Temperature? From
CarbonBrief.org. To get a complete picture of the earth’s temperature,
scientists combine measurements from the air above land with measurements of
the ocean’s surface collected by ships, buoys and sometimes satellites also.
The temperature at each land and ocean station is compared daily to what is
‘normal’ for that location and time of day, normal typically being the
long-term average over a 30-year period. The differences are called ‘anomalies’
and they help to scientists to evaluate how temperature is changing over time.
A positive anomaly means the temperature is warmer than the long-term average;
a ‘negative’ anomaly means its cooler. Daily anomalies are averaged together
over a whole month. These, in turn, are used to work out temperature anomalies
from season-to season and year-to-year.
See a Climate &
Environment at Imperial post by the Grantham Institute, 2015, Taking the
Planet’s Temperature: How Are Global Temperatures Calculated?
Granthaminstitute.com. The World Meteorological Organization recommends
defining the temperature of a location for a 24-hour period as the average of
the maximum and minimum temperatures recorded during that period. This is
practiced in many countries. Although
not the best calculation now available, it is the easiest to apply consistently
for the calculation of temperature anomalies. For a better method, see Ma and
Guttorp, University of Washington, Seattle, WA, USA and Norwegian Computing
Centre, Oslo, Norway. Some countries use a linear combination of measurements
taken at different times during the day.
That same Grantham
Institute website indicates that land and sea surface temperature data are
quality-checked and adjusted to remove biases from each different measurement
process. On land, these adjustments include changes in the time of day of
observations and moves or changes to measurement locations. Observations from
modern, well-sited, automated equipment are treated as accurate and historical
data are adjusted to use the baseline set by these modern observations. For sea
surface temperatures from ships, one of the checks is that consecutive readings
recreate a sensible ship’s course, allowing time or location errors to be
spotted. Generally, land surface temperature adjustments increase the global
land temperature trend slightly (discussed in detail below). Sea surface
temperature adjustments decrease the sea temperature trend considerably
(discussed in detail below). Overall, the surface temperature adjustments cause
a significant reduction in trends over a century or more, while making little
difference to the conclusion that global warming is real. The surface temperature adjustments make
the calculated extent of global warming less, not more.
We cannot measure
the earth’s absolute temperature with high accuracy because the that would
require an extremely dense network of temperature of temperature measurements.
But we can measure the temperature anomaly with high accuracy, high enough to
see that global warming is happening. Particular regions can experience climate
extremes (e.g. heat waves) that are completely independent of climate change.
These local variations are usually balanced by an opposite extreme somewhere
else. By averaging over the globe, we rid ourselves of most of this more-local
weather variability and more clearly isolate the smaller climate signal.
Working Up the Data into Temperature
Anomalies
After working out
the average annual temperature anomalies for each land and ocean station, the
next step is to divide the earth’s surface into grid boxes. Scientists work out
the average temperature for each grid box by combining the data from all available
stations in that grid box. The smaller the grid box, the better the determined
temperature of the box will reflect the actual temperature at any given point,
leading to a more accurate estimate of the global temperature when you add them
all together. The greater the number of temperature measurements within a grid
box, the better the determined temperature of the box will reflect the actual
average temperature for that grid box.
There are four main
data sets available for global temperatures, discussed in more detail below.
The NASA GISTEMP record is the most detailed of the four data sets, with each
grid box two degrees longitude by two degrees latitude. The other three data sets
have grid boxes which are each five by five degrees.
The four data sets
differ in the number of land stations they have around the world. HadCRUT4 has
about 5500 stations, GISTEMP about 6300, MLOST has 7000 stations. The number of
land stations for JMA was not given.
HadCRUT4 stretches
back the furthest in history, to 1850. GISTEMP and MLOST both begin in 1880.
JAL starts in 1891.
By combining the
results for all the grid boxes, scientists calculate the average temperatures
for the northern and southern hemispheres. The contribution of each grid box to
the global average temperature is adjusted to account for the fact that a degree
of longitude is bigger at the equator than at the poles. Taken together, the
two hemispherical values provide an estimate of the global average temperature.
It’s not as simple as adding the two hemispheres together, however. To avoid
the better sampled northern hemisphere dominating the temperature record,
scientists take the average of the two hemispheric values.
Satellites are used
as a quality check. As well as measuring the temperature of the earth’s
surface, satellites can collect data from the bottom 10 kilometres of the
earth’s atmosphere, the lower troposphere. Unlike the surface temperature
record, tropospheric temperatures only extend back to the start of the
satellite era in 1979. Lower troposphere temperature is different from the
temperature at the surface of the earth but not much. The influence of the El
Nino weather phenomenon is much larger, for example. Scientists can use lower
troposphere measurements as a further evidence of a changing climate.
The four main groups
listed above all keep track of tropospheric temperature and all four show a
warming trend over the last 30 years.
The Global Temperature Record
The most detailed
temperature information exists since 1850, when methodical thermometer-based
records began.
The web post Global
Temperature Record on en.m.wikipedia.org (last edited Nov 3, 2019) indicates
that proxy methods can be used to reconstruct the temperature record for the
historical period, before recent times. Quantities such as tree ring widths,
coral growth, glacial length variations, borehole temperatures, and isotope
variations in ice cores, ocean and lake sediments, cave deposits, fossils and
ice cores are correlated with climate fluctuations.
But hey, a proxy is like
kissing a picture of your sister.
The website
indicates that proxy reconstructions extending back 2000 years have been
performed, but reconstructions for the last 1000 years are supported by more
and higher quality independent data sets. The reconstructions indicate:
·
Global
mean surface temperatures over the last 25 years have been higher than any
comparable period since AD 1600, probably since AD 900.
·
There
was a Little Ice Age centred in AD 1700.
·
There
was a medieval Warm Period centered on AD 1000; the exact timing and magnitude
are uncertain and have regional variation.
Early Temperature Measurements in the
Arctic
I started by trying
to find some good information on the temperature at the north pole, to use in
my musing of the temperature driving force for the Polar Jet Stream. I looked
at the website Climate4you, Polar Temperatures. It has information published by
NASA Goddard Institute for Space Studies on their GISS website, which also
provides a limited historical background. It also has data published by the
Climate Research Unit (CRU) which provides the mean annual surface air
temperature (MAAT) anomaly for the global region 70 degrees north to 90 degrees
north.
The GISS website
indicates that the first effort at measuring and recording temperatures in the
Arctic was by the Russians in 1923, through their Development of Russian Arctic
Research Stations. The GISS website states that the total number of stations in
the Russian Arctic remained small until 1929 and the quality of the equipment
was low; the GISS website does not state the number of stations, or comment on
the amount of data. The Russians also collected air temperature data on a few
of their icebreakers. By 1933, six ships were deployed with weather stations in
the Russian Arctic. In 1933, Russia added 15 new Arctic weather stations, but
again, the total number of stations at that time was not given in the GISS
website. In 1934, Russia added 26 more weather stations and another 10 in 1935.
No more history was given for the Arctic.
The GISS website
indicates that 1957 was the initiation of widespread meteorological
observations in Antarctica.
How We Measure Ocean Temperatures
This is discussed in
the web post Why Do Scientists Measure Sea Surface Temperature, by the National
Ocean Service, National Oceanic and Atmospheric Administration, US Department
of Commerce, oceanservice.noaa.gov, last updated Nov 15, 2019.
Oceans cover 71% of
the earth’s surface. To measure sea surface temperature (SST), scientists
deploy temperature sensors on buoys, ships and ocean reference stations. They
also use satellites and marine telemetry. The NOAA-led US Integrated Ocean
Observing System (IOSS) and NOAA’s Centre for Satellite Applications and
Research (STAR) merge their data to provide SSTs worldwide.
Marine telemetry
involves attaching tags to a wide range of marine species, from salmon smelts
to 150-ton whales. The tags allow the marine animals to be tracked, where they
are and where they go. Some of these tags measure water temperature. The
signals from these tags are picked up by research vessels, buoys, satellites
and other tracking networks.
Historically, ocean
temperatures were measured off ships by dipping a bucket in the water, near the
surface, hauling the bucket up and sticking a glass thermometer in the bucket
of water, waiting and then noting the observed temperature. This method resulted
in temperatures that are different than the actual temperatures. The temperature difference is caused by warming of the
water in the bucket by warmer ambient air or cooling of the water in the bucket
by cooler ambient air. See some rigorous work by Carella et al., 23 May 2017,
Measurements and Models of the Temperature Change of Water Samples in
Sea-Surface Temperature Buckets, Quarterly Journal of the Royal Meteorological
Society, Volume 143, Issue 706. The wet bulb temperature of the ambient air was
the biggest factor in errors, then the wait time between sampling and observing
the temperature. The difference between
the reported temperature and the actual temperature at the time could be as
much as 5oC. Furthermore, there was little standardization of the
time of day that these temperatures were measured. Beware of these historical
data; they could well be inaccurate.
Broadly, the evolution over time of the types
of buckets used to measure SST on ships was from wooden buckets (partly
insulated) to canvas (uninsulated) and then to plastic or rubber buckets
(typically well insulated)
The proportion of ships making bucket
observations has decreased over time with the introduction of engine room
intake and hull manifold measurements.
Starting about 1993,
the bucket method was replaced by installing dial thermometers in the water
intake piping of ships. The dial thermometers were typically installed
downstream of the intake pump; pumps inherently heat the water. Dial
thermometers typically have 1 degree C accuracy and are next-to-never
calibrated. Because of the heating introduced by the upstream pump and heat
from the warm engine room, these measured temperatures are typically 0.6oC
warmer than the actual temperature. Using this data as an historical reference
hides some of the apparent global warming. If these historical data are
compared with recent information, then today’s apparent global warming is
overstated. See Emery and Thomson, 2001, Data Analysis Methods in Physical
Oceanography, EoS Transactions, 80, Gulf Professional Publishing, pages 24 and
25. The water intake method also has the shortcoming that the water depth of
the intake port varies from ship to ship; in a stratified ocean, these
different depths can have different temperatures.
See a paper by Saur,
14 January, 1963, A study of the Quality of Sea Water Temperatures Reported in
Logs of Ship’s Weather Observations, US Bureau of Commercial Fisheries,
Biological Laboratory, Stanford, California. He found biases ranging from minus
0.5oF to plus 3.0oF in measurements, from each of 15
ships examined.
In some ships,
thermocouples are installed in the hull of the ship, just below the water line,
away from any heat sources. These sensors provide reliable and accurate
continuous sea surface temperature data. See the web post, Sea Surface
Temperature Sensors for Australian Vessels, imos,org.au.
Ship-based
measurements have provided rigorous sampling along major shipping routes but
these routes cover only a small fraction of the surface of the world’s oceans.
There is a dearth of ship-based information for the vast majority of the
world’s oceans.
The sea surface
temperature can now be measured by special satellites, tuned to measure the
temperature of the water surface. SSTs are measured from approximately 10
microns below the surface of the water (infrared bands) to 1 mm (microwave
bands) depths, using radiometers. I wonder how the whitecaps are allowed for in
the pyrometry.
Since the 1980s,
most of the information about global SST has come from satellite observations.
Instruments like the Moderate Resolution Imaging Spectroradiometer (MODIS) on
board NASA’s Terra and Aqua satellites orbit the earth approximately 14 times
per day, enabling these satellites to gather more SSTR data in three months
than all other combined SST measurements taken before the advent of satellites.
See Sea Surface Temperature, NASA, Jet Propulsion Laboratory, California
Institute of Technology, earthobservatory.nasa.gov.
See Satellite
Temperature, on the web at sciencedrect.com. It has a lengthy section of a
chapter from a book Taking the Earth’s Temperature by Schneider et al. 2019.
They indicate that a number of land surface and sea surface data sets are
available. Weather satellites do not measure temperature directly. They measure
radiances in various wave length bands. The measured radiance data have to be
converted to temperature using a radiative model that accounts for atmospheric
effects of the signal acquired by the satellite’s sensor at the top of the
earth’s atmosphere. Besides radiometric calibration, there are three other
factors that are critical to proper temperature derivation: emissivity,
atmospheric properties at the time and topography. Land cover, snow cover, soil
moisture, rocks, surface water, etc. all effect the emissivity of the surface
the satellite is looking at.
Today, in addition
to satellite and shipboard measurements, there are thousands of floats in the
oceans measuring temperature and salinity. These floats are used to validate
satellite measurements, in addition to sampling at depth in the water. The
floats include both larger anchored buoys and smaller floating-free buoys. The
surface drifters from the Global Drifter Program (GDP) provide about 60,000
night-time SST measurements per month at a shallow depth of 0.2 metres,
becoming the biggest contributor to the in-situ global SST, Ocean Currents and
Salinity measurements.
A major
accomplishment in the distribution of satellite derived SSTs occurred with the
Group for High Resolution Sea Surface Temperature (GHRSST) project. The project
provides all SST data sets in a common format that allows easy accessibility
across different computer platforms and operating systems.
How We Measure Land Temperatures
Thermometers for
measuring land temperature are not in contact with the ground, but within the
air about 1.5 metres above the ground, in a shaded weather station. Thus,
strictly, land temperatures should be referred to using the term “near surface
temperatures”. Readings are now automated, but previously were taken manually.
Satellites are also
used to estimate air temperatures just above the ground.
See Urban et al.,
May 2013, Comparison of Satellite-Derived Land Surface Temperature and Air
Temperature from Meteorological Stations on the Pan-Arctic Scale, Remote
Sensing, on the web at remotesensing-05-0. Land surface temperature information
from AVHRR (Advanced Very High Resolution Radiometer), MODIS (Moderate
Resolution Imaging Spectroradiometer) and (A)ATSR (Advanced Along Track
Scanning Radiometer) were compared to in situ air temperatures (Tair)
from the National Climate Data centre (NCDC). MODIS agreed best with ground
level air temperatures, Tair. (A)ATSR and AVHRR tended to indicate warmer and
colder temperatures, compared to Tair, for the positive and negative
temperature ranges. MODIS minorly overestimated temperatures in the positive
range. (A)ATSR had outliers around the freezing point. The errors were
associated with differentiating between clouds and snow, as well as ice-covered
regions. Generally, LST indications from all three systems were in good
agreement with the Tair data for the winter months, with deterioration in
agreement towards the summer months. AATRS had the highest inter-annual
variability.
See Tomlinson, 2012,
Comparing Night-time Satellite Land Surface Temperature from MODIS and Ground
Measured Air Temperature Across a Conurbation, Remote Sensing Letters, Volume
3, Issue 8. The article describes a pilot project over the summer of 2010 using
Moderate Resolution Imaging Spectroradiometer (MODIS) to compare land surface
temperature data and measured air temperature data from a custom network of
data bloggers across the conurbation of Birmingham, UK. Their results showed
that the night-time air temperature measured in meteorological stations was
consistently higher than the satellite-indicated land surface temperature, but
there was significant station-specific variability. The web did not provide the
full paper, so I could not look at the extent of the discrepancies.
See Kenawy, et al.
2019, An Assessment of the Accuracy of MODIS Land Surface Temperature over
Egypt Using Ground-Based Measurements, Remote Sensing, 2019, 11, 2369, on the
web at remotesensing-11-0.During the night, MODIS tended to under-estimate the
minimum ground level air temperature by minus 1.3oC during the
winter, minus 1.2oC during the spring and minus 1.4oC in
the fall. During the daytime, MODIS
markedly overestimated the maximum temperature in all seasons, with
discrepancies mostly above 5oC.
Satellite-Estimated Temperatures
Versus Surface Temperature Sensors
Satellites actually
measure the average temperature over the lowest 8 km of the atmosphere.
Climate change deniers like Ted Cruz have said that
satellite-based temperature measurements are the best we have, better than
surface temperature sensors.
The web blog by tamino.wordprees.com Surface Temperature
and Satellite Temperature, 2 February 2018 addresses this notion and provides
information to show that the notion is wrong. They concluded, “Surface temperature data are more reliable than
satellite atmospheric temperature data. For satellite data, RSS is far more
reliable than UAH and the notion that satellite data are “better’ is just
nonsense.” RSS TLTv4 is lower troposphere data from Remote Sensing Systems. UAH
TLTv6 is lower troposphere data from the University of Alabama at Huntsville.
From what I have
looked at, summarized above, land and sea temperatures derived from satellite
measurements are not as good as sensors which sit directly in the air above the
land and in the ocean. Satellites do not measure temperature directly. They measure
radiance. The measured radiance is converted into a temperature via a
horrendously complicated computer algorithm, which few humans understand. Why
get complicated when it can be so simple with a sensor? Why go indirect,
through a vague black box, when you can go direct? All that said, I see a place
for satellite-based temperature measurements when it’s all you got.
Use of Satellites to Measure
Temperature in the Troposphere
See the web,
Wikipedia, en.m.wikipedia.org, last updated Nov 3, 2019, Satellite Temperature
Measurements. Since 1978, microwave sounding units (MSUs) on National Oceanic
and Atmospheric Administration polar orbiting satellites have measured the
intensity of upwelling microwave radiation for the atmospheric oxygen, which is
related to the temperature of broad vertical layers of the atmosphere.
Measurements of infrared radiation from the ocean surface have been used to
infer sea surface temperature since 1967. This website indicates that over the
past four decades, the troposphere has warmed and the stratosphere has cooled.
They say that both these trends are consistent with the increasing atmospheric
concentrations of greenhouse gases.
Both the instrumental temperature record and satellites
show global warming.
Some Hic-cups With Satellite Derived
Temperatures
It is sometimes
argued in parts of the media (for example see comments by Bob Carter presented
in Skeptical Science) that the opposite is true – that the troposphere is
warming more slowly than the earth’s surface, even slightly cooling. This
argument stems from research published in 1990 by Christy and Spenser from the
University of Alabama. Other scientists investigated and found errors in the
methods used by Christy and Spenser to adjust the data. Also, the satellite
must pass over the same spot on the earth at the same time every day in order
to get a reliable temperature average. In reality, the time that the satellite
passes a certain spot drifts slightly as the orbit of the satellite decays. To
compensate for that, the data had to be adjusted. The MSU data are collected
from a number of satellites which provide daily coverage of about 80% of the
earth’s surface. Each day, the orbit shifts, so that 100% coverage is obtained
every three to four days. The microwave sensors in the satellites do not measure
temperature directly; rather they measure the radiation given off by the oxygen
in the earth’s atmosphere. The intensity of this radiation is directly
proportional to the temperature of the air and is therefore used to estimate
global temperatures. There are also differences between the sensors that were
onboard each satellite and merging all the data into one continuous record is
not easily done.
It took 13 years
after the original papers that that adjustments that Christy and Spenser
applied were found to be incorrect. See Mears et al. (2003) and Mears et al.
(2005). When the correct adjustments to the measurements were applied, the data
matched much more closely the trends expected by climate models. The corrected
data were also more consistent with the historical record of troposphere
measurements obtained from weather balloons. Once corrected, the differences
between the tropospheric and surface temperatures diminished – and a warming
trend was then clear for the troposphere.
Deniers often
disdain the concept of adjusting the data. But the global warming game is so
complex that adjustments are indeed valid and needed from time to time. We all
make mistakes. The MSU people made some mistakes, likely unintentionally.
Nonetheless, this experience is an example of deniers latching onto something
that fits their mindset, despite their disdain for data adjustments. And then
we find the very thing the deniers ranted about, people adjusting data, had
inherent mistakes that created a false impression of the truth.
See a more recent
evaluation by Mears et al., October 2017, A Satellite-Derived
Lower-Tropospheric Atmospheric Temperature Dataset Using an Optimized
Adjustment for Diurnal Effects, Journal of Science, Volume 30, Issue 19, pages
7595 to 7718. They indicated that previous versions of the dataset used general
recirculation model output to remove effects of drifting time of local
temperature measurement on the measured temperature. This cited paper presented
a method to optimize these adjustments using information from the satellite
measurements themselves. This newer method found a global-mean land diurnal
cycle that peaks later in the afternoon, leading to improved agreement with
measurements made by co-operating satellites. The changes result in global-scale
warming (global trend 70 degrees south to 80 degrees north, 1979 to 2016) =
0.174oC per decade, about 30% larger than their previous version of
the data set, which showed 0.134oC per decade. They said the new
dataset shows more warming than most similar datasets constructed from
satellites or radiosone data. However, comparisons with total column water
vapour over the oceans suggest that the new dataset may not show enough warming
over the tropics.
We see Mears had to
make two rounds of corrections to his calculations. The first round changed a
cooling impression into a warming trend. The second round increased the warming
extent. I wonder what deniers are saying about Mears now.
The Four Major Temperature Data Sets
Scientists use four
major data sets to study global temperature:
·
The
United Kingdom Meteorological Office Hadley Centre and the University of East
Anglia’s Climate Research Unit Jointly produce HadCRUT4.
·
In the
USA, the GISTEMP series comes via the NASA Goddard Institute for Space Sciences
(GISS)
·
In the
USA, the National Oceanic and Atmospheric Administration (NOAA) creates the
MLOST record.
·
The
Japan Meteorological Agency (JMA).
The internet
publication by Explainer: How do Scientists Measure Global Temperature? From
CarbonBrief.org, January 2015, presents a graph which compares the results for
the average annual global temperature over the past 130 years. The results from
all four data sets are quite consistent. They all show a warming trend, albeit
with some year-to-year variability. Generally, they all show 0.5oC
of warming, globally, by 2014, compared to the average temperature for the
period 1951 to 1980.
For further
discussion of the consistency of the various data sets (referred to as
“reconstructions”) see Skeptical Science, Are Surface Temperatures Reliable?
updated July 2015.
The GISTEMP data set
shows the fastest warming. JMA tracks slightly lower than the others. The JMA
is about 0.2oC lower than GISTEMP.
The main reason the
four data sets differ somewhat lies in how the different data sets deal with
having little or no data in remote parts of the world. Measurement errors,
changes in instrumentation over time and other factors make capturing global
temperature less than a straight-forward task.
Data coverage likely
has the biggest influence. NASA GISTEMP has the most comprehensive coverage,
with measurements over 99% of the world. By contrast, JMA covers just 85% of
the globe, with particularly poor data in the poles, Africa and Asia.
NASA’s GISTEMP uses
statistical methods to fill in gaps, using surrounding measurements. How much
each measurement influences the final value depends on how close it is
geographically to the missing point. NOAA follows a similar process for the
MLOST data set.
HadCRUT4 is the only
data set to leave regions with missing data blank, rather than try to fill them
in. This effectively assumes temperatures in the blank spots are in line with
the global average. This would not be an issue if the world were warming at the
same rate everywhere. But data suggests that the Arctic, for example, is
warming twice as fast as the global average. A missing Arctic data point could
lead to a global temperature that’s lower than the real world. For example,
updates to an old version of the temperature record (HadCRUT3) to include
better Arctic data resulted in a determined northern hemisphere temperatures
rise by 0.1oC.
Data gaps still
exist: most of Greenland, the Amazon basin, parts of Central Africa, and
Antarctica, mainly. This is a good place
for satellite-measured temperatures.
The internet
publication by Explainer: How do Scientists Measure Global Temperature? From
CarbonBrief.org, January 2015 discusses a 2013 paper that describes an
attempted fix using satellite data to reconstruct the holes in the surface
temperature record. Doing so suggested that the earth’s surface warmed twice as
much over the past 15 years than the HadCRUT4 suggests. The 2013 paper was by
Cowtan and Way, Coverage Bias in the HadCRUT4 Temperature Series and its Impact
on Recent Temperature Trends, 12 November, 2013, Quarterly Journal of the
Meteorological Society, Volume 140, Issue 683.
HadCRUT4 provides
gridded temperature anomalies across the world, as well as hemispherical and
the globe as a whole; data are available as monthly and annual values. HadCRUT4
is a combination of CRUTEM4 (a land temperature data set) and HadsetSST3 (ocean
temperature data set). See Met Office Hadley Centre Observations:
crudata.uea.ac.uk, Climate Research Unit Data. The data sets were developed by
the Climate Research Unit (CRU) at the University of East Anglia, in
conjunction with the Hadley Centre of the United Kingdom Meteorological Office.
Although the sea surface temperature data set was developed solely by Hadley. Data sets are updated monthly.
Temperature Results
Polar Region Temperatures
See North Pole Climate: Average
Temperature, Weather by Month. en.climate-data.org. Some monthly average
temperatures for the north pole: minus 24.2oC for January, minus 1
for April, plus 15 for June, plus 15.9 for July, plus 6.9 for September, minus
4.1 for October, minus 23.3 for December. The average annual temperature for
the North Pole region is minus 3.4oC.
See Wikipedia, Climate of the Arctic,
updated November 11, 2019. The coldest location in the northern hemisphere is
not in the Arctic, but rather in Russia’s far eastern interior. This is due to
the region’s continental climate, far away from the moderating influence of the
ocean. The coldest recorded temperature is minus 67.7oC.
At the South Pole, the mean annual
temperature is minus 49.5oC. Some daily average monthly data: minus
28.4oC for January, minus 53.7 for March, minus 58 for May, minus
59.8 for July, minus 59.1 for September, minus 38.2 for November, minus 28.0
for December. See Wikipedia, South Pole, updated November 2019.
Why so much colder at the South Pole?
It is at the centre of a large landmass, away from the moderating influence of
the ocean and at an elevation of 2835 metres, 9301 feet. Using the standard
adiabatic lapse rate of 9.8oC per km, we have to add 27.8oC,
50.0oF to get the equivalent temperature at sea level. Taking the
mid-winter temperature of minus 59.8oC and adding 27.8oC,
we get minus 31.0oC, about 7oC colder than the North Pole
it its mid-winter. The annual average temperature of the South Pole, corrected
to sea level is minus 21.8oC, much colder than the minus 3.4oC
for the North Pole.
While we are here, some interesting
information on Antarctica. It is the coldest, driest, windiest place on earth. It
has the highest average elevation of all continents. It is the 5th
largest continent, bigger than Australia.
Polar Regions
Temperature Anomalies
The GISS website
provides a graph with temperatures for the region 70 degrees north to 90
degrees north latitude, the Arctic region, for the period 1910 to 2010. In
looking at my discussion of the data below, keep in mind that the GISS website
indicates that the early data are sparse and of questionable accuracy.
·
The data
for both the annual average temperature and the average autumn to early winter
temperature show similar trends, with similar yearly variabilities. The annual
average temperature is more or less consistent at minus 6oC, the
autumn and early winter temperature is more or less constant at minus 4oC,
but both have a noticeable upswing that begins about 2000. By 2010, both
periods sit at about minus 3oC. These are actual temperatures, not
anomalies.
·
The data
for mid-winter show a downward trend from about minus 12oC in 1910,
to about minus 14oC in 2000, followed by an upswing to about minus10oC
by 2010.
These data show
temperatures increasing in the Arctic beginning about 2000.
On February 20,
2019, the Climate Research Unit (CRU) updated temperature data under the tag
CRU4. That tag provides the mean annual surface air temperature anomaly for the
region 70 to 90 degrees north latitude, the Arctic region. The data span 1910
to 2018. In this case, the anomaly is the temperature at each point, relative
to a selected WMO normal period of 1961 to 1980. WMO stands for the World
Metrological Organization, an intergovernmental organization with a membership
of 193 States and Territories.
·
Between
1910 and 1920, the temperature anomaly was about 1oC less than the
WMO. It was a bit colder then.
·
From
1920 to 1960, the temperature anomaly was variable but generally about 0.8oC
warmer than the reference period. The temperature anomaly was consistently
above zero.
·
1960 to
1990 is the WHO baseline, where temperature anomalies are consistently close to
zero.
·
From
1990 to 2018, all the temperature anomalies are above the WMO baseline period,
and there is a definite trend of increasing temperature anomaly, to about plus
1.8oC in 2018.
There is also CRU4
information for Antarctica. Between 1955 and 2010, the temperature anomaly has
been essentially constant, with little variability. But from 2010 onwards,
there has been a slight increasing trend to about plus 0.2oC in
2018.
The GISS website
also provides data from remote sensing systems. The National Oceanographic and
Atmospheric Administration provided data from their NOAA TIROS-N satellite for
the monthly average temperature anomaly of the lower troposphere since 1979. The
GISS website displays these data and includes a 27-month rolling average trend
line.
For the Arctic
region, 60 degrees to 82.5 degrees north latitude, the temperature anomalies
were fairly constant between 1979 and 1993. Since 1993, there has been an
increasing trend, up to about plus 1.3oC in 2018.
For the Antarctica
region, 60 degrees to 70 degrees south, the temperature anomaly has been
more-or-less constant between 1979 and 2018.
Global Average
Temperature Anomalies
See Met Office Hadley Centre Observations: crudata.uea.ac.uk, Climate
Research Unit Data.
·
1850 to
1910, global average temperatures were about 0.3oC below the WMO
baseline period of 1960 to 1990.
·
1910 to
1940, the temperature anomaly rose from minus 0.3oC to zero.
·
1940 to
1975, the temperature anomaly dropped to minus 0.1oC and then back
to zero.
·
1975
onwards, the temperature anomaly rose steadily to plus 0.7oC in 2019
Northern
Hemisphere Average Temperature Anomalies
See Met Office
Hadley Centre Observations: crudata.uea.ac.uk, Climate Research Unit Data.
·
1850 to
1920, northern hemisphere average temperatures were about 0.2oC
below the WMO baseline period of 1960 to 1990.
·
1920 to
1940, the temperature anomaly rose from minus 0.2oC to zero.
·
1940 to
1970, the temperature anomaly was zero, plus / minus a bit.
·
1970 to
1975, the temperature anomaly dropped to minus 0.2oC.
·
1975 to
1980, the temperature anomaly rose to zero.
·
1980 to
1985, the temperature anomaly was zero, plus / minus.
·
1985 to
2019, the temperature anomaly rose steadily to plus 0.9oC in 2019.
Southern
Hemisphere Average Temperature Anomalies
See Met Office
Hadley Centre Observations: crudata.uea.ac.uk, Climate Research Unit Data.
·
1850 to
1920, southern hemisphere average temperatures were about 0.3oC
below the WMO baseline period of 1960 to 1990.
·
1920 to
1940, the temperature anomaly rose from minus 0.3oC to zero.
·
1940 to
1960, the temperature anomaly was zero, plus / minus a bit.
·
1960 to
1980, the temperature anomaly dropped to minus 0.2oC and came back
to zero.
·
1980 to
1985, the temperature anomaly was zero, plus / minus.
·
1985 to
2019, the temperature anomaly rose steadily to plus 0.5oC in 2019.
The Impact of Urban Areas on Global
Temperatures
Three percent of the world’s land area is urban. See List of Urban Areas by Population,
Wikipedia, 2019.
See the EPA website,
Learn About Heat Islands, 2019, epa.gov. Buildings, roads and other
infrastructure replace open land and vegetation. Surfaces that were once
permeable and moist become impermeable and dry. These changes cause urban
regions to become warmer than rural surroundings, forming an island of higher
temperatures in the landscape. Heat islands occur on the surface and in the
lower atmosphere. On hot sunny days, the sun can heat dry exposed urban
surfaces such as roofs and pavement to temperatures 27oC to 50oC
hotter than the air. Meanwhile, shaded or moist surfaces in rural surroundings
remain close to the air temperature. Surface urban heat islands are typically
present day and night, but tend to be strongest during the day, when the sun is
shining. The annual mean air temperature of a city of a million people or more
can be 1oC to 3oC warmer than its rural surroundings. On
a calm clear night, the temperature difference can be as high as 12oC.
See Urban Heat
Island, 2019, Wikipedia, en.m.wikipedia.org. The main cause of the urban heat
island effect is the modification of land surfaces. Waste heat generated by
energy usage is a secondary contributor.
See Skeptical
Science, July 2015, Global Warming & Climate Change Myths, Does Urban Heat
Island Effect Exaggerate Global Warming Trends? Also, see another similar post
the same month. They first present a ‘Climate Myth’: A paper by Ross McKittrick
and Patrick Michaels concludes that half of the global warming trend from 1980
to 2002 was caused by urban heat islands. They then comment on this claim. They
say that when compiling temperature records, NASA, GISS go to great lengths to
remove influences from urban heat islands. Scientists have compared the
temperature data from remote stations (nowhere near human activity) to data
from urban sites. The process is described in detail on the NASA website
(Hansen et al.). Hansen et al. found that in most cases, urban warming was
small and fell within uncertainty ranges. Forty-two percent of city trends
investigated were cooler relative to their country surroundings, as weather
stations were often sited in cool islands within the city, say a park, rather
than in warmer industrial areas. NASA is aware of the urban heat island effect
and rigorously adjust for it when analysing temperature records.
Continuing from the July 2015 Skeptical Science sites. Jones et al.
2008, looked at sites across rural and urban China, which has experienced rapid
growth in urbanization over the past 30 years and therefore should show the
urban heat island effect. One of the studies had 40 / 42 sites, the other had
728 urban and rural sites. The differences between urban and rural areas were
very small, about 0.1oC. They continue by looking at where the
majority of global warming has occurred across the globe. They look at the 2006
global temperature anomaly. The greatest differences in temperatures from the
long-term averages occurred across Russia, Alaska, far north Canada and
Greenland, where there is very little urbanization, not where major
urbanization has occurred. They conclude that the urban heat island effect has
had no significant influence on the record of global temperature trends.
More from the July
2015 Skeptical Science sites. Parker 2006 plotted 50-year records of
temperatures observed on calm nights, versus windy nights. He found
temperatures over the land rose as much on calm nights as windy nights. From
that, he concluded that the observed warming was not a consequence of urban
development.
Think about how the average global temperature is
calculated. Grid by grid. With only three percent of the world’s surface area
covered by urban centres, how can elevated temperatures in these areas make
much difference? The representative temperature for each grid is a combination
of available temperatures throughout that grid. Even the grid that has New York
City in it likely has a large portion of rural area in it.
Challenges to the Adjustment of Raw
Temperature Data
See Skeptical
Science, Explainer: How Data Adjustments Affect Global Temperatures, posted
July 25, 2017, skepticalscience.com. This is a repost from carbon Brief by Zeke
Hausfather.
I am just going to
quote most of the article:
“Over the past
two centuries, the times of day, locations and methods of measuring temperature
have changed dramatically. For example, where once researchers lowered buckets
over the side of ships to collect water for measuring, we now have a global
network of automated buoys floating around the oceans to measuring the water
directly.
This complicates
matters for scientists putting together a long-term, consistent estimate of how
global temperatures are changing. Scientists must adjust the raw data to take
into account all the differences in how, when and where measurements were taken.
These adjustments
have long been a heated point of debate. Many climate skeptics like to argue
the scientists “exaggerate” warming by lowering past temperatures and raising
present ones.
Christopher
Brooker, a climate skeptic writing in The Sunday Telegraph in 2015, called them
“the greatest scientific scandal in history”. A new report (see Wallace et al.
June 2017, On the Validity of NOAA, NASA and Hadley CRU Global Average Surface
Temperature Data & The Validity of EPA’s CO2 Endangering Finding, Abridged
Research Report), ef-gast-data-research) from the right-wing US think-tank, The
Cato Institute, even claims that adjustments account for “nearly all the global
warming” in recent historical record.
But analysis by
Carbon Brief comparing raw global temperature records to the adjusted data
finds that the truth is more mundane: adjustments have relatively little impact
on global temperatures, particularly over the last 50 years.
In fact, over the
full period when measurements are available, adjustments actually have the net
effect of reducing the impact of long-term warming that the world has
experienced
Land and ocean
temperatures are adjusted separately to correct for changes in measurement
methods over time. All the original temperature readings from both land-based
weather stations and ocean-going ships and buoys are publicly available and can
be used to create a “raw” global temperature record.”
The post then
presents a figure which shows the global surface temperature record created
from only raw temperatures with no adjustments applied. It also provides the
adjusted land and ocean temperature record produced using adjusted data from
the US National Oceanic and Atmospheric Administration (NOAA). The figure also
shows the difference between the adjusted data and the raw data. The data scan
the period 1880 to 2016.
The figure
illustrates that adjustments to the data have little effect on global
temperatures after about 1950. The rate of warming between 1950 and 2016 in the
adjusted data is just under 10% faster than the raw data. And only 4% faster
since the start of the modern warming period in 1970.
The adjustments that have a big impact on the surface
temperature record all occur before 1950. Here, past temperatures all adjusted
up – significantly reducing the apparent warming over the last century. Over
the 1800 to 2016 period, the adjusted data actually warms more than 20% slower
than the raw data. The large adjustments before 1950 are almost entirely due to
the way ships measured temperatures. Between 1880 and 1950, the raw data
temperatures were about 0.2oC colder than the adjusted data. By
using the raw data, the apparent extent of global warming is increased, not
decreased. In other words, using unadjusted temperature data indicates more
global warming than the adjusted data do.
Challenges to the Reliability of
Surface Temperature Records
Example 1 – An Overview
See Skeptical
Science, 15 August 2017 Are Surface Temperature Records Reliable?
They start with a
‘Climate Myth’: Temp Record is Unreliable, from a 2009 post by Watts:
“We found (US
weather) stations located next to exhaust fans of air conditioning units,
surrounded by asphalt parking lots and roads, on blistering hot rooftops, and
near sidewalks and buildings that absorb heat. We found 68 stations located at
wastewater treatment plants, where the process of waste digestion causes
temperatures to be higher than surrounding areas. In fact, we found that 89% of
the stations, nearly 9 out of 10 – fail to meet the National Weather Service’s
own siting requirements that stations must be 30 metres (about 100 feet) or
more away from an artificial heating or radiating / reflecting heat source.”
Before I give the
rebuttal published by Skeptical Science, I will say that I have been in sewage
treatment plants. The anaerobic tanks used to digest the sewage sludge are
normally well insulated to help maintain the process temperature. Once you are
few feet away from these tanks, no radiant heat is noticeable. So, I find the
claim that waste digestion causes higher temperatures in surrounding areas
quite far-fetched bullshit. Regarding the 30-metre criteria, normally when
criteria / guidelines are set, safety factors are included, perhaps a doubling
in instances like this. I wonder just how far away the stations were from
artificial heating or radiating / reflecting heat sources, not 30 meters, but
what?
Skeptical Science
replied with the following:
“Temperature data
is essential for predicting the weather. So, the U.S. national Weather Service,
and every other weather service around the world wants temperatures to be
measured as accurately as possible. To understand climate change, we also need
to be sure we can trust historical measurements. A group called the
International Surface Temperature Initiative is dedicated to making global land
temperature available in a transparent manner. Surface temperature measurements
are collected from about 30,000 stations around the world (Rennie et al. 2014).
About 7000 of these have long, consistent monthly records. As technology gets
better, stations are updated with newer equipment. When equipment is updated or
stations are moved, the new data is compared to the old record to be sure
measurements are consistent over time.”
“In 2009, some
people worried that weather stations placed in poor locations could make the
temperature record unreliable. Scientists at the National Climatic Data Centre
took those critics seriously and did a careful study of the possible problem.
Their article, “On the reliability of the U.S. surface temperature record”
(Menne et al. 2010) had a surprising conclusion. The temperatures from stations
that critics claimed were “poorly sited” actually showed slightly cooler
maximum daily temperatures compared to the average.”
“In 2010, Dr.
Richard Muller criticized the “hockey stick” graph and decided to do his own
temperature analysis. He organized a group called Berkeley Earth to do an
independent study of the temperature record. They specifically wanted to answer
the question: “Is the temperature on land improperly affected by the four key
biases (station quality, homogenization, urban heat island and station
selection)?” Their conclusion was NO. None of those factors bias the
temperature record. The Berkeley conclusions about urban heat effect were
nicely explained by Andy Skuce in a SkS post in 2011.”
The Skeptical
Science post included a figure from Hausfather et al. 2013, covering 1890 to
2010, which shows that the USA land temperature network does not show
differences between rural and urban sites.
The Skeptical
Science site ended with:
“Temperatures
measured on land are only one part of understanding the climate. We track many
indicators of climate change to get the big picture. All indicators point to
the same conclusion: global temperature is increasing.
The adjustments
NASA makes to temperature data are fully documented and available on line. See
GISS Surface Temperature Analysis, National Aeronautics and Space
Administration, Goddard Institute for Space Studies at data.giss.nasa.gov.
Tamino explains it in a more digestible form in “Best Estimates” (looks pretty complicated and opaque to me).
The Skeptical
Science site includes a figure with red dots on a map of the world, showing
land temperature stations with at least one month of data in the Global
Historical Climatology Network (GHCN-M). The figure is from Rennie et al 2014.
It shows 7280 stations which were used during the period 1991 to 2013 in the
global surface temperature data bank. There is an extremely high concentration
of stations in the USA, on the south eastern side of Australia, and in Japan,
then lesser concentrations in Central Europe, lesser in Western Australia,
Central Africa, then places like Canada, South America, most of Africa. There
are only a few stations in Antarctica and Greenland. This figure is well worth
a good look.
Example 2 – Poorly Located
Temperature Stations
See Skeptical
Science 22 January 2010. I am going to just quote most of it.
“The website surfacestations.org enlisted an
army of volunteers, travelling across the U.S. photographing weather stations.
The point of this effort was to document cases of microsite influence – weather
stations located next to car parks, air conditioners, and airport tarmacs and
anything else that might impose a warming bias. While photos can be compelling,
the only way to quantify microsite influence is through analysis of data. This
has been done in On the Reliability of the U.S. Surface Temperature Record
(Menne 2010), published in the Journal of Geophysical Research. The trends from
poorly sited weather stations are compared to well-sited stations. The results
indicate that yes, there is a bias associated with poor exposure sites. However
the bias is not what you expect.”
“Weather stations are split into categories:
good (rating 1 or 2) and bad (ratings 3, 4 or 5). Each day, the maximum and
minimum temperatures are recorded. All data goes through a process of
homogenization, removing non-climatic influences such as relocation of the
weather station or change in the Time of Observation. In this analysis both the
raw, unadjusted data and homogenized adjusted data are compared.”
The post presents a
figure that compares the annual-average unadjusted maximum temperature from the
good and bad sites. There is another figure comparing the annual-average
unadjusted minimum temperature for good and bad sites. The time scale is 1980
to 2010.
“Poor sites showed a cooler maximum
temperature compared to good sites. For minimum temperature, the poor sites are
slightly warmer. The net effect is a cool bias in poorly sited stations.
Considering all the air conditioners, BBQs, car parks and tarmacs, the results
is somewhat a surprise. Why are poor sites showing a cooler trend than good
sites?”
“The cool bias occurs primarily during the
mid and late 1980s. Over this period, about 60% of USHCN sites converted from
Cotton Region Shelters (CRS otherwise known as Stevenson Screens) to electronic
Maximum / Minimum Temperature Systems (MMTS). MMTS sensors are attached by
cable to an indoor readout device. Consequently, limited by cable length, they
are often located closer to heated buildings, paved surfaces and other
artificial sources of heat.”
“Investigations into the impact of the MMTS
on temperature data have found that on average, MMTS sensors record lower daily
maximums than their CRS counterparts and conversely, slightly higher daily
minimums (Menne 2009). Only about 30% of the good sites currently have the
newer MMTS-type sensors compared to about 75% of the poor exposure locations.
Thus, it’s the MMTS sensors that are responsible for the cool bias imposed on
poor sites.”
“When the changes from CRS to MMTS are taken
into account, as well as other biases such as station relocation and Time of
Observation, the trend from good sites shows close agreement with poor sites.”
The close agreement
is shown in two figures, one for maximum temperature and the other for minimum
temperature over the period 1980 to 2010.
“Does the latest analysis mean that all the
work at surfacestations.org has been a waste of time? On the contrary, the
laborious task of rating each individual weather station enabled Menne 2010 to
identify a cool bias in poor sites and isolate the cause. The role of
surfacestations.org is recognized in the paper’s acknowledgements in which they
“wish to thank Anthony Watts and the many volunteers at surfacestations.org for
their considerable efforts in documenting the current site characteristics of
USHCN stations.” A net cooling bias was perhaps not the result
surfacestaions.org volunteers was hoping for but improving the quality of the
surface temperature record is surely a result we all should appreciate.”
So there we go again. After due consideration and
site-specific corrections, the rate of global warming is based on the
unadjusted data is higher, not less. When the adjusted data are used, the
calculated rate of global warming is less.
Recent Trends in Global Temperatures
Temperatures get measured at various places around the
world, some land based, some ocean water temperatures. These measurements,
certainly historical data, tend to be more numerous in developed countries.
Some of the land-based measurement points are in or near big cities. Cities
tend to create heat islands, resulting in false-high temperatures. Nonetheless,
I think an ocean temperature is relatively simple and realistic.
In about 1970, Environment Canada established a temperature
measuring station at the Alert airport, the closest point in Canada to the
North Pole.
Sea temperatures are measured in the top mm of water.
Most world temperature data are displayed as anomalies,
which is the departure from historic average conditions.
Global Science has a bar chart of the yearly average global
temperature anomaly between 1880 and 2015. These data show a more or less
steadily increasing trend; there is an upward blip in the early 1940s, but
those temperatures are less than temperatures observed in recent years. These
are long-term data.
Wikipedia has a chart of the temperature anomaly of the land
and the sea. The temperature of the oceans was more-or-less constant between
1950 and 1970; since then it has increased steadily and was plus 0.7oC
in 2018. The land-based data shows the same trend, with plus 1.3oC
in 2018. The warm anomaly intensifies in the Northern Hemisphere, compared to
the Southern Hemisphere. The northern polar area has a higher and more
extensive temperature anomaly than the southern polar region.
The International Panel on Climate
Change claims that at the equator, the upper ocean has warmed by 0.09oC
to 0.13oC per decade over the past 40 years. Superimposed on this
are the El Nino and La Nina climate cycles which can influence weather patterns
across the globe. These cycles happen at irregular intervals roughly every
three to six years, causing sea surface temperatures in the Pacific Ocean,
along the equator, to be cooler or warmer than normal. El Nino can increase sea
surface temperatures by 2 to 3oC.
It is hard to find a number for the
annual average temperature of ocean water at the equator, but 30oC
(86oF) is typical for ocean surface water in the tropics. See NOAA
Ocean and Exploration Research, How Does the Temperature of Ocean Water Vary,
oceanexplorer.noaa.gov.
Since 1980, the Northern Hemisphere has warmed faster than
the Southern Hemisphere. There are a few possible contributors:
·
The Northern Hemisphere has more land area and
less ocean surface than the Southern Hemisphere. Ocean warms less slowly than
land, given its great depth; heat can be mixed down.
·
Global ocean currents tend to transport heat
from southern waters into the North Atlantic and North Pacific.
·
Starting in the 1970s, there has been a
significant reduction in aerosol emissions from countries in the Northern
Hemisphere, especially Europe and North America. See Smith et al.2011 (9), Wang
et al. 2015 (10), and Crippa et al. 2016 (11). The Southern Hemisphere never
had significant aerosol emissions, so decreasing them is not going to have the
same impact as in the Northern Hemisphere. The aerosols are formed from sulphur
dioxide, as discussed above in the section on volcanoes. Sulphur dioxide is formed
by burning fossil fuels, especially coal and high-sulphur diesel fuel. The
reduction in aerosol emissions from countries in the Northern Hemisphere is due
to a move away from high-sulphur fuels to reduce acidification of water,
especially fresh water lakes.
All considered, there is a growing concern that tropical
rainfall patterns are shifting northwards.
The temperature anomalies are higher in the Arctic region
than the Antarctic.
The ten warmest years on record are: 2016 is the highest,
then 2015, 2017, 2018, 2014, 2010, 2013, 2005, 2009 and 1998.
I looked at a report on weather in the Mid-western USA (US
National Assessment: Midwest Technical Input Report: Historical Climate Sector
White Paper, 2012). They provided data on the length of the growing season,
bracketed by frosts. Prior to the 1930s, the growing season was 155 to 160
days. Between 1930 and 1980, it was 160 days. In recent years (2012), the
growing season has been 167 days. The report also provided data that show while
the average temperatures are higher now, the frequency of extreme high
temperature periods (4 days and longer) and extreme cold periods has decreased
between 1930 and recent times (2012); currently less temperature extremes, not
more.
In the USA, the 1930s had very high temperatures, with a big
drought in the mid-west. 1934 ranked 6th, behind 2012, 2016, 2015
and 1998, for the highest annual temperature in the contiguous 48 states. But
the associated land area is only 2% of the earth’s surface. Globally, the
annual average temperature for the 1930s was cooler than the average for the 20th
century. The 1930s drought in the USA was a geographically isolated event, not
representative of the whole world. People who hold up the USA temperatures of
the 1930s are myopically cherry-picking with bias.
The world’s oceans contain about 1,351,000,000 cubic metres
of water. If all of the incoming solar energy, 160 watts per m2 of
earth surface, went into the oceans on a completely mixed basis, the
temperature of the oceans would increase by 0.5oC per year. This is
an extreme, unrealistic calculation. Nonetheless, it illustrates the
temperature buffering power of the oceans; they are finite in their ability to
moderate temperature rise.
End of This Blog
Blackie Manana
Comments
Post a Comment