This may be partially due to low sun angles associated with latitude and the 24 hour day that Antarctica experiences in summer, along with snow albedo. I’m looking into the full paper and will have a follow up post later. See the map below, there are a lot of these AWS. As I’ve always said, if you don’t fully understand the micro-site conditions in which the measurements are made, and the instrumentation limitations, you can’t be certain of the data it produces.
Atmospheric temperature measurements biases on the Antarctic plateau
Christophe Genthon, Delphine Six, Vincent FavierLaboratoire de Glaciologie et Géophysique de l’Environnement, CNRS/UJF, Saint Martin, d’Hères, France
Matthew Lazzara, Linda KellerAntarctic Meteorological Research Center, University of Wisconsin-Madison, Madison, USA
Abstract
Observations of atmospheric temperature made on the Antarctic plateau with thermistors housed in naturally (wind) ventilated radiation shields are shown to be significantly warm biased by solar radiation. High incoming solar flux and high surface albedo result in radiation biases in Gill (multiplate) styled shields that can occasionally exceed 10°C in summer in case of low wind speed.
Although stronger and more frequent when incoming solar radiation is high, biases exceeding 8°C are found even when solar is less than 200 Wm−2. Comparing with sonic thermometers, which are not affected by radiation but which are too complex to be routinely used for mean temperature monitoring, commercially available aspirated shields are shown to efficiently protect thermistor measurements from solar radiation biases. Most of the available in situ reports of atmospheric temperature on the Antarctic plateau are from automatic weather stations that use passive shields and are thus likely warm biased in the summer. In spite of low power consumption, deploying aspirated shields at remote locations in such a difficult environment may be a challenge. Bias correction formulae are not easily derived and are obviously shield dependent. On the other hand, because of a strong dependence of bias to wind speed, filtering out temperature reports for wind speed less than a given threshold (about 4–6 ms−1 for the shields tested here) may be an efficient way to quality control the data, albeit at the cost of significant data loss and records biased towards high wind speed cases.
Received: May 13, 2011; Revised: August 01, 2011
=================================================================
h/t to the Hockey Schtick
“steven mosher,
Your comment is valid for a single station over the life of that specific instrument. Maybe. Maybe not. The instrument error may not be constant over time. As mentioned in the subject paper, low wind during summer resulted in maximum error. OK, you may say, on a year to year basis the bias is the same. That would hold true if windiness and cloudiness are constant year over year. Are you willing to assume that much?
####
I would not assume any of that. I think it’s important to understand the ways in which the bias can matter as well as the ways in which it would matter less than most people imagine. So, for example. If I have a scale that aways reads 10 lbs too heavy and I’m trying to measure your weight gain or loss, and for 30 years you hit the scales at 210 ( when truth is at 200) then that bias, since it is constant, doesnt impact my estimate of the trend, which is 0. So, obviously if windiness and cloudiness change in non uniform ways you’ve got an issue. duh. (They had similar problems with arctic instruments as well.) Personally, I’d lean toward dumping the bad data rather than trying to adjust it or trying to assume away impacts. However, It’s always fascinating to see how the answers dont change much when you take the three different paths:
1. keep the data. 2 dump the data. adjust the data.
In the end I would always look at all three paths and report out the answer. I would not assume anything. I’d just do the sums. But philosophically i lean toward dumping bad data rather than adjusting it and living with the uncertainty that sparseness creates. CLS in polar regions is very seasonally dependent ( at least in the arctic where it varies between 300km and 1000km depending on the season. )
There is a bit of chatter here on how hard it is to measure temperature. And this is most true. We develop a methodology of artifices that incorporate a presumption of physical perfection to often measure a sense rather than a constant, the subject of which is changing temperature constantly. And all are given to supposition.
Thus I find it intriguing that Warmists can measure the temperature of an entire planet or any sub-part thereof down to a tenth of a degree and project the same seasonally, annually, and even out to a century with absolute certainty.
Dave Wendt,
Might be true but I am having a problem understanding how that is possible. Water vapor is a much more powerful GHG than CO2. Any significant water vapor in the atmosphere basically swamps CO2 to the point where CO2 is a negligible contributor to greenhouse warming.
Summer at the poles see a breakdown of the circumpolar jet, mixing of the upper atmosphere with gas from lower latitudes and a great increase in atmospheric water vapor. Winter sees the polar atmosphere more sequestered from moisture in lower latitudes, what moisture there is precipitates out, the air gets extremely cold, extremely dry, and CO2 (which doesn’t change in atmospheric content between winter and summer). So in winter CO2 is a much larger proportion of total atmospheric GHG than in summer, the additional evaporation of snow in winter notwithstanding.
I can’t see any way CO2 is going to result in 1/3 of the greenhouse warming during any season, there just isn’t enough CO2 to do that. We are talking about a trace gas here, not something that makes up any significant portion of the atmosphere, and a gas that isn’t really that potent of a GHG.
I would venture that the surface temperature at the south pole in winter is due more to adiabatic warming than any “greenhouse” warming. In fact, there can’t be any greenhouse warming at the South pole in winter. Greenhouse warming means solar radiation is converted to heat, this IR is captured and re-radiated back to Earth. There is no solar radiation reaching the surface in austral winter so there can be no greenhouse warming.
The metaphor I prefer is, “one more arrow in the elephant.” At some point a “last straw” arrow will drop the beast to its knees, and it’ll be all over but the dining.
Excellent! I’m going to use that in place of CAWG in the future.
Anymore on Steig’s latest paper about Rosby waves? That one’s the real coffin nail for his 09 paper in Nature.
Spot o, Steve T and Latitude.
In my opinion ultimately, a defective instrument is just that, defective, giving wrong data. Any data or reading from a defective instrument is considered null and void in any other field of science. Only in climate science we always hear the reply ” but it does not matter or alter the overall conclusions ” when defective data and dubious methods are uncovered.
If they are defective because of direct sun shine, I have always found a that shade fixes the problem. Put a roof over them, a little parasol designed for 200 mph winds and the problem fixed.
A long time ago, at the times of school a good professor of physics used to say to us:
“Keep in mind, thermometers always measure the temperature of themselves! When you’ll use a thermometer always handle with care the result of the measurements”
Looks to me that too much people in the climate arena should learn that teaching.
Just to help out Real Climate with a response here…..
“A more likely cause of the warming is due to the increased concentration (AG) of CO2 inside the AWS’s Stevensen Screen which now retains more of the radiated heat entering from outside. This increased heat also melts more of the ice inside the screen raising the water vapour concentration and amplifying the increase a further 3 x. These findings are in line with IPCC model predictions and provide further proof of man made greenhouse gas induced warming.”
There Gavin – That should fix it for you …. 🙂
@Mosher: “If the sensor has a bias and you never change the sensor, then taking an anomaly effectively removes the bias because what you are concerned about is the trend and not the absolute temperature.”
We aren’t really interested in thermometer measurements over the last 20 years because they are too short to have any significance, so presumably we are comparing mercury thermometer readings with electronic thermometer readings and voila we have a discrepancy of over 10 Celsius in readings “on sunny days” which presumably is many days in Antartica since the poles have very low precipitation. Hence we have a major bias in the data that occured in the last 20 years towards greater “warming” in the readings as taken. Pretty obvious really that this was a possibility and any decent scientist would have taken this possibility of sytematic error into account but it isn’t even mentioned by Team AGW.
I notice that the electronic thermometer at Ross-on-Wye in UK shows readings 1-2Celsius higher than the mercury thermometer that was replaced when the station was revamped and re-opened by TV weatherman Ian McCaskill in 1985. It’s obviously a widespread problem. Technological progress will ensure that not only is there likely to be more carbon burnt but more of the old manual thermometers will be replaced by electronic thermometers – an excellent source of systematic error .
Are the data averaged over each grid? And what size are the grids? A number of 250,000 sq.km. grids appear to have no stations or at most one. Others (mainly on or near the West Coast, which is most affeted by wind) are reasonably populated. If we are questionning the reliability of the data, should we attach the same weight to areas where there are no, or relatively few, thermometers as we do to those where we have a number of (hopefully) fairly consistent readings?
If there are a number of thermometers in the same grid we can discard the results from those giving extreme readings, not just the outliers but any outside, say, the interquartile range. Then unless the bias is widespread we should obtain some fairly meaningful results. The trouble is that unless there are a reasonable number of thermometers in the grid this would not be possible and with none or only two or three there is no way of ensuring that bias has been eliminated or even reduced. Hence to give the same weight to those grids where thermometers are plentiful as to those where they are rare or non-existant is to introduce yet another bias.
When I read, as I did some time ago, that in cases where for some reason or other there are no data or some data is missing and therefore the data has been extrapolated (or intrapolated) from adjacent data, I begin to wonder how any climate scientist can ascribe a high level of probability to their models.
This implies that, now that the transition to electronic thermometers is about complete, global temperatures should stop rising. If greater cloudiness occurs, thanks to the Svensmark effect, a sharper-than-expected decline in temperatures will occur.
Silly me! I would just throw out the bad data and start over.
We can’t assume the inaccuracies and biases just somehow average out. That’s poor science.
@ur momisugly “Mike says: August 10, 2011 at 12:05 pm [Fixed, and insults snipped. ~dbs, mod.]”
You ‘fixed’ one occurrence. The other is clearly intentional. I’ll stand by my ‘insults’ but you snip whatever you wish.
Watts: “…false warming signal.”
crosspatch says:
August 10, 2011 at 10:21 pm
Dave Wendt,
Might be true but I am having a problem understanding how that is possible. Water vapor is a much more powerful GHG than CO2. Any significant water vapor in the atmosphere basically swamps CO2 to the point where CO2 is a negligible contributor to greenhouse warming.
H2O is indeed the dominant GHG. As I understand it the only reason that CO2 is responsible for such a seemingly significant portion of DLR at the South Pole is that it’s one of the driest places on Earth. Even at its seasonal max H2O is still incredibly scarce there. Although the CO2’s contribution of one third to total DLR seems significant, the total DLR at the Pole is never more than a fraction of what is seen elsewhere (60-120W/m2 compared to well over 400W/m2 over much of the Tropics) See Figs. 8 & 9 in the paper I linked above. Data that Evans and Puckrin collected using similar techniques suggest that when H2O is responsible for more than 200W/m2 of DLR CO2’s contribution is dramatically suppressed. A situation that prevails over most of the planet, most of the time
http://ams.confex.com/ams/Annual2006/techprogram/paper_100737.htm
Let’s get this straight. We are going to spend $28 trillion to save us from warming, that is caused by a $2 heat shield?
.
The recent figures of a temperature anomaly of warming, it was stated that though the northern hemisphere showed cooling the southern hemisphere showed that the world was warming? This with the sparseness of thermometers in the south would give the warming!!! So cooling is happening it is a measurement anomaly, that shows warming.
Having worked at the Pole during the ’03 Summer, the local and remote weather sensors are often times erroneous due to ice crystals, wind, ice berms etc.,
The ‘inspection’ of the remote stations is so random it’s a true shame from a climatological standpoint.
I and another weather person volunteered and was flown to dig out these stations and their sensors. This is obviously not implemented during the winter.
These stations are dated, never calibrated and troubleshot. Then again, Raytheon Polar Services Corporation, the NSF and SPAWAR (i.e. government contract, good enough for government work-type) run the Pole’s day-to-day, shoddily.
The next year I did a Summer at McMurdo Station, whereas the meteorological operations is a private company, Scientific Research Corporation. Lo and behold, a wonderful experience (though the food at Pole is infinitely better!) due to having state-of-the-art equipment, not DOS-era that is THEE software of the Pole. You know, unlike a government/government funded (tax dollar siphoned..) where accountability, balanced budget and strong work ethic is expected.. unlike Pole’s joke of a ‘weather ops’ facility.
Several people – including the original writer – have mentioned “low angles” of the sun, but nobody (yet) has actually called out what those “low angles” actually are.
Angles will vary based on latitude and time-of-day – just as they do in the high Arctic – but the following is typical:
Latitude 70 south: Sept, max angle, at noon during the equinox, = 20 deg.
Latitude 80 south: Sept, max angle, at noon during the equinox, = 10 deg.
Latitude 70 south: Dec, max angle, at noon during the summer, = 43.5 deg.
Latitude 70 south: Dec, “sunset” angle, at 1800 during the summer, = still 23.5 deg.
Latitude 70 south: Dec, min angle, at midnight during the summer, = 3.5 deg.
Latitude 70 south: Dec, “sunrise” angle, at 0600 during the summer, = back at 23.5 deg.
Latitude 80 south: Dec, max angle, at noon during the summer, = 33.5 deg.
Latitude 80 south: Dec, “sunset” angle, at 1800 during the summer, = still 23.5 deg.
Latitude 80 south: Dec, min angle, at midnight during the summer, = 13.5 deg.
Latitude 80 south: Dec, “sunrise” angle, at 0600 during the summer, = back at 23.5 deg.
Second, the “low angle” of the sun during that whole period of time that the sun is above the horizon during all dates between September and March – is amplified because all of the reflected energy from the (typical) ice and snow around almost every Antarctic station bounces back “up” off of the ground ice and snow at the same low incidence angle it came in from.
Looking at the weather screen in the link above, that means all the reflected energy from the sun for 6-18 hours of every day feeds back “up” into the thermometer right between the down-ward-sloping sun “screens”. This will NOT happen further north, back up in civilized areas where the thing was designed and tested and calibrated.
What are the differences? Much less ground reflection (no ice, no snow all year); what energy IS reflected from the limited ice and snow present during limited parts of the year further north is coming in at higher angles, that energy is reflecting up at higher angles, and so both the inbound energy and the reflected energy is better blocked by the downward-sloping screens. Also, at virtually all locations further north, reflections are blocked by nearby buildings, trees, rocks, and hills/environment. An isolated weather station will have fewer buildings and no trees and hills blocking the sun. (Most can even get placed so there is no blocking buildings or trees at all. Thus, in the Antarctic field installation, the sun is coming in from all angles all day and night during parts of the year.
This far south, the sun “rotates” all the way around the horizon: so the sun is not coming in from only the “expected” east-in-the-morning, west-in-the-evening – but will be coming in at midnight from the south, at 0600 from the east, at noon from the north, and at 1800 from the south again. 24 hours of heat radiation, all of it doubles from the sky and the ground.
Yet another error in AGW that is in favor of them.