NIWA issued a response statement regarding the charges leveled by The NZ Climate Science Coalition here:
http://www.niwa.co.nz/our-science/climate/news/all/niwa-confirms-temperature-rise
They say:
Warming over New Zealand through the past century is unequivocal.
NIWA’s analysis of measured temperatures uses internationally accepted techniques, including making adjustments for changes such as movement of measurement sites. For example, in Wellington, early temperature measurements were made near sea level, but in 1928 the measurement site was moved from Thorndon (3 metres above sea level) to Kelburn (125 m above sea level). The Kelburn site is on average 0.8°C cooler than Thorndon, because of the extra height above sea level.
I’m not too impressed, especially when you see where the weather station for National Institute of Water and Atmosphere (NIWA) is, right on the rooftop next to the air conditioners:
Here is the station survey: NIWA_station_survey (PDF) and the Google Earth KML file
Thanks to: Dieuwe de Boer who did a good portion of station surveys in New Zealand last year.
The NZ Climate Science Coalition responds:
NIWA’s explanation raises major new questions
The NIWA climate controversy took a new twist tonight with the release of new data from the government run climate agency.
Reeling from claims that it has massaged data to show a 150 year warming trend where there isn’t one, NIWA’s chief climate scientist David Wratt, an IPCC vice-chair on the 2007 AR4 report, issued a news release stating adjustments had been made to compensate for changes in sensor locations over the years.
While such an adjustment is valid, it needs to be fully explained so other scientists can test the reasonableness of the adjustment.
Wratt is refusing to release data his organisation claims to have justifying adjustments on other weather stations, meaning the science cannot be reviewed. However, he has released information relating to Wellington temperature readings, and they make for interesting reading.
Here’s the rub. Up until 1927, temperatures for Wellington had been taken at Thorndon, only 3 m above sea level and an inner-city suburb. That station closed and, as I suspected in my earlier post, there is no overlap data allowing a comparison between Thorndon and Kelburn, where the gauge moved, at an altitude of 135 metres.
With no overlap of continuous temperature readings from both sites, there is no way to truly know how temperatures should be properly adjusted to compensate for the location shift.
Wratt told Investigate earlier there was international agreement on how to make temperature adjustments, and in the news release tonight he elaborates on that:
“Thus, if one measurement station is closed (or data missing for a period), it is acceptable to replace it with another nearby site provided an adjustment is made to the average temperature difference between the sites.”
Except, except, it all hinges on the quality of the reasoning that goes into making that adjustment. If it were me, I would have slung up a temperature station in the disused location again and worked out over a year the average offset between Thorndon and Kelburn. It’s not perfect, after all we are talking about a switch in 1928, but it would be something. But NIWA didn’t do that.
Instead, as their news release records, they simply guessed that the readings taken at Wellington Airport would be similar to Thorndon, simply because both sites are only a few metres above sea level.
Airport records temps about 0.79C above Kelburn on average, so NIWA simply said to themselves, “that’ll do” and made the Airport/Kelburn offset the official offset for Thorndon/Kelburn as well, even though no comparison study of the latter scenario has ever been done.
Here’s the raw data, from NIWA tonight, illustrating temp readings at their three Wellington locations since 1900:
What’s interesting is that if you leave Kelburn out of the equation, Thorndon in 1910 is not far below Airport 2010. Perhaps that gave NIWA some confidence that the two locations were equivalent, but I’m betting Thorndon a hundred years ago was very different from an international airport now.
Nonetheless, NIWA took its one-size-fits all “adjustment and altered Thordon and the Airport to match Kelburn for the sake of the data on their website and for official climate purposes.
In their own words, NIWA describe their logic thus.
- Where there is an overlap in time between two records (such as Wellington Airport and Kelburn), it is a simple matter to calculate the average offset and adjust one site relative to the other.
- Wellington Airport is +0.79°C warmer than Kelburn, which matches well with measurements in many parts of the world for how rapidly temperature decreases with altitude.
- Thorndon (closed 31 Dec 1927) has no overlap with Kelburn (opened 1 Jan 1928). For the purpose of illustration, we have applied the same offset to Thorndon as was calculated for the Airport.
- The final “adjusted” temperature curve is used to draw inferences about Wellington temperature change over the 20th century. The records must be adjusted for the change to a different Wellington location
Now, it may be that there was a good and obvious reason to adjust Wellington temps. My question remains, however: is applying a temperature example from 15km away in a different climate zone a valid way of rearranging historical data?
And my other question to David Wratt also remains: we’d all like to see the metholdology and reasoning behind adjustments on all the other sites as well.
“It isn’t an “alleged” station,”
So show me where it says what the thinggummy on the roof measures.
“the picture *is* a good example of incompetence. ”
To claim this you’ll have to know what it actually is that one sees on the image and what it is used for.
“A difference of .1C is a very significant difference, ”
Depends on what you want to show.
You where saying that the difference is “nowhere near 0.79”, which is rather wrong.
Besides, you are looking at daily maximum and minimum values. So to be able to actually compare the values from two different stations you have to assume that they are pretty similar (i.e. have their min. and max. temperatures at the same time).
Given the short time span and the fact that you are looking at min/max temperatures, I would say that the data fits quite well.
“and the data posted is only for a year, and not for the period of the early 20th century.”
The data you posted is for a month and not a year.
“airports needing accurate temperatures”
I said that the weather services need accurate measurements.
“Frank, please provide me with evidence (not claim) of which stations NIWA labeled “Airport”:”
I don’t know which of the airport stations NIWA has used in their analysis, but here’s their list of Wellington airport stations:
(search for wellington on http://cliflo.niwa.co.nz/pls/niwp/wstn.get_stn_nodt)
3445 E14387 Wellington Aero 01-Dec-1959 – -41.322 174.804
3446 E14388 Wellington Aero South 31-Jan-1980 01-Nov-1988 -41.338 174.804
3447 E14389 Wellington Aero North 31-Jan-1980 01-Nov-1988 -41.322 174.807
3437 E1438A Wellington,Air Nz 22-Nov-1991 – -41.332 174.808
10331 E1438C Wellington Aero Aws 01-Jun-1994 – -41.335 174.805
Frank (17:45:08) :
“It isn’t an “alleged” station,”
So show me where it says what the thinggummy on the roof measures.
Reply: I already provided that information once to you in Glenn (17:24:08)
“the picture *is* a good example of incompetence. ”
To claim this you’ll have to know what it actually is that one sees on the image and what it is used for.
Reply: Yes, and Anthony explain what it actually is in Nick Stokes (14:45:40)
“A difference of .1C is a very significant difference, ”
Depends on what you want to show.
You where saying that the difference is “nowhere near 0.79″, which is rather wrong.
Besides, you are looking at daily maximum and minimum values. So to be able to actually compare the values from two different stations you have to assume that they are pretty similar (i.e. have their min. and max. temperatures at the same time).
Given the short time span and the fact that you are looking at min/max temperatures, I would say that the data fits quite well.
Reply: How can you say “nowhere near” is “rather” wrong? In that short time, by your own figures is off a very significant difference in terms of differences of tenths of a degree, which is the scale of temperature trends. Unlike you, I assume little.
“and the data posted is only for a year, and not for the period of the early 20th century.”
The data you posted is for a month and not a year.
Reply: yes, I took a year from the db but only posted a month. Shoot me. Funny you didn’t reply to the fact that the period was not for the early 20th century.
“airports needing accurate temperatures”
I said that the weather services need accurate measurements.
Reply: Well goody for you.
“Frank, please provide me with evidence (not claim) of which stations NIWA labeled “Airport”:”
I don’t know which of the airport stations NIWA has used in their analysis
You didn’t have anything to say in this post, you made no real arguments, you avoided and snipped important points. That’s “unequivocal”.
Frank (17:45:08) : The “thinggummy” measures temperature. See here Richard (17:58:31)
And why are we getting sidetracked in things that do not matter?
NIWA MUST REVEAL EXACTLY HOW THEY GET A WARMING OF 1.9 C FROM THE RAW DATA THAT SHOWS NO WARMING. IT IS NOT ENOUGH TO GIVE THE WELLINGTON EXAMPLE. THE WELLINGTON EXAMPLE GIVEN BY NIWA SUFFERS FROM THE FLAW I HAVE POINTED OUT.
Summing up about this NIWA affair:
1. It was noticed in the raw data, which was available, that in 150 years there was NO warming whatsoever.
2. The ADJUSTED temperatures show a huge warming of 1.9 C over this period.
3. None of the details of this adjustment is available.
4. But when challanged – NIWA puts up a reply saying
a) We have already explained this to you – irrelevant – please send the details again, we dont have it.
b) look in Wellington we had to adjust upwards to the tune of 0.79 C (some merit in this, but flawed as the station used today as a proxy for the one discared in the 1920’s should be a lot warmer – why? because temperatures go up due to heat island effect and not only when you come down from an altitude). Thus the final adjustment will probably be less than 0.79 C
5. You have made an average adjustment of 1.9 C over all the stations. You have pointed out one station where you adjusted 0.79 C. But for the average to come upto 1.9 C the adjustments in the remaining stations have to be much larger.
This raises doubts. They didnt choose a typical station where they had to adjust up by say 2.5 C but one where the adjustment was less than half the average. Were all the stations adjusted upwards? None downwards?
The bottom line:
NIWA must reveal the adjustments made to the temperature readings at all the stations that they have done so, the raw data involved in the making of these adjustments and the reasoning and logic behind each of these adjustments.
Instead of waffling about inane side issues come up with a good reason why they shouldnt do so to remove our legitimate doubts.
Update: “NZ temperature rise clear” says NIWA
“1. Warming analysis is backed up by other observations
(a) measurements from ships [sea surface temperatures?]
(b) measurements from climate stations which have never been shifted
Dr Jim Salinger has identified from the NIWA climate archive a set of 11 stations with long records where there have been no significant site changes. When the annual temperatures from all of these sites are averaged to form a temperature series for New Zealand, the best-fit linear trend is a warming of 1°C from 1931 to 2008. We will be placing more information about this on the web later this week.”
Question: How come the complete raw data shows shows NO temperature rise if “measurements from climate stations which have never been shifted” show a rise of 1 C?
Surely these could not be all the stations?
In fact if there is no temperature rise in the raw data then there must be other stations (more than the 11 handpicked) that show a decline to balance out the rise in these stations.
Richard (09:52:01) :
Has Salinger actually released the adjusted data used in the original reconstruction NIWA claims he holds? By released I mean made public.
Glenn (10:23:41) : … Has Salinger actually released the adjusted data used in the original reconstruction NIWA claims he holds? By released I mean made public
No I think he (NIWA) needs to do that.
There seems to be some justification in their claim that they did point out to members of the NZ Climate Science Coalition that they had adjusted some temperatures and pointed out the elevation change in Wellington in 2006.
http://www.niwa.co.nz/news-and-publications/news/all/nz-temperature-rise-clear
The NZ Climate Science Coalition weaken their case by making wrong allegations in this regard. (maybe bad communication among the members, the left hand not knowing what the right is doing – but whatever it is a bit shoddy it appears)
What NIWA needs to do now is post exactly how they have arrived at their graph. (Make that public, just like their raw data). The complete adjustments, methodology, reasons and logic for anyone to analyse.
Then if anyone can point out any legitimate errors or concerns, these should be addressed by NIWA.
If not we will accept NIWA as being correct, but not until we have examined the adjustments.
PS The 11 stations that Dr Salinger has identified have “no significant site changes.”
Some changes may have been made. One has to see if they are “significant” or not.
If one of them for example has been moved “only” 200 metres, with no altitude change but next to an airconditioning exhaust, this could be significant in accounting for the 1 C rise.
And has the heat island effect been adequately adjusted for? (in other words some or many adjustments should be made downward and not upwards).
The NIWA owes proof that their work is correct to those who pay for it when it has been questioned. They have give a case for Wellington, but only Wellington. And Wellington was not selected by an indpependent party. They picked it.
No reasonable person would accept their work based only on data for one station. They need to provide their explantion of their adjustments for the other 10 stations. Any reasonable person would be suspicious that they have not doen so after several days. After much longer, any resonable person would assume that they are lying.
http://www.niwa.co.nz/news-and-publications/news/all/nz-temperature-rise-clear
“NIWA’s unadjusted climate data is available to anyone at no charge, through web access to the NIWA climate database. This has been the case since 1 July 2007.”
This is a blatant, deceptive lie.
3391 Thorndon 15July1912-31Dec1927 Station type: “Rain”
No temperature data for any date, but has rainfall data.
3385 Kelburn 1Jan1928-present Station type: “Climat Standard”
No temp data for 1928, 1929, 1930 Temp data available for later years
25354 Kelburn AWS (airport) Apr 2004-present station type: AWS
temp data available.
12443 Kelburn2 Nov1995-Nov1996 station type “Climat standard”
No temp data
13905 Kelburn Nov1995-Nov1996 station type: Rain
No temp data
As a skeptic of AGW alarmism, I find the original attack on NIWA’s adjustment of Wellington data, and this comment section, very discouraging. I fell a lot better when those on my side use valid criticism, especially when alleging intentional scientific misconduct.
The adjustments to the Wellington data are in fact exactly what I would do if the only information I had were the coordinates and the altitude. The adjustment uses the dry adiabatic lapse rate of the atmosphere, which is a pretty good estimate for height differences. That lapse rate is used daily by meteorologists all over the world, and is a strongly established atmospheric constant (it’s a direct result of simple thermodynamics).
While all of the other comments about questionable sites, etc, are valid, the initial attack on the adjustment as being somehow evidence of improper behavior is simply and clearly wrong. The best that can be said is that a better time series could perhaps be achieved by considering other factors, such as local effects.
Glenn makes a valid point (which I haven’t verified, but would be interested in) in claiming to find stations which are close together but differ in altitude where the lapse rate does not explain the variations, but that wasn’t part of the original attack. If that is a normal case, then the simple adiabatic adjustment has higher error bars. However, if the average of the errors is small (the errors have different signs, among other things), then the impact on the larger issues of wide area temperature measurement may be small (the errors would integrate out). Obviously, those making political recommendations based on this data must shoulder a huge burden of proof, which they have failed to do.
I think a more important line of attack in general, rather than questioning the motives of the NIWA scientists, is to challenge the use of surface temperature data in general as a proxy for the warming of the earth. The valuable volunteer “open science” project to analyze the official sites in various countries is helpful in this regard. However, even then, too many people, including skeptics, are clinging to the idea that the Earth’s surface temperature. A far better proxy is ocean energy storage, which we are just starting to measure.
The adjustment uses the dry adiabatic lapse rate of the atmosphere, which is a pretty good estimate for height differences.
Now is that applicable for the environment?
I didn’t know that New Zealand was dry enough to have a zero humidity?
Has it been moved to the Sahara on the quite, away from the ocean?
The dry adiabatic lapse rate (9.8C/km) is not the lapse rate of dry air – it is the lapse rate of unsaturated air. It might better just be called the adiabatic lapse rate. There is another lapse rate – the wet pseudo-adiabatic lapse rate.
The difference is this:
The (dry) adiabatic lapse rate is that which you would get by lifting a parcel of air where no condensation occurs. The wet (saturated) lapse rate (~5C/km) is what you get when lifting a parcel of air during condensation. It has “pseudo” attached to it because lifting air which is condensing is not adiabatic – energy leaves the system in terms of the latent heat of condensation.
On this particular topic, the Wikipedia article appears to be accurate and unbiased.
The dry adiabatic lapse rate is appropriate under the following conditions:
1) no condensation (by definition)
2) the air is adequately mixed (not stratified, no thermal inversion, no differential movement (advection) of air into the column at different altitudes in the area of interest.
Needless to say, these criteria may or may not be met in actual meteorological conditions, and criticism of the adjustment may be appropriate based on those issues (and others – such as the fact that the air is measured 2M above the ground and thus is very sensitive to micro-climate issues such as evapotranspiration, nearby asphalt, etc, or even the intermittent presence of condensation).
Additional note: I use the meteorological terminology here because it is in weather forecasting that I use the lapse rate.
John Moore (13:12:11) :
“Glenn makes a valid point (which I haven’t verified, but would be interested in) in claiming to find stations which are close together but differ in altitude where the lapse rate does not explain the variations, but that wasn’t part of the original attack.”
It was included in Anthony’s post. I disagree with your regard for adjusting station temps using lapse rates. If there were no change in the weather, an adiabatic rule could be used to adjust temps to different altitudes. But then whether the weather has changed is what we’re trying to find out. I haven’t checked the actual temp data out, but here’s a post from the more recent WUWT thread:
“Richard (22:32:56) :
… the CliFlo database reveals that Hokitika seems to be made up of Hokitika South from 1866 to 1965, followed by Hokitika Aero until the present. There was a decent 14- month overlap during the closing of South and the opening of Aero.
..
The interesting thing to note is that South is 0.3ºC cooler than Aero. Yet South is at the lower altitude of 4 m and Aero is higher, at 39 m. According to NIWA’s Wellington Thorndon explanation, the higher station should be cooler, based on the expected environmental lapse rate.”
You can look at station details, open this and click “get station list”
http://cliflo.niwa.co.nz/pls/niwp/wstn.get_stn_html?cstype=lat&cs_val1=-42.73&cs_val2=170.957
And from janama in the newer thread:
“Hokitika Aero”
http://data.giss.nasa.gov/cgi-bin/gistemp/gistemp_station.py?id=507936150000&data_set=1&num_neighbors=1
. If there were no change in the weather, an adiabatic rule could be used to adjust temps to different altitudes.
================
Which?
Dry? Saturated? Something in between? How do you decide?
My view is that you can’t.
That’s why homogenising different temperature series to join them up is bonkers.
Why do you need to do this at all?
After all, if one series for a period says the temperature has risen or fallen, you have information. That’s true for all series.
Why try and exclude, include, merge to make one series, and then merge again?
Each time you merge or fiddle with the data you add something that isn’t there or take something away that is.
Nick
Glenn, that is an interesting anomaly. As I said, I haven’t studied the data in general, just the specific critique. There is a reason to raise the temperatures from a higher altitude, absent contrary information. Put another way… if you do that on the average for a lot of stations, you will get better information (by far) than just joining the data without adjustment, and somewhat better information (i.e. >0) than throwing away the earlier data altogether.
Nick,
The lapse rate is the same all the way from dry to saturated. Only at saturation does it change, and then the new lapse rate is the same until the air mass becomes unsaturated. So it isn’t a liitle of this and a little of that. I am not saying (see above) that the lapse rate gives you perfect information, just that it is better than doing nothing at all.
As for homogenising different temperature series, the reason is obvious: to get more information than if you didn’t. It is normal scientific practice in many disciplines, and we do get valuable information form it. Again, that doesn’t mean that this particular homogenization was correct – just that the procedure was an appropriate first attempt (lacking other information, like data from other stations, meteorological information – say, if the second station is often above the cloud bases, etc).
John Moore (16:57:41) :
“Glenn, that is an interesting anomaly. As I said, I haven’t studied the data in general, just the specific critique. There is a reason to raise the temperatures from a higher altitude, absent contrary information. Put another way… if you do that on the average for a lot of stations, you will get better information (by far) than just joining the data without adjustment, and somewhat better information (i.e. >0) than throwing away the earlier data altogether.”
There is no reason why you would get better information, nor is there any way to test the accuracy of applying an arbitrary value absent a comparable concurrent record. And I’m not sure the Hokitika example should be regarded as an anomaly. It is an example of a similar location near the ocean in the same region.
Your argument here would apply to a situation where only one station measured actual temps, and all other locations were guessed at using temp lapse rates. I hope you can see from this example why applying an arbitrary value on a larger area would only invite less reliable data. Temps are different even between stations very close together at the same altitude! The two stations at the Airport, same altitude, a third of a mile apart are reported to be averaged .4F different.
You may get better information than without adjusting stations, but then again science isn’t about playing horseshoes, and “close enough” doesn’t cut it. In the Wellington example, we’re looking at a few tenths of a degree in a hundred years making the difference in a trend, and that difference could be, and in this case was, spread over multiple stations. The result could very well be worse than doing nothing at all.
You get better information because, *on average* the unsaturated adiabatic lapse rate is a pretty good estimate – because the atmosphere tends to be mixed (much more so farther above ground than the 1M thermometers, unfortunately). That does not mean that, for one station, you get better information. It means that, in a large average, you should. Not correcting for altitude, on average, is guaranteed to get you worse data, which is my whole point. But caveat… on any particular case, such as you raise, it may not be good, and may in some cases even be in the wrong direction.
And science is indeed about getting better information. It is only when it is used for policy issues that we have to draw the line and say that better isn’t enough. When we are talking climate policy, the “better information” that I advocate is probably nowhere near good enough.
In fact, I’m of the opinion that surface atmospheric temperature data in general is a lousy way to establish global temperature trends. I would much rather see ocean energy measurement and satellite sounder data than the very noisy and difficult surface thermometry, and before that, dendroclimatological data.
So John, let me ask you a question.
Why does one need to join up different temperature records to get one temperature record?
The obvious answer would be it’s obvious. 🙂
However,think about the question. GW is not that the temperature is 10C on average, but that the average has increased.
So if you put all the records into the pot, didn’t join them up, can you still get an average increase in temperature?
I have no idea. Averaging all the records, without adjusting, would be a silly exercise.
The basic idea that these guys are doing is not unreasonable. It is trying to squeeze as much as they can out of poor data… and scientists have to do that all the time. Likewise, the use of the dry lapse rate is not an unreasonable approximation, if you have no better data.
This leaves two issues:
(1) Are they doing the job right?
(2) Even with it done right, how useful is the data?
The answer to 1 is: I don’t know, but I have seen a couple of suspicious examples.
The answer to 2 is: better than not doing it. Not nearly good enough, as far as I can tell, to support the AGW hypothesis.
Niwa says, rather blandly, that there is no overlap between the Thorndon data and the Kelburn data. The fact is, overlap data was collected in 1927. So why does Dr Wratt not explain what happened to the data?
As an historical aside a biography of Dr Sir Edward Kidson, director of the NZ meteorological service at the relevant period, alludes to the poor quality of records prior to his arrival from Australia in 1927 (the previous director DC Bates having been sacked). What weight should be put on any pre-1927 data if it is unreliable?
I’m not suggesting averaging in that way,
What concerns me is that different temperature records are manipulated to get a temperature record for a grid point, that is then further maniplulated, and so on. Each stage moves you farther from reality.
So lets look at some of the things that are done.
Do they really need to be done?
1. Missing data. New values are subsituted.
2. Different temperature records from separate sites are merged to make one.
3. Some sites are dropped.
4. Sites are adjusted up and down. (mainly up it seems)
Of these I can see that there is an argument for 3. You could drop bad sites based on some sort of criteria. Like being urban or not conforming to the standards for measurement.
Time of observation adjustments are relevant.
Adjustments that bias a site are more difficult. I can’t see any argument for adjusting a site that meets the quality criteria. For those that don’t they need to be dropped.
I see no reason for selection one of a choice of two or more just because they are close subject to the other criteria.
I think averaging a months worth of data into one, and then processing even more the results is wrong too.
If we have a site that changes instruments, or moves, treat it as a second series. ie. No join the dots, and no distortions.
Then the deal is that you come up with delta temperature on a day by day basis by averaging all included sites deltas
It’s important that the delta is averaged, and not the temperature. It means you can combine different temperature records without resorting to joining the dots.
Nick