Another Smoking Gun from Australia? How GISS adjusts temperature records in two adjacent sites
Ken Stewart, January 2010
(Updated November 2010 with better graphs)
Despite its assurances, GISS has adjusted the temperature records of two sites at Mackay to reverse a cooling trend in one and increase a warming trend in another. This study presents evidence that this is not supportable and is in fact an instance of manipulation of data.
I decided to have a look at the temperature records of the weather stations closest to where I live, near Mackay in North Queensland. The Bureau of Meteorology lists 3 current stations: Mackay MO, Mackay Aero, and Te Kowai Exp Station, plus the closed station Mackay Post Office. GISS has a list of nearby stations. One is “Mackay Sugar Mill Station”. I had never heard of it. Te Kowai Exp Station, only a few kilometres from Mackay, is in fact at the same co-ordinates as Mackay Sugar Mill. I checked on AIS for the GHCN site, and there is Mackay Sugar Mill on the map. The co-ordinates given by GHCN put it in the middle of a cane paddock 600m to the south of Te Kowai Sugar Experiment Station, so that’s definitely it! (If not, it’s identical in every other way!) And that is the closest weather station to my home, so I became even more interested.
Te Kowai is an experimental farm for developing new varieties of sugar cane, run by scientists and technicians since 1889. It has a temperature record of over 100 years with only a couple of gaps. So in fact it’s an ideal rural station for referencing a nearby urban station, as it should have a similar climate.
I plotted data from BOM for maxima and minima and obtained the means for Te Kowai, all Mackay city stations, all GHCN stations in our 5 x 5 grid, and several other towns and cities with long records (Te Kowai’s starts at 1908). This is because “ In our analysis, we can only use stations with reasonably long, consistently measured time records.”
GISS combines GHCN data from all urban stations at the same location, and then homogenises this with data from neighbouring rural stations. So I then plotted the same-location data and the post-homogenisation data.
A problem that appeared immediately is that the GISS annual mean runs from December to November, while BOM’s raw data is for calendar years. Most of the time it matches pretty well, but there are several examples of poor quality data. Another problem is that BOM does not compute a mean for any year with even one month of data missing, while GISS tolerates several missing months.
Here are graphs of the results.
First, BOM data for Mackay:
Mackay- composite of the Met Observation office from 1959, Mackay Airport, observations from the old Post Office, and a short Comparison series.
- The airport, on the outskirts of Mackay, is cooler by nearly 0.5 degree- (More on this later).
- You can see why they abandoned the PO after 1949!
As an aside, add in Te Kowai’s BOM figures to compare:
- The airport’s temperature from 1999-2009 differs from Te Kowai’s by an average of 0.1 degrees warmer, whereas the Met Office is an average of 0.5 warmer than Te Kowai for the same period.
- The PO tracks Te Kowai for a few years but about 0.25 degree warmer, then goes haywire. Apart from the anomolous jump in the 1940s, the warmest year in both was 1931.
Surely GISS would have used Te Kowai data when analysing Mackay, wouldn’t you think? As they say,
“The GHCN/USHCN/SCAR data are modified in two steps to obtain station data from which our tables, graphs, and maps are constructed. In step 1, if there are multiple records at a given location, these are combined into one record; in step 2, the urban and peri-urban (i.e., other than rural) stations are adjusted so that their long-term trend matches that of the mean of neighboring rural stations. Urban stations without nearby rural stations are dropped.”
Well, you’d be wrong.
This is what happened. Step 1: the unadjusted GISS data:
- The unadjusted data tracks BOM very closely, except it trusts the airport temperatures over the MO! So far, so good.
- They completely disregard the PO data.
Now, Step 2: adjust with the mean of neighbouring rural stations
- What’s going on? The homogenised data starts out cooler and catches up.
- The adjustment makes the slope of the adjusted data trend line STEEPER! The trend increases from 0.9 to 1.2 degrees C over 58 years, or 2 degrees per 100years!
- If they had included and adjusted PO data to the same trendline, 1908 would have been more than 1 degree cooler.
Here’s the plot of adjustments from the GHCN data:
Now let’s look at Te Kowai, in the country about 7km from Mackay.
Te Kowai/ Mackay Sugar Mill Station
- Look how hot it was in the 1920s and 1930s!
- Also note a steady increase of say half a degree from 1950-2008, and a linear trend of less then 0.2 degrees per 100 years.
Let’s look at what GISS does with this.
- The unadjusted temperatures pretty well match the BOM data, except that like most rural stations GISS stops the record in 1992.
- The linear trend of the unadjusted data slopes down, although the BOM data trends up again.
- Temperatures were much higher in the 1920s and early 30s. The hottest year was 1931. (The hottest summer was 1922-23). The next highest year after the 1930s was 1973. Only in the last 10 years have temperatures approached (but not reached) these levels.
- Some would say it’s a “Travesty”.
But have a look at what they do next! Here’s the graph of the homogenised data:
- Wow- when they adjust, they don’t muck around!
- Once again, the homogenised data is obviously different, showing much cooler temperatures in the first 40 years and higher at the end, with the result that the linear trend of the homogenised data slopes UP!
- There is now a warming trend of nearly 1 degree per 100 years.
- They have reduced the earlier warming but not eliminated it-1931 is still the hottest year.
Below is a graph of the adjustments GISS has made to the GHCN data. Notice the step pyramid shape you’ve seen before at Darwin. 1908 is suddenly 0.9 degrees Celsius cooler. The adjustments finally peter out to 0 in 1976.
Homogenisation adjustments to GHCN
So why didn’t GISS use the Te Kowai data to adjust the Mackay data, and make the Mackay trend less, not more? After all, Mackay has an Urban Heat Island effect, whereas Te Kowai’s should be very small. And why did they adjust the Te Kowai data at all, and so much?
Supposedly, they compare with nearby sites and make adjustments. But what nearby sites? To adjust Te Kowai, they must surely use only those with records from the early decades of the 1900s, like Clermont, Bowen, Charters Towers, Emerald, Barcaldine (too far away in my book). These are small-ish towns from the same 5 x 5 grid area as Mackay, even though they have vastly different climates- much drier for a start. Because they are “rural” GISS does not adjust their data at all. The good thing is, I’m familiar with all of them. I also looked at Ayr and Mt Morgan, “nearby” rural towns but outside the 5 x 5 grid, and Willis Island, out in the ocean.
Clermont (250km southwest) has a 0.8 increase/100years and was warm in the 1920s-1930s.
Charters Towers (328km north west)
0.7 degrees? But note the jump (UHI effect?) in the 1980s and the step down with the change to the airport, so handle cautiously. Also, warm years earlier.
Bowen (159 km north, on the coast)
Again note the big jump (UHI?) in the 1980s, and the big drop with the move to the airport. Use cautiously. Say 0.9 degrees per 100 years? (Or flat if you join the beginning to the end of the airport data.)
Emerald (289 km south west)
0.8 degrees over 100 years? Note the large step up (UHI effect?), after a declining trend, from the 1970s onwards, when the town boomed with expansion of irrigation, then coal mining. Also note the warm period in the early 20th Century. Airport data seems flat or lower. Handle cautiously.
Ayr (256km north west)
A huge jump of over one degree in early 1970s after a declining trend, so probably a change in surroundings- UHI, it’s a council depot after all- when compared with the DPI research farm just 6.3 km away. Also, this is evidence that GISS did NOT combine all records at this location. Not reliable. If used, an increase of 1 degree over 80 years, or 1.2 degrees per 100 years.
Mt Morgan (Walter Hall) (304km south)
Many gaps. A small decrease (1920s definitely hotter).
Barcaldine (487km inland)
Far inland, Barcaldine has a continental climate. Nearly 1 degree rise over 100 years, dropping lately. Too far away to be seriously considered as “nearby”.
I also looked at Willis Island, far out in the Coral Sea, and only from 1923, but the nearest island site with a long record.
Willis Island (542km north east)
Flat as a tack. BOM shows a very small rise. Should be a good site for checking ENSO effects.
There are no other sites within 500km with a long record.
The mean of the warming trends of all land based rural stations excepting Te Kowai is 0.77 degree Celsius approximately per 100 years. Leave out Mt Morgan it is 0.93. With Te Kowai it is 0.7.
First, an observation:
- Two unexpected consequences of this study are I believe I have discovered an Urban Heat Island signal in small towns, and it seems airports are not necessarily warmer places.
So: The GISS adjustments result in increasing the warming trend in Mackay, and introducing a non-existent warming at Te Kowai. Te Kowai and Mackay both end up with warming trends of 1 degree and 2 degrees per century. Now, GISS probably have a perfectly satisfactory algorithm which applies over most of the world, which averages the data of nearby stations to correct UHI effects in urban areas. However, only ONE “neighbouring” station has a warming trend of over 1 degree per 100 years, and it has a noticeable UHI effect of its own and in my opinion shouldn’t be relied on, (and another high one is hundreds of km inland.) All of the other nearby sites with long records have much less than 1 degree warming, and most have noticeable UHI effects. Most, like Te Kowai, show a warm period in the early 20th Century. The mean of the trends of all neighbouring long term stations (0.77 degree Celsius approximately per 100 years) does not warrant these adjustments. Even if they use the trend of the long term mean of the rural stations this is still only 0.9 degrees, and less than the homogenization adjustment for Te Kowai. Further, the UHI of Mackay is corrected UPWARDS.
How can GISS justify their manipulation of the data, which they claim not to do? How have they arrived at the adjustments for Te Kowai and Mackay? And besides, don’t they say that urban stations are adjusted by comparing with rural stations, not the other way around? They don’t adjust any other rural stations that I’m aware of.
Here’s the explanation, and THIS IS THE SMOKING GUN-
They are allowed this manipulation simply because GHCN has declared at some stage “Mackay Sugar Mill Station” (Te Kowai) to be a “small town”, with a population of 35,000. Exactly the same population as Mackay.
Well, actually it’s surrounded by cane fields which have not changed much for the last 130 years, is about 7 km from Mackay and 2.5km from the nearest suburb (Ooralea).
There are 2 or 3 buildings around it for the manager, greenhouses, and laboratories, and its population on a good day might be… 10.
This has allowed them to adjust its data, when other rural sites are not adjusted. It also allows them to NOT include the pre-1950 data from Te Kowai with the other rural stations, when adjusting Mackay’s data.
Why do I suspect this is deliberate? Because the Te Kowai GHCN data ends in 1992, just like all the other rural stations in this area. It was once classified as rural.
What would be the reason for this adjustment? Who knows, but the effects are obvious:
- Early warmer temperatures are artificially cooled.
- These two stations now have an extra warming trend and now do their bit to support the warming trend at the other 2,300 stations worldwide.
And it’s happening in my own backyard! I’m furious!