Jennifer Marohasy,
My good friend Patrick Lloyd, passionate about the new Artificial Intelligence ChatGPT, sent me the Tasmanian maximum temperature as an individual annual average of all daily values. all 155 weather stations since 1882. He downloaded these values from the Australian Bureau of Meteorology website with the help of ChatGPT. He wants to know from me if they show global warming.
After entering the numbers into my statistics program, I plotted a scatter plot with a suitable regression line, Figure 1. It shows that the maximum temperature in Tasmania is not as high as it was in the past. early 20th century and that there was a cooling period until about 1950.
The regression equation indicates that overall, the temperature has decreased at a rate of 0.0051 degrees Celsius per year. Converting this to climate change, we have a cooling rate of 0.5 C per 100 years. This doesn’t sound right, does it?
Official historical reconstructions for Tasmania also show temperatures dropping around 1960, and then rising to the lowest levels since 2000, it became very hot, Figure 2.
It is clear that the official maximum temperature reproduction for Tasmania is not just a merger of all maximum temperature series. Charts 1 and 2 show very different overall trends, although both show a cooling trend around 1960.
Is the Bureau’s historical reconstruction (Graph 2) a more accurate representation of Tasmania’s historic temperatures than the annual average of all the daily values ChatGPT provided Patrick (Chart 1)?
Each temperature series has its limitations, and sometimes problems only get more complicated when multiple temperature series, starting at different times and from different locations and altitudes, are combined. . That’s what Chart 1 represents, a simple compilation of everything available. It needs checking.
Patrick suggested a possible solution is to combine the longest and most complete maximum temperature series, perhaps the best 10% of the 155 weather stations, and discard the rest.
But there is an additional potential problem.
Since November 1, 1996, the Bureau has switched from mercury thermometers to probes in automated weather stations that measure this temperature data. Probes in automated weather stations represent a fundamentally different way of measuring temperature than traditional mercury thermometers, where daily maximums are used for manual readings at 9 a.m. each morning and then put mercury back.
Is it possible that the automatic weather stations are recording hotter for the same weather?
Considering only Larapuna (also known as Eddystone Point, station number 092045), as an example, this weather station has a long and relatively competitive temperature record. Larapuna will represent one of the ‘best 10 per cent’ from Tasmania.
Figure 3 shows the average annual maximum temperature (heterogeneous) measured at Larapuna for the first time with a mercury thermometer and since 1997 by probes in an automated weather station.
Considering the maximum temperature for this location, it seems to have increased by more than 1 degree Celsius after about 1997.
How can we tell if this apparent order change around 1997 was due to a change in the weather or a change in the equipment used to measure temperatures at Larapuna?
I told Patrick that before we start combining the individual temperature sequences, I wanted to tackle the problem of the mercury-to-probe shift. In fact, this problem has prevented me from continuing to reconstruct Australia’s temperature history for several years now.
Probes are generally considered to be more responsive to temperature fluctuations and, therefore, capable of recording both hotter and colder in the same weather. This is stated in the Research Report of the Bureau No. 032 Australian Climatic Observation Reference Network – Surface Air Temperature (ACORN-SAT) Version 2 (October 2018) by Blair Trewin:
In the absence of any other effects, devices with faster response times will tend to record higher maximum temperatures and lower minimum temperatures than devices with slower response times. than. This is most evident by the increase in the mean diurnal range. In most locations (especially in arid regions), it will also lead to a slight increase in mean temperatures, as short-term temperature fluctuations are generally larger during the day than at night.” (Page 21)
Maybe because Tasmania isn’t particularly hot, nor is it particularly arid, we don’t need to worry about change? I wonder what ChatGPT thinks? I wonder if ChatGPT can solve the device change problem?
How much hotter and how much colder the temperature of a recording probe is than an old-fashioned mercury thermometer will depend on the intrinsic time constant of that device and also how it was calibrated. Probes in automated weather stations can be calibrated so that the displacement is uniform in one direction.
If only we had parallel data for Larapuna, it is the measurements taken from the mercury thermometer and also the probe in the same Stevenson screen, we can know the degree of equivalence between the temperatures measured with different types of devices. So far, the Bureau has kept the parallel data private; The Bureau has a policy of not disclosing parallel data.
What is a number? other What are the possible problems we might face when combining temperature series to find the most accurate way of reconstructing temperature history for Tasmania?
****
UPDATE
ChatGPT, via Patrick, just provided me with a scatterplot for Larapuna, as of yesterday.
*****
This is the first in a new series of blog posts about reconstructing Tasmania’s temperature history with the help of Patrick and ChatGPT.
If you want to follow as the three of us try to understand how and why the two temperature series (Chart 1 vs 2) are so different, and try to reproduce our own temperature history for Tasmania, then, consider registering at my websitespecifically click on the ‘Temperature Trends’ box and you will start receiving information on this topic.
This photo of Patrick and I in front of the weather station in Tewantin, Queensland.
Related
