Thursday, May 03, 2007

Statistical Proof of sun caused global warming in South Australia Part I

Abstract: Statistical analysis of the rate of temperature change between times of proximity has shown that the rate of increase of temperature as a measure of time has significantly increased up until 3pm and then decreased until 9am in South Australia over the last 50 years. If Co2 levels were the major cause of global warming, then no rate of change in temperature between near times should exist, eg they should all increase at the same rate. Our analysis proves that this is not the case, and that temperature is increasing at a greater rate when the sun is at it’s peak in the middle of the day. Hence our conclusion is that the major force behind global warming is not Co2 levels, but the sun.


When looking at the graphs at 3am,6am and the minimum for south east southern Australia, something puzzled me. Whenever 3am or 6am had a negative anomaly (eg a lower than expected temperature for that year), the minimum temperature for that year was average. Whenever 3am or 6am had an average temperature for the year, the minimum temperature was significantly higher than normal.

This pattern was even more extreme in more recent years. It was because of this that I decided to look into the rate of change of temperature anomalies between neighboring times. We actually said a few things about this in the last post on the area:

We even found that temperature anomalies at 3pm were significantly higher than at Noon (p<0.01) and that temperature anomalies at 9pm were significantly lower than at 6pm (p<0.01). This indicates, that the temperature around south east southern Australia is heating up during the day, but more importantly is heating up at a greater rate when the sun is at its hottest. Does this not clearly indicate an increase in solar heating?

But actually looking at graphs of this made the pattern that more obvious than simple significant figures. But first lets hypothesise what would happen to the rate of change, if we believe that the sun has been the major cause of most recent warming.

We have said shown in a very early post on this blog, that in the last 5 years of more recent warming, almost all of this has occurred during the day when the sun it up. The pattern is clearly identifiable in the graph below. The rate of change of temperature anomalies have increased rapidly up to 3pm and then decreased. Even at midnight, the temperature is still a little hotter than normal, largely to due the extra hot sun heating up the atmosphere. By 3am and 6am, the atmosphere it would seem is not influenced by the sun.

Similarly from 1947 to 1976 when we saw a slight decrease in overall global (and Australian) temperature, most of the decrease occurred during the heat of the day around 3pm. This indicates again, then when the overall temperature is cold, it is because it is a lot colder during the heat of the day and not equally hot overall.

If Co2 were the major cause of global, and Australian warming, then we would expect an equal increase in temperature at all times of the day and night. Maximum and minimum temperatures agree with this, however, the analysis shown here strongly disagree. In fact it disagrees so much, that it is clearly obvious that Co2 could not be the major cause. What else heats the world up with increasing rates up to around 3pm and then decreasing? The sun of course.

So our hypothesis on rate of temperature anomaly change between neighbouring times in south eastern south Australia, given that the sun is the major cause would be very similar. We should find little difference in the overnight temperatures of 3am and 6am, but should find significant increases in the rate of change of temperature anomaly’s leading up to 3pm and then significant decreases in the rate of change of temperature anomaly's. This of course is looked at over time (in this case years). The rate of change should actually grow stronger as the years go on (as we have seen stronger increases in maximum and minimum temperatures).

So does the data suggest this? Let’s find out.

Whilst temperature data at midnight is limited, we do find that the rate of change of temperatures at 3am is on average 0.1 degrees Celsius less than that at midnight. In other words, the temperature at 3am is cooling down a rate of 0.1 degrees quicker than it is at midnight. Perhaps this is because the influence of the sun is now very limited? If Co2 was the major cause of warming, we should not see any pattern at all in this graph of limited years.

As a surprise, the rate of change of temperature anomalies at 6am have decreased significantly from 3am. In fact this rate of decrease has been at 0.5 degrees per 100 years. Why is 6am getting a lot colder with relation to 3am?

When looking at changes between 9am and 6am, we find no major change with the exception of the latest 10 years all being warmer. So the last 10 years, we have seen temperatures at 9am increase significantly more than at 6am. This isn’t surprising as this is when the sun increases, although I have to admit I would have thought the increase would have been more substantial (if the sun was the major cause of global warming).

But now the pattern, as predicted, is starting to hit in. Temperatures at Noon were significantly increasing as compared to 9am at a rate of 0.4 degrees per 100 years and the significant increase in temperatures at 3pm as compared to Noon was even greater (1.2 degrees per 100 years).

No major changes were found in the relationship between 6pm and 3pm, although it must be noted that a large negative trend was found in the last 6-7 years (eg. 6pm has been cooling quicker than at 3pm). And when the sun is starting to lose it’s influence of the temperature (between 9pm and 6pm) in south east southern Australia, we see that the rate of change more recently has been negative at a rate of 0.5 degrees per 100 years. This pattern, though a small sample size, is still obvious when looking at changes in temperature anomalies over time between Midnight and 9pm.

So believe it or not, we have proven exactly what our hypothesis predicted. That the rate of change in neighboring times would increase more significantly over time when the sun is getting hotter (around 3pm) and then decrease with time later in the day. I strongly recommend you view the linked graphs above to see it for yourself.

This is a clear indication, that the major driving force behind temperature change is the sun. Should Co2 levels been the main force, then would have seen no patterns at all in these graphs, but the patterns are clearly obvious and only point to one possible conclusion.

Of course you are asking, this is only a small sample of Australian weather stations yet alone the world, and yes we will get to them. I can’t let all the eggs out the basket at once!

But this analysis still doesn’t look at why minimum temperatures have increased despite early morning temperatures having not done so. We have hypothesized in the past how the sun is influencing minimum temperatures as well, but can we prove that statistically?

You’re just going to have to wait till the next article for Part II on Statistical Proof of sun caused global warming in South Australia.


Phil_B said...

So you are going to tease us with your explanation of why minimum temps are increasing, while night time temps are not. I think I know the answer, but I look forward to seeing what data you have.

BTW, showing that increasing minimum temperatures are not a good proxy for nightime temps (which are not increasing) is significant and should rattle the climate modellers cages.

Phil_B said...

Jonathan, one point. The sun isn't hotter in the afternoon. Maximum solar heating (i.e. when the sun is hottest) will be when the sun is at its highest point. Depending on the location's time zone's relation to dawn and dusk, this will be around mid-day.

However, maximum cumulative solar heating (relative to cooling) will be mid to late afternoon, around 3pm. That is the time when solar heating throughout the day will show its maximum effect on temperature (in the day). So you are correct in predicting temperatures at this time will show the largest increase from an increase in solar heating throughout the day.

IanS said...

Is this not all simply measuring the ozone layer effect?

Anonymous said...

The observables that it is typcially warmest in midafternoon and coolest pre-dawn are in the "intuitively obvious to the most casual observer" category. The graphs have some interesting trends, I think. The statistical analysis is eagerly awaited.

Jonathan Lowe said...

couple of things,
phil - i know the sun isn't hottest in the afternoon. It is afterall around the same temperature all day long, but it effects us most around midday and then the atmosphere heats up the most a bit after that.

Ias? Ozone layer effect? seriously?

Anonymous. I knew someone would say this. of course it's hotter at 3pm then at 6am, but this is not what this is measuring. it's not even measuring the difference from the average at these times, but rather it is measuring the rate of change of the difference from the average temperature at this time. Completely different.

viagra online said...

I'd seen a similar article in other blog, and I think is true, but scientists say that this trouble will be fixed around 20 years.