patrick.net

marcus says

Onvacation says

It just seems ridiculous to talk about a worldwide average temperature that can be measured down to hundredths of a degree.

When you average hundreds or thousands of numbers that are accurate to 1 or 2 tenths of a degree, you get a number that should probably be rounded to the nearest thousandth of a degree, for comparative purposes. Certainly at least to the one hundredths place.

Here's a mini example for you. Average 5 and 6 and you get 5.5. How can that happen when I'm averaging two integers ? Shouldn't the average be an integer too ?

Not sure why I'm explaining this to the same guy that thinks a brief cooling trend in the midst of a thirty year uptrend with higher highs and higher lows (thats the definition of an uptrent - the "higher lows" refer to the lows of down swings, which by definition occur in an uptrend)

marcus says

Not sure why I'm explaining this

Me neither.

Please link to the calculations where thousands of thermometers are used to find the average year long temperature for the entire world with accuracy down to hundredths of a degree for the last century.

Onvacation says

Please link to the calculations where thousands of thermometers are used to find the average year long temperature for the entire world with accuracy down to hundredths of a degree for the last century.

Even if you were averaging 100 numbers that were accurate to plus or minus on tenth of a degree, you would get an average that is accurate to one hundredth of a degree. The fact that this is hard for you to grasp is consistent with the fact that you think a downswing within an uptrend constitutes a significant downtrend.

That is, unless they always rounded up on the readings. But even then for year to year comparison purposes (considering differences from year to year) they would be accurate to the one hundredths place.

But hey, it takes all kinds as they say. Maybe there's an audience with a similar ummm, perspective (let's call it) to yours, that will appreciate your insights.

marcus says

Even if you were averaging 100 numbers that were accurate to plus or minus on tenth of a degree, you would get an average that is accurate to one hundredth of a degree.

Assuming you were measuring the same thing and you weren't taking the temperature of the water in a bucket hauled up to the rail of a whaling ship in the 1890 arctic.

marcus says

The fact that this is hard for you to grasp

Like I said, when the alarmist have no answers they go personal.

Still waiting for a link to
Onvacation says

calculations where thousands of thermometers are used to find the average year long temperature for the entire world with accuracy down to hundredths of a degree for the last century.

But suspect I will get more personal attacks.

Onvacation says

marcus says

The fact that this is hard for you to grasp

Like I said, when the alarmist have no answers they go personal.

100%

The process is fundamental to the debate. Is there a link summarizing how these measurements are collected?

marcus says

Even if you were averaging 100 numbers that were accurate to plus or minus on tenth of a degree, you would get an average that is accurate to one hundredth of a degree.

I will play.

So if I took a thousand readings from all over the globe that were acurate down to 1/10 of one degree and added them together and then divided the sum by one thousand we would get the average temp of all those places down to hundredths of a degree? What does that even mean? Does anyone have a link to the methodology of measuring average global temperature?
The illogic of the alarmists is religious.

Onvacation says

Like I said, when the alarmist have no answers they go personal.

No, I was honestly surprised.

Maybe you're just TPB under a different name, and you're trolling. You think it's a victory if you can get me to explain something simple to you. But okay, if that's the case you win.

Here's an example, simpler, but the same concept. Say you use a random number generator to generate numbers between 6 and 7 inclusive, accurate to the tenths place. Suppose these are true random numbers, 6.0, 6.1, 6.2, 6.3, 6.4, ........6.9, 7.0

But you always round to the 6 or seven, whichever is closest. If it's 6.5, you flip a coin, heads you call it 6, tails you call it 7. You do this one thousand times, using true random numbers to generate the tenths place.

So you have one thousand numbers and each is either 6 or 7 due to rounding by as much as .5., you add these 6 and 7s together and divide by 1000, getting an average very close to 6.5, give or take a few hundredths. Which will be extremely close to the actual average of the original numbers between 6 and 7, without rounding them.

How can that be ? If all the numbers were 6s and 7s, how can I get an average that so much closer to the actual average than 6 or 7 are (the actual inputs in to the average) ?

Or how about this. Say you have two groups of people, and each group has 1000 people in it. The heights of all the people are measured carefully, always rounded to the nearest inch. If it looks exactly half way in between 2 inch values, to the extent they can't say, then they round up. This procedure is done the same way by the same people using the same devices to measure.

So say that the average for one group is .41 inches higher than the other. Do you think the one average really is higher than the other ? That is, that the average height of the one group really is .41 inches taller than the other? Even though the resolution of the measurement was to the nearest inch ?

Yes or no ?

marcus says

Yes or no ?

So no links to the Onvacation says

calculations where thousands of thermometers are used to find the average year long temperature for the entire world with accuracy down to hundredths of a degree for the last century.

?

Aphroman says

What is the optimal temperature for humans.

I like 72 degrees Fahrenheit.

Onvacation says

Aphroman says

What is the optimal temperature for humans.

I like 72 degrees Fahrenheit.

Necrophiliac?

HeadSet says

Necrophiliac?

I like it hot. Global cooling is not good for humans. Where would you rather live, tropics or tundra?

______________
I still see a danger and a risk. I don't have to have absolute certainty about the exact magnitude of the danger in order to believe policies are justified.
______________
And that's the danger of "justification." We can ignore any findings because the "science is settled." Then the narrative becomes a belief system like religion. Outsiders are labeled as Deniers and dismissed. I may be dismissed, but I still have a vote.

The graph above is indicative of this trend - flawed data that is used to promote the narrative. When NOAA, for example, stops "correcting" old data to match the narrative, then I will begin believing the findings.

Read Crichton's 2004 book "State of Fear" - a great novelization of the AGW sham (with several pages of bibliography - in a novel!!) In the back there is an appendix chapter of how science becomes politicized. Oh, you didn't know several states in the USA had a policy of forced sterilization for "morons" and some mentally ill groups?? It happened. All the "smart" folks of the day believed it back in the 20's and 30's. It's happening now with AGW/Global warming.

Onvacation says

HeadSet says

Necrophiliac?

I like it hot. Global cooling is not good for humans. Where would you rather live, tropics or tundra?

This "Necrophiliac" quip was joke, a play on words. "Optimal temperature for humans" can interpreted to mean the optimal climate temperature, or the optimal body temperature. I think most people prefer an outside temp of 72 degrees over freezing, but one who prefers someone with body temperature of 72 degrees would be the Necro.

HeadSet says

This "Necrophiliac" quip was joke,

I got it.
I still like it hot.

marcus says

Even if you were averaging 100 numbers that were accurate to plus or minus on tenth of a degree, you would get an average that is accurate to one hundredth of a degree.

No, that is incorrect. To calculate an average of measured values, the correct method is (1) to add the values and then (2) to divide by the number of measurements. The sum "can contain no more decimal places than the least precise measurement." The quotient "should have the same number of significant figures as the quantity having the least significant figures entering into the calculation." The precision of the measurements thus limits the precision of the average.

marcus says

The fact that this is hard for you to grasp is consistent with the fact that you think a downswing within an uptrend constitutes a significant downtrend.

If this is hard for you to grasp, then please ask a math teacher to explain it to you. Getting it wrong is consistent with not actually being a math teacher IRL.

P.S. I see that marcus is now the only PatNet user who ignores me. Some people prefer ignorance, which they consider bliss. Any other user may feel free to copy and paste this comment so that marcus will see it, thus replacing blissful ignorance with the painful knowledge of having been proved wrong yet again.

marcus says

CBOEtrader says

uggesting multi-trillion dollar worldwide initiatives

That's a little extreme.

"[T]he developed world will provide $100 billion a year," NPR's Christopher Joyce reports.

But that amount is identified as a "floor," not a ceiling.

"Developed countries won inclusion of language that would up the ante in subsequent years," he explains, "so that financial aid will keep ramping up over time."

marcus says

But it does deny certain big interests some of their near and medium term profits.

These costs would be in addition to the $100bn+/year direct transfers described above.

marcus says

But you have to realize that investing a lot of money in 4th or 5th generation nuclear, possibly thorium, or other cutting edge and fairly efficient forms of enerrgy, and giving the fossil fuels a rest, is probably good for humanity regardless of the impact on AGW.

The people who used to live in or near Fukushima might disagree with you about that.

bob2356 says

What is your backup plan if man made co2 is the problem and spirals out of control?

Which major party or parties in the USA, or anywhere for that matter, have a plan actually to manage the climate?

In answering this question, please consider:
a) multiple factors can affect the climate, and climate change reflects the net effect of all these factors;
b) any viable solution would need to manage the results, rather than merely addressing one alleged component of the process.

Democrats tend to value stated intent over results. Micro-managing CO2 would make little or no practical difference, but Democrats present it as a way of signaling good intent, and then pretend it can somehow stop the climate from changing, even though nobody who looks at the long history of climate change could really agree with that pretense.

Republicans deny the whole thing, and that process might seem selfish or dishonest, but it prevents the worse results that Democrats campaign on.

I would like to see a major party in the USA or anywhere campaign on a serious proposal to gain control of the climate, but I have yet to see anything like that. So, again, if you know of any examples, please list them.

curious2 says

marcus says

Even if you were averaging 100 numbers that were accurate to plus or minus on tenth of a degree, you would get an average that is accurate to one hundredth of a degree.

No, that is incorrect. To calculate an average of measured values, the correct method is (1) to add the values and then (2) to divide by the number of measurements. The sum "can contain no more decimal places than the least precise measurement." The quotient "should have the same number of significant figures as the quantity having the least significant figures entering into the calculation." The precision of the measurements thus limits the precision of the average.

curious2 says

No, that is incorrect. To calculate an average of measured values, the correct method is (1) to add the values and then

MY statement of accuracy to 100th may be an exaggeration, but it obviously is more accurate, as long as the vlaues are "accurate" meaning just as likely to be obove the true value as below. See your own links description of accuracy versus precision. And even if not accurate, but ever so slightly skewed one way, the year to year comparative differences will be more precise than the initial measurements, as long as the same bias is there each year.

My example above should make this clear.

marcus says

Here's an example, simpler, but the same concept. Say you use a random number generator to generate numbers between 6 and 7 inclusive, accurate to the tenths place. Suppose these are true random numbers, 6.0, 6.1, 6.2, 6.3, 6.4, ........6.9, 7.0

But you always round to the 6 or seven, whichever is closest. If it's 6.5, you flip a coin, heads you call it 6, tails you call it 7. You do this one thousand times, using true random numbers to generate the tenths place.

So you have one thousand numbers and each is either 6 or 7 due to rounding by as much as .5., you add these 6 and 7s together and divide by 1000, getting an average very close to 6.5, give or take a few hundredths.

In this example, the numbers used were the nearest integer, and yet the average comes out extremely close to 6.5. This same effect holds for anything that is measured "accurately," see the bullseye description of precision versus accuracy in your linked article. As for the significant figures in an average, this rule makes sense in being cautious and where the number of individual data point may be relatively small, and when they may not be precise.. If the measures are inaccurate (biased) but in the same way, then of course the average doesn't improve the accuracy of measurement. But if say the measurement are innacurate, in the same direction (say due to always rounding up to the next 10th whenever the reading is too close .05 in between to tell to the next highest 10th). Then for comparative purposes, that is comparing 1000s of valuesone year to 1000s of values the next year, the difference is going to be accurate to a much to much less than one 10th of a degree, even though the measurements were done to the nearest tenth.

Common sense, no ? Use reasoning, and consider my integer example. The same thing is going to happen when the data values are always rounded to the nearest tenth.

MY initial response was about data from say 120 years ago, when they were using thermometers. The thermometers give readings in 10ths, but there is space in between the 10th marks leading to people using some sort of rounding strategy, for calling the temp to the nearest 10th.

If temp is read to the nearest 10th a true temp of 63.337 is more likely to be read as 63.3 than 63.4. In fact, the likelihood is such that with thousands of repeated measurements, guess what the average is going to be ? Maybe not exactly 63.34. But far closer to this than 63.3

I'm just using reasoning here. Maybe you can find a better web site to make your argument ?

marcus says

MY initial response was about data from say 120 years ago, when they were using thermometers. The thermometers give readings in 10ths, but there is space in between the 10th marks leading to people using some sort of rounding strategy, for calling the temp to the nearest 10th.

The thermometers were on whaling ships!
Does anyone really think you can get hundredths, tenths, or even full degree accuracy from readings taken from a bucket of water hauled up to the rail of a sailing ship?
And if these readings were so accurate why does Hansen et al have to "adjust" these temperatures in order to "prove" the globe is heating out of control.

marcus says

.If the measures are inaccurate (biased) but in the same way, then of course the average doesn't improve the accuracy of measurement. But if say the measurement are innacurate, in the same direction (say due to always rounding up to the next 10th whenever the reading is too close .05 in between to tell to the next highest 10th). Then for comparative purposes, that is comparing 1000s of valuesone year to 1000s of values the next year, the difference is going to be accurate to a much to much less than one 10th of a degree, even though the measurements were done to the nearest tenth.

Common sense, no ? Use reasoning, and consider my integer example.
***
In fact, the likelihood is such that with thousands of repeated measurements, guess what the average is going to be ?
***
I'm just using reasoning here. Maybe you can find a better web site to make your argument ?

Asking me questions is inconsistent with pretending to ignore me.

Your comment quoted above seems also to show ignorance of the relevant difference between counting and measuring. The integer example is inapposite because a hypothetical list of integers contains, by definition, exact numbers: each infinitely accurate and precise. In contrast, a list of measurements is necessarily a list of approximations. If you can grasp these discrete concepts, then your comment seems to ignore the fact that math applies different rules to averaging each.

To the extent your attempts at sentences quoted above are intelligible, they seem to be trying to make an argument similar to James Surowiecki's The Wisdom of Crowds. That argument reasons inductively from observations that, for example, if you ask many people to guess how many jelly beans are in a jar, then the average of their guesses may tend to be more accurate than asking a single expert.

Please visit the nearest GED program and ask a math teacher to explain the difference between counting and measuring, and the implications with respect to accuracy and precision when calculating an average. If you believe the rules of mathematics should change to suit your "reasoning," or better yet The Wisdom of Crowds, then ask for help in phrasing and suggesting that change. Also, ask an English teacher for help in understanding the word "ignore," and the resulting words "ignorance" and "ignorant." You asked me for a website, so here is one that might help:

https://www.wikihow.com/Get-a-GED

GED programs are inexpensive, and it is not too late to pursue a real diploma and maybe someday live IRL the online fantasy of becoming a math teacher.

Meanwhile, since I have now answered your question, perhaps you can try answering mine. Otherwise, I don't see the point of answering your questions when you prefer ignorance.

curious2 says

prefer ignorance.

Alarmist pretend to be able to measure the "worldwide average temperature" down to hundredths of a degree. The reality is the numbers have been adjusted so much that they don't reflect reality.

Catastrophic Manmade Global warming is the biggest fraud ever perpetuated on science. Only the ignorant still believe in it.

curious2 says

our comment quoted above seems also to show ignorance of the relevant difference between counting and measuring. The integer example is inapposite because a hypothetical list of integers contains, by definition, exact numbers: each infinitely accurate and precise. In contrast, a list of measurements is necessarily a list of approximations. If you can grasp these discrete concepts, then your comment seems to ignore the fact that math applies different rules to averaging each.

And your comment seems to ignore the effect of sample size on statistics.

curious2 says

Democrats tend to value stated intent over results. Micro-managing CO2 would make little or no practical difference, but Democrats present it as a way of signaling good intent, and then pretend it can somehow stop the climate from changing, even though nobody who looks at the long history of climate change could really agree with that pretense.

So, you are the expert and can state categorically that a reduction in CO2 would have no effect on climate?

LeonDurham says

So, you are the expert and can state categorically that a reduction in CO2 would have no effect on climate?

I would categorically state that the supposed increases in CO2 have had no noticeable impact on climate, so I would infer that reducing CO2 would have no noticeable effect.

Please stop trying to put the skeptics on the defensive because the models aren’t panning out in the alarmist favor.

Malcolm says

I would categorically state that the supposed increases in CO2 have had no noticeable impact on climate, so I would infer that reducing CO2 would have no noticeable effect.

Maybe you aren't paying enough attention then. There has definitely been a noticeable impact.

http://www.greenfacts.org/en/climate-change-ar4/l-3/4-observed-impacts.htm

LeonDurham says

And your comment seems to ignore the effect of sample size on statistics.

No, that assertion is also incorrect, because sample size affects a different aspect of statistics: margin of error. The margin of error depends on standard deviations from the mean (mean = average in this context). That's college level statistical math, while Marcus was failing to grasp the mean itself, which is school (or GED) level arithmetic math.

jazz_music says

doing something to save themselves

Adding emotional intensity does not make a false claim true. I am still waiting for someone to answer my question above. Otherwise this whole thread is yet another example of everybody talking about the weather and nobody doing anything about it.

LeonDurham says

There has definitely been a noticeable impact.

So how much has the temperature risen as the co2 has doubled?

Onvacation says

So how much has the temperature risen as the co2 has doubled?

I'm not going through this with you again. Your questions have been asked and answered a million times on here. You don't believe the measurements anyway, so what's the point in showing you the rise?

jazz_music says

Oh bullshit!

A target can be established with arbitrary accuracy.

Translation please?

jazz_music says

The discussion of significant figures

Is very significant when your measuring hundredths of one degree.

Onvacation says

Is very significant when your yalking of hundredths of one degree.

No--it's rightly analyzed statistically. Not with sig figures. We're trying to understand if there's a significant difference in the mean temperature in two populations of data.

jazz_music says

In fact, the scientists who prostitute themselves for oil companies

You don't have to work for an oil company to see that the earth is not warming and the sea level is not rising catastrophically.

LeonDurham says

No--it's rightly analyzed statistically. Not with sig figures. We're trying to understand if there's a significant difference in the mean temperature in two populations of data.

That comment makes no sense. A person unable to state either of two means would consequently be unable to state the difference (if any) between them. Analyzing statistically significant differences (if any) depends on the prerequisite predicate: measuring the values and calculating the means.

Otherwise you would be trying to istall the roof of a house before building the foundation and framing. You can shape the roof any way you like, but it won't stand up until you have the necessary support in place.

That sounds like a lot of the "debate" on this topic. Too many people start with the result that they want, and then fudge (or "model") whatever they need to support it. Many get so emotionally overwrought that they fail to suggest anything that might effectively enable managing the climate. Politicians and others cash in with proposals to transfer huge sums to their corrupt cronies, and ultimately themselves.

LeonDurham says

. You don't believe the measurements anyway, so what's the point in showing you the rise?

No. I don't believe that you can measure the worldwide average temperature at all, much less down to hundredths of one degree.
The "rise" is very little.

LeonDurham says

Your questions have been asked and answered a million times on here.

That's CAGW level hyperbole!

curious2 says

That comment makes no sense. A person unable to state either of two means would consequently be unable to state the difference (if any) between them. Analyzing statistically significant differences (if any) depends on the prerequisite predicate: measuring the values and calculating the means.

Come on now. This isn't that hard. Look up t-test or paired t-test.

Of course you need to calculate the means of the populations. Why would you think that I am saying otherwise?

Onvacation says

No. I don't believe that you can measure the worldwide average temperature at all, much less down to hundredths of one degree.
The "rise" is very little.

So why would you ask for someone to show you the measurements then? That is just being purposely difficult.

LeonDurham says

So why would you ask for someone to show you the measurements then?

Alarmists claim to have them?