AuthorTopic: Climate Science Questions  (Read 29178 times)

Offline Palloy

  • Sous Chef
  • ****
  • Posts: 3754
    • View Profile
    • https://palloy.wordpress.com
Re: Climate Science Questions
« Reply #90 on: January 27, 2015, 07:28:21 PM »
Quote
I'll show you mine if you show me yours.

Or how about YOU show us what YOU want to say, instead of challenging everybody else to do a lot of work first.  And it would help if you say where you charts are plucked from, and give a link to the dataset.

Quote
... the beauty about data is that it is agnostic ...

You clearly didn't look at the dataset underlying the Greenland chart, instead you simply assumed Lappi knew what he was doing with the data.  But the trick lies in not charting the data properly.

Same with the Idaho data - which I assume was correct.  But the data had nothing to do with the argument you were making, T(Min) and T(Max) not being T(mean).
The State is a body of armed men

Offline MKing

  • Contrarian
  • Sous Chef
  • *
  • Posts: 3354
    • View Profile
Re: Climate Science Questions
« Reply #91 on: January 27, 2015, 08:24:43 PM »
Quote
I'll show you mine if you show me yours.

Or how about YOU show us what YOU want to say, instead of challenging everybody else to do a lot of work first. 

Well....okay.

Here are the distributions of average daily temperature, by year, for Aberdeen Idaho.



Here is the distribution of the means of all those years, with the top four fits based on the Akaike Information Criterion (AIC).



So when people proclaim they have confidence intervals of fractions of 1F or 1C, I am stunned at such certainty when a 90% confidence interval based on the empirical evidence covers a range of 3.36C. Obviously just for a single station, but depending on how the uncertainty propagates, it does become a bit sketchy to pretend high levels of certainty when you can't even find it at higher levels of data resolution.

I should mention that including the TOBS correction caused the range to widen up and become nonsensical in certain cases, and I am currently investigating whether or not that effect is real, or just my lack of knowledge about what is considered appropriate use of the correction within the given data.

I currently have about 60 of the longest temperature records in Idaho to test for the validity of assumptions of correlation across large areas, because for some strange reason I can't find the correlation matrix between stations provided by NOAA, or anyone else for that matter, for the stations within an area, even though folks certainly say things are correlated all the damn time. I can find the ASSUMPTIONS of correlation at the macro scale, but for some reason folks don't want to demonstrate the validity of their assumptions with the actual data...preferring instead to get all artsy-fartsy with the central limit theorem, and to be honest, that one isn't always holding up all that well either.

Let me know if you wish to discuss your method of calculating the range of means using the assumption of fully independent distributions, as I did here, as compared to perfectly independent and perfectly positively correlated distributions.

The data for the Aberdeen Idaho station (GHCND:USC00100010) was obtained from here:

http://www.ncdc.noaa.gov/cdo-web/datatools/findstation

A free software package to do statistical analysis can be found here:

http://www.r-project.org/

and after that it is just your level of experience with exploratory data analysis I suppose. A common textbook on the topic can be found here:

http://www.amazon.com/Exploratory-Data-Analysis-John-Tukey/dp/0201076160

Tukey rocks by the way.




Sometimes one creates a dynamic impression by saying something, and sometimes one creates as significant an impression by remaining silent.
-Dalai Lama

Offline Palloy

  • Sous Chef
  • ****
  • Posts: 3754
    • View Profile
    • https://palloy.wordpress.com
Re: Climate Science Questions
« Reply #92 on: January 28, 2015, 01:58:04 PM »
I don't know what you are trying to prove here - I think we may be talking at cross-purposes.  I said:
The confidence level in the trends is certainly better than +/- 0.1 °F, so what's with this +/- 8°C?

You haven't tested the confidence level in the trends, but the confidence level in the means of individual years.
Since one big storm over the Arctic, at the time of year when the sea ice is breaking up, can break up and disperse a lot more ice, which can have cascading effects on the jet stream meanderings, which can have "polar vortex" effects on Idaho, I think the variation in the means of individual years could be measuring mostly weather effects, not climate change.  This is only a guess though.

Quote
... for some strange reason I can't find the correlation matrix between stations provided by NOAA ...

I think this might be because the stations are divided into climate divisions, not States. Idaho is split into 10 divisional types, with some types found in more than one area. http://www.esrl.noaa.gov/psd/data/usclimdivs/data/map.html#Idaho
The State is a body of armed men

Offline Petty Tyrant

  • Cannot be Saved
  • Sous Chef
  • *
  • Posts: 4573
    • View Profile
Re: Climate Science Questions
« Reply #93 on: January 28, 2015, 03:26:21 PM »
And so is accuracy. David Lappi  was a notorious climate change denier, now deceased. The original for this graph appears to be from Joanne Nova's website,  which also has a notorious anti-climate change bias.

Fortunately it relies on data. Chuck out the current information if you'd like, but the beauty about data is that it is agnostic to your implication of an ad hom just because of where such data resides.

Another example being Kobashi's work, and my challenge to go calculate the mean and confidence interval of a single station if you want to prove to yourself...using data..of the magnitude of the variation we are talking about. But trying to discredit information by blaming its location on the web...please...that is just so...Global Research-ish.

Its not the location he criticised, its the fact the graph used data inconsistently, the way you could to plot a chart suggesting a man grew steadily for 90 years until he died. compress the last 70 years into a 1 year interval. Honest science does not do that.
« Last Edit: January 28, 2015, 03:35:31 PM by Uncle Bob »
ELEVATE YOUR GAME

Offline MKing

  • Contrarian
  • Sous Chef
  • *
  • Posts: 3354
    • View Profile
Re: Climate Science Questions
« Reply #94 on: January 28, 2015, 07:38:03 PM »
I don't know what you are trying to prove here - I think we may be talking at cross-purposes.  I said:
The confidence level in the trends is certainly better than +/- 0.1 °F, so what's with this +/- 8°C?

You haven't tested the confidence level in the trends, but the confidence level in the means of individual years.

Darn right. How is anyone supposed to determine statistical significance of a trend if you haven't figured out the natural variation the trend is hiding within? You see, if you claim a mean trend of +0.3C or something (the signal), it becomes a bit unbelievable if the uncertainty around the mean from which the trend is drawn is +/- 5C (the noise). And we are talking about MEANS here, not as though I am even trying to incorporate the full range of variability available down at the daily or monthly level. I am doing that right now, part of the data necessary to determine actual correlation across grids of various sizes, while accurately keeping track of the full distribution. Computer took about 4 hours this afternoon to run that one, 36,500 distributions in that dataset, haven't even figured out how to display the thing on forums yet.

Chicken and egg kind of problem really.

How does uncertainty of natural variability propagate up into highly certain estimates? And what assumptions are necessary to make the full range of uncertainty disappear along the way? And how does THAT match up for empirical observations?

Just one of those interesting questions you ask when trying to figure out what the climate folks are up to.

Quote from: Palloy
Since one big storm over the Arctic, at the time of year when the sea ice is breaking up, can break up and disperse a lot more ice, which can have cascading effects on the jet stream meanderings, which can have "polar vortex" effects on Idaho, I think the variation in the means of individual years could be measuring mostly weather effects, not climate change.  This is only a guess though.

Guesses are as reasonable a place to start as any. But the pieces I am looking for are quite a bit more fundamental.

Quote from: Palloy
Quote
... for some strange reason I can't find the correlation matrix between stations provided by NOAA ...

I think this might be because the stations are divided into climate divisions, not States. Idaho is split into 10 divisional types, with some types found in more than one area. http://www.esrl.noaa.gov/psd/data/usclimdivs/data/map.html#Idaho

So now we are talking about aggregations on top of aggregations....which means that any correlation matrix would be at quite bit higher level than I am looking for. You see, those mean answers everyone likes? Depending on the correlations at higher levels of data resolution...they can CHANGE. Moving to higher aggregations signifies just that many more assumptions on top of the aggregations and correlations and interpolations and extrapolations...and just that many more assumptions that someone better had backed up with empirical observations.

So empirical observations seemed like a good place to start!!

Would you like to venture a guess on how much the range of uncertainty around the mean expands in THIS station depending on whether or not you assume complete independence between years (as I did), or perfect positive correlation?
Sometimes one creates a dynamic impression by saying something, and sometimes one creates as significant an impression by remaining silent.
-Dalai Lama

Offline JRM

  • Sous Chef
  • ****
  • Posts: 3190
    • View Profile
Re: Climate Science Questions
« Reply #95 on: January 29, 2015, 01:10:08 PM »
Is the world's ice (e.g., glaciers and ice caps) melting at an accelerating rate in recent decades, or isn't it?

Why is it melting more rapidly lately than a hundred years ago?
My "avatar" graphic is Japanese calligraphy (shodō) forming the word shoshin, meaning "beginner's mind". --  http://en.wikipedia.org/wiki/Shoshin -- It is with shoshin that I am now and always "meeting my breath" for the first time. Try it!

Offline azozeo

  • Master Chef
  • *****
  • Posts: 9469
    • View Profile
Re: Climate Science Questions
« Reply #96 on: January 29, 2015, 01:26:05 PM »
Is the world's ice (e.g., glaciers and ice caps) melting at an accelerating rate in recent decades, or isn't it?

Why is it melting more rapidly lately than a hundred years ago?

The earth's core is becoming more active. When that spinning ball of plasma
becomes more active it heats up. When it heats up, the surface of the planet
shows the effects. The glacial remains will melt. Simple stuff, hope this helps.
Why is Earth's core more active ? It is receiving more energy from the Sun.
You may want to follow along with this channel on youtube for Earth/Sun
activity.

https://www.youtube.com/user/killerazazello

Glad your still hangin' with us JRM.
« Last Edit: January 29, 2015, 01:29:35 PM by azozeo »
I know exactly what you mean. Let me tell you why you’re here. You’re here because you know something. What you know you can’t explain, but you feel it. You’ve felt it your entire life, that there’s something wrong with the world.
You don’t know what it is but its there, like a splinter in your mind

Offline Palloy

  • Sous Chef
  • ****
  • Posts: 3754
    • View Profile
    • https://palloy.wordpress.com
Re: Climate Science Questions
« Reply #97 on: January 29, 2015, 02:17:45 PM »
Quote
MKing: How is anyone supposed to determine statistical significance of a trend if you haven't figured out the natural variation the trend is hiding within? You see, if you claim a mean trend of +0.3C or something (the signal), it becomes a bit unbelievable if the uncertainty around the mean from which the trend is drawn is +/- 5C (the noise).

We must be homing in on why we are at cross-purposes, because I still don't get what you are worried about.
Sure the T(annual mean) data has a lot of noise in it, and T(daily mean) would have a lot more, but so what?
As long as the trend type chosen (linear, sum of differences or sum of square of differences) is sensible for the dataset being studied,
and the number of data points in the data set is sufficiently large, which it is,
I don't see how those evil climate scientists could be putting a spin on the trend.
Why is it "a bit unbelievable"?

I can enter what I want at the NOAA website, but it doesn't have a "Download" button or an "add to cart" button, so I am still without the data.
The State is a body of armed men

Offline MKing

  • Contrarian
  • Sous Chef
  • *
  • Posts: 3354
    • View Profile
Re: Climate Science Questions
« Reply #98 on: January 29, 2015, 04:52:56 PM »
Is the world's ice (e.g., glaciers and ice caps) melting at an accelerating rate in recent decades, or isn't it?

Why is it melting more rapidly lately than a hundred years ago?

Good questions. Perhaps because when the Little Ice Age ended, the world did what it does after a cooling cycle? Which is the opposite of what it does after a warming cycle? Wouldn't that be funny if modern instrumentation had been invented right at the peak of the MWP? We would all be standing around screaming about the coming ice age, and how we can geoengineer the planet to be warmer. But instead, we invented all the cool instrumentation on the other side of the cycle instead. So now we have been able to really, really, really carefully watch global warming not bother this particular temperature station in Aberdeen Idaho much.
Sometimes one creates a dynamic impression by saying something, and sometimes one creates as significant an impression by remaining silent.
-Dalai Lama

Offline MKing

  • Contrarian
  • Sous Chef
  • *
  • Posts: 3354
    • View Profile
Re: Climate Science Questions
« Reply #99 on: January 29, 2015, 05:12:27 PM »
Is the world's ice (e.g., glaciers and ice caps) melting at an accelerating rate in recent decades, or isn't it?

Why is it melting more rapidly lately than a hundred years ago?

The earth's core is becoming more active.

How? More active implies energy input. Where might the energy input be coming from?

Quote from: azozeo
When that spinning ball of plasma
becomes more active it heats up.

Energy can neither be created nor destroyed, and things don't heat up just because they feel like it one afternoon.

Quote from: Azozeo
When it heats up, the surface of the planet
shows the effects.

And is it this same effect that has caused all the prior warmings, and was there some change in the core that then caused the accompanying coolings?

Quote from: azozeo
The glacial remains will melt. Simple stuff, hope this helps.

The glacial remains have been melting for quite some time. If this is all so matter of fact, why do you think folks are so excited about it?

Quote from: azozeo
Why is Earth's core more active ? It is receiving more energy from the Sun.
You may want to follow along with this channel on youtube for Earth/Sun
activity.

Yeah, Youtube videos as a substitute for coherent thought, no footnotes, references, etc etc, doesn't work for me much better than folks pretending to know something about on renewable resources without at least a cost of supply curve or some comprehension of hotellings rule.
Sometimes one creates a dynamic impression by saying something, and sometimes one creates as significant an impression by remaining silent.
-Dalai Lama

Offline MKing

  • Contrarian
  • Sous Chef
  • *
  • Posts: 3354
    • View Profile
Re: Climate Science Questions
« Reply #100 on: January 29, 2015, 05:38:45 PM »
Quote
MKing: How is anyone supposed to determine statistical significance of a trend if you haven't figured out the natural variation the trend is hiding within? You see, if you claim a mean trend of +0.3C or something (the signal), it becomes a bit unbelievable if the uncertainty around the mean from which the trend is drawn is +/- 5C (the noise).

We must be homing in on why we are at cross-purposes, because I still don't get what you are worried about.

I'm not worried about anything. Are you not familiar with exploratory data analysis, or forensic statistics?

I trying to establish A) the size and shape of natural variability in temperature data at a single station, B) the size and shape of the uncertainty around the mean that this natural variability leads to, C) the correlation between the movement of those means at multiple stations across a wide area, and D) assuming C) is true, the distance at which C) holds true.

Think of it like a semi-variogram, I am trying to find the kernel and sill for temperature.

I realize that this is all probably Climate 101 stuff to the climate guys, but amazingly I haven't been able to find such basic analysis, only the conclusions and results of heaps of correlations between huge blocks of area.

Quote from: Palloy
Sure the T(annual mean) data has a lot of noise in it, and T(daily mean) would have a lot more, but so what?

Calculate the uncertainty around the mean with, and without correlation, and the range of uncertainty changes. That matters when you are trying to find a signal of +0.3C and your confidence interval is +/- 5C. Or worse.

Quote from: Palloy
As long as the trend type chosen (linear, sum of differences or sum of square of differences) is sensible for the dataset being studied,
and the number of data points in the data set is sufficiently large, which it is,
I don't see how those evil climate scientists could be putting a spin on the trend.

The establishment of statistical significance comes before claims of validity.

Quote from: Palloy
I can enter what I want at the NOAA website, but it doesn't have a "Download" button or an "add to cart" button, so I am still without the data.

NOAA has a WONDERFUL way of collecting the data. After you find what you are interested in (I can provide the station ID for Aberdeen ID if you need it) you do add it to the cart, and then you go to the cart, choose what you might be interested in (I just collect temperature data) and provide an email address. They then email you a link when the product has been assembled into a CSV file, and provide you a link in the email. You then download your data.

Collected a 100MB download the other day to begin assembling a profile for Idaho using stations that have long consistent strings. Using 25+ stations, I now have distributions for every day for the past century for Idaho..can't figure out how to show 36,000+ distributions in one shot though. It's about 1 million data points, might have some software issues whenI try and do the entire Pac NW.
Sometimes one creates a dynamic impression by saying something, and sometimes one creates as significant an impression by remaining silent.
-Dalai Lama

Offline Palloy

  • Sous Chef
  • ****
  • Posts: 3754
    • View Profile
    • https://palloy.wordpress.com
Re: Climate Science Questions
« Reply #101 on: January 30, 2015, 09:47:55 PM »
Quote
I trying to establish A) the size and shape of natural variability in temperature data at a single station, B) the size and shape of the uncertainty around the mean that this natural variability leads to, C) the correlation between the movement of those means at multiple stations across a wide area, and D) assuming C) is true, the distance at which C) holds true.

Why is that interesting? 
It could be useful for estimating the temperature for places that don't have their own temperature measuring equipment, but why bother?
The State is a body of armed men

Offline MKing

  • Contrarian
  • Sous Chef
  • *
  • Posts: 3354
    • View Profile
Re: Climate Science Questions
« Reply #102 on: January 31, 2015, 09:08:05 AM »
Is the world's ice (e.g., glaciers and ice caps) melting at an accelerating rate in recent decades, or isn't it?

You got me. The main melting is already done, are you aware of the origins of the Great Lakes? THAT was melting, when Lake Connecticut burst through an ice damn and created Long Island? Have we seen ANYTHING like that? No way!! So why get all excited over piddly stuff?

Quote from: JRM

Why is it melting more rapidly lately than a hundred years ago?

Because.. when the Little ICE age ended.... the same thing happened that always happens after ICE ages end? You know...the planet warms...the planet cools...this has been going on for millions of years..these are referred to as interglacials...are you sure you haven't heard these before?

http://en.wikipedia.org/wiki/Interglacial
Sometimes one creates a dynamic impression by saying something, and sometimes one creates as significant an impression by remaining silent.
-Dalai Lama

Offline MKing

  • Contrarian
  • Sous Chef
  • *
  • Posts: 3354
    • View Profile
Re: Climate Science Questions
« Reply #103 on: January 31, 2015, 09:20:57 AM »
Quote
I trying to establish A) the size and shape of natural variability in temperature data at a single station, B) the size and shape of the uncertainty around the mean that this natural variability leads to, C) the correlation between the movement of those means at multiple stations across a wide area, and D) assuming C) is true, the distance at which C) holds true.

Why is that interesting? 

Are you familiar with the issue of significant figures and why they matter?

Quote from: palloy
It could be useful for estimating the temperature for places that don't have their own temperature measuring equipment, but why bother?

Natural variability has far more to do with those significant digits than it does "fill in the blank" interpolations used in climate science. Determining the correlation certainly might have use within "fill in the blank" exercises, but that isn't what I want it for.

Are you familiar with semi-variograms?
Sometimes one creates a dynamic impression by saying something, and sometimes one creates as significant an impression by remaining silent.
-Dalai Lama

Offline Palloy

  • Sous Chef
  • ****
  • Posts: 3754
    • View Profile
    • https://palloy.wordpress.com
Re: Climate Science Questions
« Reply #104 on: February 01, 2015, 04:08:27 PM »
You seem to be avoiding answering a perfectly straightforward question by pretending that your "significant digits" are somehow really important, when your original intention was to cast doubt on climate science with your trick chart.
The State is a body of armed men

 

Related Topics

  Subject / Started by Replies Last post
14 Replies
4676 Views
Last post July 07, 2014, 01:30:39 PM
by Randy C
10 Replies
1268 Views
Last post December 18, 2016, 06:10:07 PM
by RE
25 Replies
6406 Views
Last post June 06, 2017, 09:31:52 PM
by RE