| Pages:
1
2
3
4 |
LearnedAmateur
National Hazard
Posts: 513
Registered: 30-3-2017
Location: Somewhere in the UK
Member Is Offline
Mood: Free Radical
|
|
Quote: Originally posted by JJay  |
I'm probably not going to read all of that, but I do know that chlorine radicals are not incredibly destructive to ozone in all circumstances, and I
don't see offhand why they would be so incredibly destructive to ozone in the atmosphere unless they have nothing else to react with....
|
The thing is though, is that the hole in the ozone layer began and rapidly expanded when CFCs were first introduced in the 1930s and has since slowed
and actually began to shrink since 1996 when they’ve been pretty much banned across most of the world. The CFCs are able to rise to the ozone layer,
which as the name implies contains a lot more ozone than other atmospheric layers, so yeah there isn’t a huge number of things to preferentially
react with Cl• up there, plus ozone is very reactive so quite happily breaks down in its presence. Because the reactions are all facilitated by UV,
which is very intense at the OL compared to sea level, this also contributes heavily.
It’s an equilibrium like most other things, oxygen will undergo homolytic fission and the resulting radicals react with oxygen to create ozone,
which easily cleaves by UV irradiation to make more ozone from oxygen, so on and so forth. CFCs disrupt this equilibrium by decreasing ozone levels
and increasing oxygen, so more UV gets through to the ground, which means more harm for life on Earth.
In chemistry, sometimes the solution is the problem.
It’s been a while, but I’m not dead! Updated 7/1/2020. Shout out to Aga, we got along well.
|
|
|
JJay
International Hazard
Posts: 3440
Registered: 15-10-2015
Member Is Offline
|
|
| Quote: Originally posted by LearnedAmateur | | Quote: Originally posted by JJay |
I'm probably not going to read all of that, but I do know that chlorine radicals are not incredibly destructive to ozone in all circumstances, and I
don't see offhand why they would be so incredibly destructive to ozone in the atmosphere unless they have nothing else to react with....
|
The thing is though, is that the hole in the ozone layer began and rapidly expanded when CFCs were first introduced in the 1930s and has since slowed
and actually began to shrink since 1996 when they’ve been pretty much banned across most of the world. The CFCs are able to rise to the ozone layer,
which as the name implies contains a lot more ozone than other atmospheric layers, so yeah there isn’t a huge number of things to preferentially
react with Cl• up there, plus ozone is very reactive so quite happily breaks down in its presence. Because the reactions are all facilitated by UV,
which is very intense at the OL compared to sea level, this also contributes heavily.
It’s an equilibrium like most other things, oxygen will undergo homolytic fission and the resulting radicals react with oxygen to create ozone,
which easily cleaves by UV irradiation to make more ozone from oxygen, so on and so forth. CFCs disrupt this equilibrium by decreasing ozone levels
and increasing oxygen, so more UV gets through to the ground, which means more harm for life on Earth. |
*shrug* I'm not really sure about that: http://www.theozonehole.com/ozoneholehistory.htm
If you mix a tiny amount chlorine gas and a large amount of pure ozone and expose it to UV, perhaps each chlorine molecule destroys a large number of
ozone molecules... but how would the effect of the chlorine change by concentration? Also, if there's something for the chlorine radicals to latch
onto, they won't last long....
|
|
|
LearnedAmateur
National Hazard
Posts: 513
Registered: 30-3-2017
Location: Somewhere in the UK
Member Is Offline
Mood: Free Radical
|
|
Well CFCs aren’t the only contributors to ozone breakdown, halons and other halogenated organics are still in wide use and also deplete the layer.
It’s like any other chemical reaction, the higher the concentration of a reagent will increase the rate of reaction because collisions are more
likely, as per kinetic theory. Take the catalytic decomposition of H2O2 for instance, if you used a gram of MnO2 it will occur a lot quicker than
using 0.1g - if you have more chlorine radicals floating around in the ozone, the ozone will decompose quicker and being an equilibrium, ozone
production is effectively decreased hence there is less in the atmosphere. And yeah, they do come across molecules which can terminate the radical
steps, but as you can see in the mechanism I added above, Cl• acts as a catalyst as it is regenerated by heterolytic fission of chlorine peroxide.
In chemistry, sometimes the solution is the problem.
It’s been a while, but I’m not dead! Updated 7/1/2020. Shout out to Aga, we got along well.
|
|
|
RawWork
Hazard to Others
Posts: 167
Registered: 10-2-2018
Member Is Offline
|
|
So CFCs belong to global warming topic, but UV doesn't?
It's easy for any hater or angry person or mentally diseased to release CFCs into atmosphere and cause massive damage with little effort, correct?
What would be the best counterattack? Is there some chemical that would remove CFCs from the atmosphere, and not cause additional damage? Some that
will itself be removed too, or act neutral on ozone after removing CFCs? Consider that sun ionizes some chemicals.
[Edited on 29-3-2018 by RawWork]
|
|
|
DrP
National Hazard
Posts: 625
Registered: 28-9-2005
Member Is Offline
Mood: exothermic
|
|
We need facts and measured data - not opinions on the subject. Science right?
| Quote: Originally posted by RogueRose | | I know I'm not alone in saying that at least since about 2007 the sun has been A LOT brighter than it was before. There were times when it seemed
as much as 20-40% brighter than what it had been just a few years ago. |
Can you point us towards the meteorologists reports that back this up please? This news report says it's due to cleaner atmosphere in Europe and that
some parts of the world are actually dimmer. I am not sure how bright the sun appears to be in the sky has any effect on the amount of energy
absorbed and kept by the earth. It is about how much is retained and not allowed to escape - which atmospherics effect.
https://www.theguardian.com/news/2017/mar/21/a-bright-sun-to...
Local variations are irrelevant, obviously, when talking about GLOBAL warming. The majority of the world's scientists say different - please don't say
it's a conspiracy theory.
Regarding the CO2 stuff - I pretty certain you know the proposed mechanism how greenhouse gasses work and trap energy from the sun by changing the
wavelength of the incident rays so they can't escape.... we studied it at school back in the 1980's - you choose to ignore all that stuff or not
understand it?
\"It\'s a man\'s obligation to stick his boneration in a women\'s separation; this sort of penetration will increase the population of the younger
generation\" - Eric Cartman
|
|
|
LearnedAmateur
National Hazard
Posts: 513
Registered: 30-3-2017
Location: Somewhere in the UK
Member Is Offline
Mood: Free Radical
|
|
| Quote: Originally posted by RawWork | So CFCs belong to global warming topic, but UV doesn't?
It's easy for any hater or angry person or mentally diseased to release CFCs into atmosphere and cause massive damage with little effort, correct?
What would be the best counterattack? Is there some chemical that would remove CFCs from the atmosphere, and not cause additional damage? Some that
will itself be removed too, or act neutral on ozone after removing CFCs? Consider that sun ionizes some chemicals.
[Edited on 29-3-2018 by RawWork] |
Eh, kinda, CFCs can be considered greenhouse gases in themselves but aren’t as strong as CO2 and CH4 for example - increase in UV penetration (which
makes up about 4% of all wavelengths released by the Sun) doesn’t really contribute to global warming, infrared is the main concern which is almost
half the output (~45%), visible light being the rest but this isn’t absorbed as well by greenhouse gases.
Probably not an individual, like how amateur chemists don’t really do much environmental damage by improper waste disposal, it’s more the
collective masses and industry where thousands of tonnes of CFCs were released in the past years, due to consumerism. There isn’t really any
counterattack other than banning or restricting their usage, over time levels have dropped naturally due to radical termination and
absorption/degradation at the surface of the Earth.
In chemistry, sometimes the solution is the problem.
It’s been a while, but I’m not dead! Updated 7/1/2020. Shout out to Aga, we got along well.
|
|
|
mayko
International Hazard
Posts: 1218
Registered: 17-1-2013
Location: Carrboro, NC
Member Is Offline
Mood: anomalous (Euclid class)
|
|
| Quote: Originally posted by JJay |
Here, the error variance is correlated to the global temperature (which we don't know, but the correlation is still evident), and while I haven't run
any diagnostics, I doubt that happened by random chance.
There are ways of dealing with these things, but the heteroskedasticity suggests that converting the temperature readings to "temperature anomalies"
introduces a loss of information which might actually be helpful for constructing a model. |
Well.... yes, subtracting the baseline from the raw temperatures removes information about the baseline. This is often desirable, since it helps to
control for spatial or seasonal differences, which can be much larger than the temporal signal of interest. It's not obvious what sort of model you
want to build which would use those baselines (if you're testing UHI as a hypothesis, surely it would be more useful to include, say, population
density as an explanatory variable?).
It's also not obvious to me how using anomalies would introduce spurious changes in variance. Two alternate explanations for your observation, at
least as plausible, come immediately to mind:
* Earlier regional values are usually based on a smaller number of surface stations, and are thus less precise. The relatively noisier sampling of a
homoskedastic process could artificially inflate variance.
* The underlying process might, in fact, be heteroskedastic.
It's not clear to me what you're trying to do here. It looks like you're trying to build a global time series from gridded ones. If you really want to
reinvent the wheel, go ahead. Personally, unless I have a pressing reason to do otherwise, I'd stick with one of the globally averaged data sets that
are already been published, such as from GISTEMP, HadCRUT, or BEST. It's also not clear why you're normalizing by month; monthly anomalies are almost
always calculated from monthly baselines.
It may be that I've badly misunderstood you, that you're actually saying that the raw temperatures don't have constant variance, and that this makes
temperature anomaly an inappropriate tool. I don't see how that bird flies either.
Temperature anomalies are ubiquitous in climatology, and are in common use by better statisticians than me. If I had evidence that they introduced
artifacts or otherwise inappropriate, I'd be tripping over myself to publish, rather than posting on sciencemadness dot org.
How does any of this get us any closer to attributing regional trends to UHI?
al-khemie is not a terrorist organization
"Chemicals, chemicals... I need chemicals!" - George Hayduke
"Wubbalubba dub-dub!" - Rick Sanchez
|
|
|
mayko
International Hazard
Posts: 1218
Registered: 17-1-2013
Location: Carrboro, NC
Member Is Offline
Mood: anomalous (Euclid class)
|
|
| Quote: Originally posted by nitro-genes | If CO2 and temperature are showing an not completely known interaction, why did they use CO2 measurements from icecores as a temperature proxy in the
first place? Considering the inertness of noble gasses and it's temperature dependend solubility in ocean water (assuming equal salinity and volume),
icecore noble gas levels make at least an interesting comparison/control. According to this article, these data may provide a much more unbiased
temperature proxy than CO2.
https://www.nature.com/articles/d41586-017-08721-4
[Edited on 29-3-2018 by nitro-genes] |
Ice core analysis generally doesn't use CO2 as a temperature proxy, but rather oxygen isotope enrichment:
https://www.scientificamerican.com/article/how-are-past-temp...
That article does look neat, though, and I'm interested to see what comes of it; thanks for bringing it up!
al-khemie is not a terrorist organization
"Chemicals, chemicals... I need chemicals!" - George Hayduke
"Wubbalubba dub-dub!" - Rick Sanchez
|
|
|
sodium_stearate
Hazard to Others
Posts: 255
Registered: 22-4-2011
Location: guard duty at the checkpoint
Member Is Offline
Mood: No mask.
|
|
The earth has been around for a very long time.
It has gone through many kinds of very long-term cycles
and will continue to do so long after humans have been
extinct for billions of years.
Right now, we humans are playing an active role in
the chemistry and the physics of our planet.
That makes sense when you take in to account the fact
that we are all made up of stuff that was here to start with.
It's just that currently, part of the process realizes what it's
doing.
"Opportunity is missed by most people
because it is dressed in overalls and it
looks like work" T.A. Edison
|
|
|
JJay
International Hazard
Posts: 3440
Registered: 15-10-2015
Member Is Offline
|
|
| Quote: Originally posted by mayko |
It may be that I've badly misunderstood you, that you're actually saying that the raw temperatures don't have constant variance, and that this makes
temperature anomaly an inappropriate tool. I don't see how that bird flies either.
|
That is exactly what I'm saying. If you don't know the variance, you don't know how anomalous the readings are; what you call an anomaly, I would call
an "error" or a "deviation" - an anomaly would probably be an unusually large error or deviation. Finding explanatory variables that account for that
variance is not sufficient for causal inference (ideally you'd use a controlled experiment), but it is necessary.
(Oh and the problem is that you can't efficiently estimate the variance from the errors if it is correlated with the measurements. I understand this
intuitively, but it is surely not hard to prove.)
[Edited on 29-3-2018 by JJay]
|
|
|
mayko
International Hazard
Posts: 1218
Registered: 17-1-2013
Location: Carrboro, NC
Member Is Offline
Mood: anomalous (Euclid class)
|
|
Walk me through this symbolically.
Given a raw temperature time series, T(t), an end goal is to decompose it into a linear combination of signals:
T(t) = F(t) + V(t) + C
where:
F(t) is net forced response, that is, the total change to the thermodynamic equilibrium of the earth system. As a special case, F(t)=0 in the absence
of long-term climate change (of any sort: solar, greenhouse, aerosol, etc).
V(t) is net unforced response; that which does not change the thermodynamic equilibrium. Examples include the perturbation of a volcanic eruption, or
the seesawing back and forth of heat between the ocean and the atmosphere. The long-term average of V(t) is 0; were it nonzero (for example, if
volcanic eruptions increased in frequency), then there would be a forced response unaccounted for in F(t), which has been defined as the net forced
response.
C is some constant representing the particulars of the record in question. The same 0.1 degree/decade warming signal plus noise is a very different
raw temperature time series in the Sahara (C ~ 20) than in Antarctica (C ~ -20).
To calculate an anomaly time series, we first calculate a baseline B from T. This is done by choosing a reference time period P and calculating the
average T(t) over P:
B = mean( T(t) for t in P)
B is a lot like a constant of integration; so long as you're consistent, choice of P doesn't really matter.
The anomaly time series is thus:
A(t) = T(t) - B = F(t) + V(t) + (C - B)
uhhhhhhhhh.... well, we know that A(t) is 'anomalous' by virtue of it being an 'anomaly', and we only need to know T(t) to know A(t). It sounds like
you think that anomalies are intended to be, by themselves, some sort of significance test, which they are not.
| Quote: |
what you call an anomaly, I would call an "error" or a "deviation"
|
You can do that if you want, but why would you and what would it matter? If you mean 'deviation from some reference value B', then yes, the terms are
isomorphic, but what utility do we get from insisting on your linguistic preferences over established jargon? If you mean 'deviation from some central
tendency', then no. The central tendency of T(t) is F(t) + C and the central tendency of A(t) is F(t) + (C - B). The deviation around both is V(t),
whose homoskedasticity is not really relevant here. You don't need to know V(t) to know A(t).
| Quote: |
an anomaly would probably be an unusually large error or deviation
|
No. "An anomaly" in this context would be A(t) at some particular t. That is what the word means here. You don't have to take it from me. If you don't like it, I'm sorry, but if it makes you feel better nobody asked me either.
| Quote: |
(Oh and the problem is that you can't efficiently estimate the variance from the errors if it is correlated with the measurements.
|
It sounds like you are saying that it is nontrivial in this case to infer V(t) from A(t) because V(t) is correlated with T(t).
We can talk about that if you want, and what it means in practice, and whether it supports the hypothesis that UHI rather than greenhouse forcing is
responsible for regional trends. But if it is indeed a problem, the problem stems from the raw temperatures T(t), not the process of calculating A(t).
Why would it be less trivial to infer V(t) from A(t) + B, compared to A(t), given that V(t) is equally correlated with both?
al-khemie is not a terrorist organization
"Chemicals, chemicals... I need chemicals!" - George Hayduke
"Wubbalubba dub-dub!" - Rick Sanchez
|
|
|
JJay
International Hazard
Posts: 3440
Registered: 15-10-2015
Member Is Offline
|
|
| Quote: Originally posted by mayko |
It sounds like you are saying that it is nontrivial in this case to infer V(t) from A(t) because V(t) is correlated with T(t).
|
No. Here's the concept: https://en.wikipedia.org/wiki/Efficiency_(statistics)
|
|
|
mayko
International Hazard
Posts: 1218
Registered: 17-1-2013
Location: Carrboro, NC
Member Is Offline
Mood: anomalous (Euclid class)
|
|
You are saying that it is inefficient in this case to estimate V(t) from A(t) because V(t) is correlated with T(t).
We can talk about that if you want, and what it means in practice, and whether it supports the hypothesis that UHI rather than greenhouse forcing is
responsible for regional trends. But if it is indeed a problem, the problem stems from the raw temperatures T(t), not the process of calculating A(t).
Why would it be more efficient to estimate V(t) from A(t) + B, compared to A(t), given that V(t) is equally correlated with both?
[Edited on 30-3-2018 by mayko]
al-khemie is not a terrorist organization
"Chemicals, chemicals... I need chemicals!" - George Hayduke
"Wubbalubba dub-dub!" - Rick Sanchez
|
|
|
JJay
International Hazard
Posts: 3440
Registered: 15-10-2015
Member Is Offline
|
|
| Quote: Originally posted by mayko |
You are saying that it is inefficient in this case to estimate V(t) from A(t) because V(t) is correlated with T(t).
We can talk about that if you want, and what it means in practice, and whether it supports the hypothesis that UHI rather than greenhouse forcing is
responsible for regional trends. But if it is indeed a problem, the problem stems from the raw temperatures T(t), not the process of calculating A(t).
Why would it be more efficient to estimate V(t) from A(t) + B, compared to A(t), given that V(t) is equally correlated with both?
[Edited on 30-3-2018 by mayko] |
I don't see any reason to assume that it is equally correlated with both. The definition that most people use for "anomaly" in data is more like this:
https://en.wikipedia.org/wiki/Anomaly_detection I think this is a reasonable definition of "anomaly." After all, if temperature two readings taken
at two different times are exactly the same, that's anomalous.
|
|
|
mayko
International Hazard
Posts: 1218
Registered: 17-1-2013
Location: Carrboro, NC
Member Is Offline
Mood: anomalous (Euclid class)
|
|
I am sorry to hear that you don't like the established jargon, but I doubt anyone reading is in a position to do anything about it, even were they so
inclined. More to the point, what the data are called has no bearing on how they were derived or their statistical properties. They could be called
potato numbers, but complaining that they don't taste good with ketchup wouldn't get you closer to attributing regional temperature trends to UHI.
| Quote: Originally posted by JJay | | Quote: | | Why would it be more efficient to estimate V(t) from A(t) + B, compared to A(t), given that V(t) is equally correlated with both?
|
I don't see any reason to assume that it is equally correlated with both. |
V(t) is equally correlated with A(t) and T(t) because A(t) and T(t) = A(t) + B only differ by a constant offset, B.
Maybe you are using "correlation" in some sense with which I'm not familiar, but for the most common definitions (eg, Pearson's coefficient,
Spearman's rank), it's trivial that (x1, x2, x3....) will be just as correlated with (y1, y2, y3....) as with (y1+k, y2+k, y3+k...), for any real
number k.
It's the variation that is correlated or uncorrelated between two data sets; adding or subtracting a constant offset doesn't add or subtract any
variation (because it's constant, not variable.)
al-khemie is not a terrorist organization
"Chemicals, chemicals... I need chemicals!" - George Hayduke
"Wubbalubba dub-dub!" - Rick Sanchez
|
|
|
JJay
International Hazard
Posts: 3440
Registered: 15-10-2015
Member Is Offline
|
|
You're assuming that A(t) is at a constant linear offset from T(t), and it's offset by a different value at each station. In other words, your
assumption is false.
|
|
|
AJKOER
Radically Dubious
Posts: 3026
Registered: 7-5-2011
Member Is Offline
Mood: No Mood
|
|
When looking at time series data of temperature, remember that the sun's irradiance fluctuates (the sun is alive, at least for a long time to come).
Sun spots, a measure of solar activity, has short term cyclic (averaging 11.1 years) properties that can introduce similar cyclic distortions in the
temperature data on earth, which is separate from other global warming trends.
Some of my prior comments on SM:
"there is an interesting correlation between sunspots and climate (see http://www-das.uwyo.edu/~geerts/cwx/notes/chap02/sunspots.ht... . To quote a few points:
"Sunspots and climate
Incidentally, the Sporer, Maunder, and Dalton minima coincide with the colder periods of the Little Ice Age, which lasted from about 1450 to 1820.
More recently it was discovered that the sunspot number during 1861-1989 shows a remarkable parallelism with the simultaneous variation in northern
hemisphere mean temperatures (2). There is an even better correlation with the length of the solar cycle, between years of the highest numbers of
sunspots". .....
Intuitively one may assume the that total solar irradiance would decrease as the number of (optically dark) sunspots increased. However direct
satellite measurements of irradiance have shown just the opposite to be the case. This means that more sunspots deliver more energy to the atmosphere,
so that global temperatures should rise.
Not only does the increased brightness of the Sun tend to warm the Earth, but also the solar wind (a stream of highly energetic charged particles)
shields the atmosphere from cosmic rays, which produce 14C (radioactive carbon 14). So there is more 14C when the Sun is magnetically quiescent. This
explains why 14C samples from independently dated material are used as a way of inferring the Sun's magnetic history.
Recent research (3) indicates that the combined effects of sunspot-induced changes in solar irradiance and increases in atmospheric greenhouse gases
offer the best explanation yet for the observed rise in average global temperature over the last century. Using a global climate model based on energy
conservation, Lane et al (3) constructed a profile of atmospheric climate "forcing" due to combined changes in solar irradiance and emissions of
greenhouse gases between 1880 and 1993. They found that the temperature variations predicted by their model accounted for up to 92% of the temperature
changes actually observed over the period -- an excellent match for that period. Their results also suggest that the sensitivity of climate to the
effects of solar irradiance is about 27% higher than its sensitivity to forcing by greenhouse gases."
My take on this is that if one is citing temperature data, one should at least adjust the data for the impact due to solar irradiance. I would also
adjust for known recent volcanic events, for example, Krakatoa, possibly using a dummy variable or a model based on volume of mass/SO2 ejected into
the upper atmosphere. To quote from Wikipedia on Krakatoa (link: http://en.m.wikipedia.org/wiki/1883_eruption_of_Krakatoa#Glo... ) :
"In the year following the eruption, average Northern Hemisphere summer temperatures fell by as much as 1.2 °C (2.2 °F).[9] Weather patterns
continued to be chaotic for years, and temperatures did not return to normal until 1888.[9] .....
The eruption injected an unusually large amount of sulfur dioxide (SO2) gas high into the stratosphere, which was subsequently transported by high
level winds all over the planet. This led to a global increase in sulfuric acid (H2SO4) concentration in high level cirrus clouds. The resulting
increase in cloud reflectivity (or albedo) would reflect more incoming light from the sun than usual, and cool the entire planet until the suspended
sulfur fell to the ground as acid precipitation.[14]"
Another possible variable would be the change in annual sales in millions of ozone degrading chemicals, possibly at a lag (more work but better, use a
percent contribution by group of chemicals rated by ability to attack ozone, and therefrom construct a potency adjustment factor). The estimated
coefficient times the lagged sales figure change should correlate to reported estimates of the change in size of the ozone hole. A functional
transformation to this area variable (like converting it to an ozone volume figure) may be required to obtain highest Pearson's correlation
coefficient, which is a measure of linearity.
--------------------------------------------------------------
Also, as heating from whatever sources go on, it is true that heat itself becomes a major issue at a point!
When is that?
Well, for instance, when the methane hydrate at the bottom of the Gulf of Mexico (and other places) hits a critical warming point to convert back to
gaseous methane!
CH4 is actually much more potent than CO2 gas (like 86 times) in creating atmospheric warming, but its half life in the atmosphere is less. Source,
see for example, https://www.scientificamerican.com/article/how-bad-of-a-gree... .
Still, that vast amount of CH4 release is going to be unpleasantly abrupt and nasty for the planet  .
Like a bad chemistry experiment gone totally rogue!
[Edited on 3-4-2018 by AJKOER]
|
|
|
JJay
International Hazard
Posts: 3440
Registered: 15-10-2015
Member Is Offline
|
|
Furthermore, the variance is calculated using a non-linear equation; even if the framework you suggested did not suffer from the shortcoming that I
pointed out earlier, if the variance is correlated with the measurements (this is *not* the same thing as assuming that the deviation is correlated
with the measurements), if using so-called "anomalies," it's still necessary to estimate a an offset parameter, which introduces an additional degree
of freedom to the model that must be estimated using non-linear techniques, introducing unnecessary bias and inefficiency.
[Edited on 3-4-2018 by JJay]
|
|
|
AJKOER
Radically Dubious
Posts: 3026
Registered: 7-5-2011
Member Is Offline
Mood: No Mood
|
|
If one is forecasting, my personal experience is that the more parsimonious a model is with good variables, the more improved accuracy of the
forecast!
Add lots of variables to impress people with your historical fit, but to really get a good forecast estimate, use but a few.
A good variable should be based on some accepted law, concept,... , and not random good/high correlation. If you understand how the variable is
contributing to the model, and the universe changes, judging the change impact on the variable aids in judging the continued likelihood of a good
forecast and/or its bias direction.
[Edited on 3-4-2018 by AJKOER]
|
|
|
mayko
International Hazard
Posts: 1218
Registered: 17-1-2013
Location: Carrboro, NC
Member Is Offline
Mood: anomalous (Euclid class)
|
|
| Quote: Originally posted by JJay | | You're assuming that A(t) is at a constant linear offset from T(t), and it's offset by a different value at each station. In other words, your
assumption is false. |
!?HUH?!
It isn't an 'assumption' that A(t) differs from T(t) by a constant; that's *trivially true*, *by construction*! If you disagree, I'm not sure what to
tell you, besides to look back at the definition, or maybe subtract A(t) from T(t) and see what you get.
In practice, the value of B will vary between station records (that's sort of the point), but that has no bearing on whether A(t) and T(t) are both
correlated with the variance of V(t).
| Quote: |
it's still necessary to estimate a an offset parameter, which introduces an additional degree of freedom to the model that must be estimated using
non-linear techniques, introducing unnecessary bias and inefficiency.
|
Accepting this momentarily for the sake of argument, in practice the difficulties of standardizing and combining records must surely be weighed
against the difficulties of using an unmerged collection of many non-standardized raw temperature records. Some of these have already been mentioned
here or in linked sites; here is another discussion from the perspective of global absolute temperature, which would presumably be implicit in any
global absolute temperature time series you produced:
http://www.realclimate.org/index.php/archives/2017/08/observ...
Again, if I had evidence of what you're claiming (that using anomalies instead of raw temperatures is inappropriate) I'd be licking stamps to get my
manuscript to GRL.
al-khemie is not a terrorist organization
"Chemicals, chemicals... I need chemicals!" - George Hayduke
"Wubbalubba dub-dub!" - Rick Sanchez
|
|
|
mayko
International Hazard
Posts: 1218
Registered: 17-1-2013
Location: Carrboro, NC
Member Is Offline
Mood: anomalous (Euclid class)
|
|
Not everything on that site is wrong, but it is a bit dated and several of the sources aren't valid anymore:
https://en.wikipedia.org/wiki/Eigil_Friis-Christensen#Solar_...
As I've mentioned, observed temperature increases are difficult to explain in terms of solar activity, given that the every solar trend has been going
the wrong direction to explain it, for the last several decades of the fastest heating. If we decided to "adjust" the temperature data, that is,
remove everything but the greenhouse signal, it would revise things upwards (surely it would be an extreme revision if it's true that "the sensitivity
of climate to the effects of solar irradiance is about 27% higher than its sensitivity to forcing by greenhouse gases"!)
Volcanic eruptions introduce real, but short-lived perturbations which don't change the ultimate trend.
It's often not necessary or informative to "adjust" the temperature data to account for these effects, but if you want to, sure, you can try to remove
known non-greenhouse and internal variabilities to try to isolate the greenhouse signal by itself. Here's an example:
http://iopscience.iop.org/article/10.1088/1748-9326/6/4/0440...
al-khemie is not a terrorist organization
"Chemicals, chemicals... I need chemicals!" - George Hayduke
"Wubbalubba dub-dub!" - Rick Sanchez
|
|
|
JJay
International Hazard
Posts: 3440
Registered: 15-10-2015
Member Is Offline
|
|
| Quote: Originally posted by mayko | | Quote: Originally posted by JJay | | You're assuming that A(t) is at a constant linear offset from T(t), and it's offset by a different value at each station. In other words, your
assumption is false. |
!?HUH?!
It isn't an 'assumption' that A(t) differs from T(t) by a constant; that's *trivially true*, *by construction*! If you disagree, I'm not sure what to
tell you, besides to look back at the definition, or maybe subtract A(t) from T(t) and see what you get.
In practice, the value of B will vary between station records (that's sort of the point), but that has no bearing on whether A(t) and T(t) are both
correlated with the variance of V(t).
| Quote: |
it's still necessary to estimate a an offset parameter, which introduces an additional degree of freedom to the model that must be estimated using
non-linear techniques, introducing unnecessary bias and inefficiency.
|
Accepting this momentarily for the sake of argument, in practice the difficulties of standardizing and combining records must surely be weighed
against the difficulties of using an unmerged collection of many non-standardized raw temperature records. Some of these have already been mentioned
here or in linked sites; here is another discussion from the perspective of global absolute temperature, which would presumably be implicit in any
global absolute temperature time series you produced:
http://www.realclimate.org/index.php/archives/2017/08/observ...
Again, if I had evidence of what you're claiming (that using anomalies instead of raw temperatures is inappropriate) I'd be licking stamps to get my
manuscript to GRL.
|
I'm not trying to be rude about this, but I have already explained three times why using temperature "anomalies" is not really a good idea, and I
didn't even get into non-linear links and so forth. If you can't follow the discussion so far, there's no point in going any further.
|
|
|
mayko
International Hazard
Posts: 1218
Registered: 17-1-2013
Location: Carrboro, NC
Member Is Offline
Mood: anomalous (Euclid class)
|
|
I can follow the discussion! I just want to understand how V(t) - A(t) is something other than B, which is constant. Can someone more patient than
JJay explain this to me? It sounds like he has a groundbreaking paper in geophysics to write!
al-khemie is not a terrorist organization
"Chemicals, chemicals... I need chemicals!" - George Hayduke
"Wubbalubba dub-dub!" - Rick Sanchez
|
|
|
AJKOER
Radically Dubious
Posts: 3026
Registered: 7-5-2011
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by mayko |
Not everything on that site is wrong, but it is a bit dated and several of the sources aren't valid anymore:
https://en.wikipedia.org/wiki/Eigil_Friis-Christensen#Solar_...
As I've mentioned, observed temperature increases are difficult to explain in terms of solar activity, given that the every solar trend has been going
the wrong direction to explain it, for the last several decades of the fastest heating. If we decided to "adjust" the temperature data, that is,
remove everything but the greenhouse signal, it would revise things upwards (surely it would be an extreme revision if it's true that "the sensitivity
of climate to the effects of solar irradiance is about 27% higher than its sensitivity to forcing by greenhouse gases"!)
Volcanic eruptions introduce real, but short-lived perturbations which don't change the ultimate trend.
It's often not necessary or informative to "adjust" the temperature data to account for these effects, but if you want to, sure, you can try to remove
known non-greenhouse and internal variabilities to try to isolate the greenhouse signal by itself. Here's an example:
http://iopscience.iop.org/article/10.1088/1748-9326/6/4/0440...
|
Perhaps a new examination of the faults in current science is more in order. First, even though years ago I saw a documentary showing the obvious
melting occurring in the usually frozen tundras regions on earth with large methane releases occurring through the unfrozen ground, only recently do I
see articles like, "Thawing permafrost produces more methane than expected" citing 'Methane (CH4) is a potent greenhouse gas that is roughly 30 times
more harmful to the climate than carbon dioxide (CO2). Both gases are produced in thawing permafrost as dead animal and plant remains are decomposed'
link: https://phys.org/news/2018-03-permafrost-methane.html . That article produced forecasts:
"first forecasts. According to the scientists: The permafrost soils of Northern Europe, Northern Asia and North America could produce up to 1 gigaton
of methane and 37 gigatons of carbon dioxide by 2100. "
Also, read "First direct observations of methane's increasing greenhouse effect at the Earth's surface ", at https://phys.org/news/2018-04-methane-greenhouse-effect-eart... .
Per my prior reference, in my opinion, there is some politics mixing in with the science as the official impact of CH4 is still currently based on a
100 year average effect! Basically, it appears that the science is just developing on the real CH4 impact, but it was early on deemed, by the science
community, to be less alarmist by toning down methane impact on global warming by employing a questionable hundred year average effect.
Another source (see https://www.edf.org/methane-other-important-greenhouse-gas?g... ) states, to quote:
"While methane doesn't linger as long in the atmosphere as carbon dioxide, it is initially far more devastating to the climate because of how
effectively it absorbs heat. In the first two decades after its release, methane is 84 times more potent than carbon dioxide. "
Also, per Wikipedia (https://en.wikipedia.org/wiki/Atmospheric_methane ) on methane:
"As methane rises into the air, it reacts with the hydroxyl radical to create water vapor and carbon dioxide. The lifespan of methane in the
atmosphere was estimated at 9.6 years as of 2001; however, increasing emissions of methane over time reduce the concentration of the hydroxyl radical
in the atmosphere."
So the impact of CH4 due to its reduce rate of conversion to CO2 with time is effectively increasing from hydroxyl radical loss. I make sense of the
latter as more CO2 (from all sources inclusive of methane) plus water vapor means increase HCO3- ions which are scavengers of hydroxyl radicals via:
HCO3- + .OH --> H2O + .CO3-
creating a less powerful (than the hydroxyl radical) carbonate radical anion. So, my opinion is that the above reaction could be increasingly in
competition with the action of .OH on CH4 as CO2 levels rise. In other words, a compounding effect which is apparently not baked into the current
widely promulgated science. Also, possible threatening enhancing effects (in apparent agreement with my comment):
"Ramanathan (1988)[81] notes that both water and ice clouds, when formed at cold lower stratospheric temperatures, are extremely efficient in
enhancing the atmospheric greenhouse effect. He also notes that there is a distinct possibility that large increases in future methane may lead to a
surface warming that increases nonlinearly with the methane concentration."
In 1985, to quote "Ramanathan and collaborators announce that global warming may come twice as fast as expected, from rise of methane and other trace
greenhouse gases", see https://history.aip.org/climate/timeline.htm .
If we don't like older models implications (which may be implying a methane induced elevated sensitivity) of what is happening, don't change to newer
models that are more politically digestible.
--------------------------------------------
Here is a recent reference to methane hydrate release from sea floor movement: https://www.sciencedaily.com/releases/2018/02/180206105852.h... .
Also, these two references to ancient events: https://www.sciencedaily.com/releases/2017/11/171120111336.h... and https://www.sciencedaily.com/releases/2010/11/101110101313.h...
Put all three articles together suggests to me a possible CO2/methane footprint to extreme warming occurring 54 million years whose release was
triggered by massive movements of land masses (formation of mountain ranges).
[Edited on 4-4-2018 by AJKOER]
|
|
|
mayko
International Hazard
Posts: 1218
Registered: 17-1-2013
Location: Carrboro, NC
Member Is Offline
Mood: anomalous (Euclid class)
|
|
It's not a secret that methane is a greenhouse gas, and its impact has been and continues to be actively researched; I'm not sure what the 'faults'
are that you want to examine. Maybe I'm reading too much into your wording, but this remark is puzzling:
| Quote: Originally posted by AJKOER |
First, even though years ago I saw a documentary showing the obvious melting occurring in the usually frozen tundras regions on earth with large
methane releases occurring through the unfrozen ground, only recently do I see articles like, "Thawing permafrost produces more methane than expected"
|
If the article was titled "Unexpectedly, Thawing Permafrost Releases Methane", then it well might suggest a glaring oversight in the context of the
earlier documentary you mention. As stated, it's simply reporting on the newest developments in the 'current science'.
Similarly, this sounds to me like you're suggesting that it just now occurred to someone to try to directly measure methane's greenhouse effect. In
reality, it was a long-running project, and the scientists began gathering the data a decade and a half ago. This dedication would seem to contradict
a systemic bias against considering the impact of methane!
| Quote: | | Per my prior reference, in my opinion, there is some politics mixing in with the science as the official impact of CH4 is still currently based on a
100 year average effect! Basically, it appears that the science is just developing on the real CH4 impact, but it was early on deemed, by the science
community, to be less alarmist by toning down methane impact on global warming by employing a questionable hundred year average effect.
|
It sounds like you're talking about Global Warming Potential, which attempts to account for the fact that a given methane emission is, in the long
run, a CO2 emission. It's a little like buying a car in installments. A 20-year GWP would be like the down payment, whereas a 100-year GWP is like the
total amount the car costs. One or the other measure might be more useful, depending on your specific budgeting concerns, but planning around one or
the other doesn't change the basic facts of financing.
This article explores the issue in more detail, but the key points I want to highlight are:
* Using a 20 vs 100 year GWP may make a difference in terms of policy and planning, but it doesn't change the physical science involved ("There is no
scientific reason to prefer a 100-year time horizon over a 20-year time horizon; the choice of GWP100 is simply a matter of convention"; this cuts
both ways!)
* A 100 year GWP isn't inherently politicized, nor is a 10 or 20 year inherently neutral
https://www.scientificamerican.com/article/how-bad-of-a-gree...
An article that might support a larger methane impact that typically thought might be this one:
https://phys.org/news/2017-01-effect-methane-climate-greater...
However, there are two important points to keep in mind:
* The results don't substantially change the big picture: "carbon dioxide remains by far the most significant gas driving human-induced climate
change"
* This is a relatively new result in an area of active research. It may be correct, but new results often aren't, especially when they appear to
radically change the current understanding.
For an example of the second point in action, one need only scroll up! A few posts ago, JJay dropped this link on us with no particular introduction
or explanation:
It's definitely an interesting story! (Again, the scientists involved didn't think that it changed the big picture: "Overwhelming evidence still
suggests that anthropogenic emissions of CFCs and halons are the reason for the ozone loss." That's not my main point here, though.)
The thing is, it's a story from 2007. As one might imagine, there was a flurry of research into the controversy, which confirmed that the reported
result was not correct, and which identified the reason why. For example:
https://www.nature.com/news/2009/070509/full/news.2009.456.h...
Now, some might say it's strange, even negligent, that he'd bring up this particular news item without mentioning any research on the topic that's less than a decade old. I think we can forgive him, though; he was probably busy singlehandedly overturning the
GISS, HadCRUT, BEST, UAH, RSS, and RATPAC datasets (to name a few), from a base level, mere days after needing that base level explained to him (LOL!)
| Quote: | So the impact of CH4 due to its reduce rate of conversion to CO2 with time is effectively increasing from hydroxyl radical loss. I make sense of the
latter as more CO2 (from all sources inclusive of methane) plus water vapor means increase HCO3- ions which are scavengers of hydroxyl radicals via:
HCO3- + .OH --> H2O + .CO3-
creating a less powerful (than the hydroxyl radical) carbonate radical anion. So, my opinion is that the above reaction could be increasingly in
competition with the action of .OH on CH4 as CO2 levels rise. In other words, a compounding effect which is apparently not baked into the current
widely promulgated science.
|
Is there any evidence that this reaction takes place at all under atmospheric conditions?
| Quote: |
Put all three articles together suggests to me a possible CO2/methane footprint to extreme warming occurring 54 million years whose release was
triggered by massive movements of land masses (formation of mountain ranges).
|
It's entirely possible. The PETM is an area of active research, and I don't think there is anything resembling a solid conclusion on the ultimate
causes of the carbon cycle perturbation and thermal spike.
al-khemie is not a terrorist organization
"Chemicals, chemicals... I need chemicals!" - George Hayduke
"Wubbalubba dub-dub!" - Rick Sanchez
|
|
|
| Pages:
1
2
3
4 |
|
![[*]](images/xpblue/default_icon.gif){kind=icon}
