|
Totally off-topic thread
|
|
| HarbingerDawn | Date: Wednesday, 30.01.2013, 22:52 | Message # 481 |
 Cosmic Curator
Group: Administrators
United States
Messages: 8717
Status: Offline
| Quote (Tragono) There do exist humans with multiple heads
That is not a human with multiple heads, that is two humans.
Someone with multiple minds can exist with only one head, it's a psychological disorder.
All forum users, please read this!
My SE mods and addons
Phenom II X6 1090T 3.2 GHz, 16 GB DDR3 RAM, GTX 970 3584 MB VRAM
|
| |
| |
| DoctorOfSpace | Date: Wednesday, 30.01.2013, 22:55 | Message # 482 |
 Galaxy Architect
Group: Global Moderators
Pirate
Messages: 3600
Status: Offline
| Quote (HarbingerDawn) Someone with multiple minds can exist with only one head, it's a psychological disorder.
Intel Core i7-5820K 4.2GHz 6-Core Processor
G.Skill Ripjaws V Series 32GB (4 x 8GB) DDR4-2400 Memory
EVGA GTX 980 Ti SC 6GB
Edited by DoctorOfSpace - Wednesday, 30.01.2013, 23:03 |
| |
| |
| Watsisname | Date: Wednesday, 30.01.2013, 23:34 | Message # 483 |
 Galaxy Architect
Group: Global Moderators
United States
Messages: 2613
Status: Offline
| Quote (midtskogen) Yes[link], in particular for the period when the instrumental record is the most accurate.
Right, this is a big point as I will elaborate on in a moment. Also I'm not sure if I understand the significance of your link -- the divergence problem is a very specific phenomenon with recent northern hemisphere tree growth and is beyond my level of comfortable knowledge. Is there something about it that you are interested in going over? If you like I could gather some research articles for you. 
Quote But isn't rather the instrumental record how you calibrate the proxies?
In general, yes. Perhaps my phrasing caused some confusion -- let me try to clarify.
Calibrating the proxies by the instrumental record means that we can figure out through empirical data what the proxy/temperature relationships are, and it shows that the proxies really are responding to the temperature changes. This demonstrates their robustness as a method of determining prehistoric temperature. So, the (relative) flatness of the pre-industrial temperature isn't a matter of statistical shenanigans or a sign that the proxies aren't working. There really is a tremendous difference in the behavior of the current climate vs. previous climate.
Further support of this is not hard to come by. You can look at how the reconstructions show the more significant past climate changes as known from geology (e.g. ice ages). You can compare it with evidence from historical events from before the instrumental record (volcanic eruptions are an excellent example). And you can compare it with what you'd expect from applying radiative physics to the evolution of atmospheric composition. (For current climate changes in this case -- past climate changes had different underlying mechanisms, such as orbital forcings, though CO2 still plays a big role in most if not all of them.)
|
| |
| |
| DoctorOfSpace | Date: Wednesday, 30.01.2013, 23:47 | Message # 484 |
|
Galaxy Architect
Group: Global Moderators
Pirate
Messages: 3600
Status: Offline
| When the robot apocalypse happens, at least they'll do it in style
Intel Core i7-5820K 4.2GHz 6-Core Processor
G.Skill Ripjaws V Series 32GB (4 x 8GB) DDR4-2400 Memory
EVGA GTX 980 Ti SC 6GB
|
| |
| |
| Watsisname | Date: Thursday, 31.01.2013, 00:44 | Message # 485 |
|
Galaxy Architect
Group: Global Moderators
United States
Messages: 2613
Status: Offline
| That just totally made my day. 
|
| |
| |
| midtskogen | Date: Thursday, 31.01.2013, 09:53 | Message # 486 |
 Star Engineer
Group: Users
Norway
Messages: 1674
Status: Offline
| Quote (Watsisname) I'm not sure if I understand the significance of your link
I referred to this as an example of things that you can run into when using proxies. I don't think we should attempt to go deeply into this one. Because of the "Nature trick" and "hide the decline" controversy and the role of tree rings in the hockey stick story, there is much discussion out there way too polarised. Which is certainly easily fuelled by that this proxy happens to involve factors that every layman already understands, that tree growth depends on the length and quality of the growing season (which in turn partly relates to temperature, spring temperature in particular I suppose), and other factors like precipitation, sunlight, soil conditions, CO2 concentration and damage from weather events like floods and storms, to name what immediately comes to mind, all of which will influence trees in a large area similarly. But it's by no means obvious to the layman how to extract only the growing season component of that, and from that the annual temperature. Cue the instrumental record. A pity, then, if it doesn't quite line up. Then you need to solve that problem, look for things to throw into the equation that will give you what you look for without contradicting something else. But in order to do so, you also have to assume what you're trying to demonstrate, that the proxy follows temperature. See the problem?
Quote (Watsisname) Calibrating the proxies by the instrumental record means that we can figure out through empirical data what the proxy/temperature relationships are, and it shows that the proxies really are responding to the temperature changes.
And my point was that when you do this, line up your proxies with the instrumental record, you can't then just say, hey, look how the proxies reproduce the instrumental record! Because you did, in fact, make them do so. They didn't entirely by themselves. The more fitting you need to line up, the more you assume which has to be checked independently. If it's just a scaling issue, you're lucky (you still have to provide a physical explanation for how your proxy works). Your reference should also be subject to similar conditions. For instance, we know that CO2 increases tree growth and the best part of the instrumental record is from a period with entirely different CO2 levels. Hence the if-given-assuming parts of the story are piling up. I do not deny the possibility that a 0.5-1 degree change in a century is the first that the world has seen in 12,000 years, but I question whether the given evidence for the claim is sound, and in the light of the obvious (perhaps more obvious here in high latitudes) that the earth is a dynamic place, the claim should also require extraordinary evidence.
NIL DIFFICILE VOLENTI
|
| |
| |
| Watsisname | Date: Friday, 01.02.2013, 02:19 | Message # 487 |
|
Galaxy Architect
Group: Global Moderators
United States
Messages: 2613
Status: Offline
| I apologize for the length of this post, but there's a lot I'd like to cover.
Quote (midtskogen) I referred to this as an example of things that you can run into when using proxies. I don't think we should attempt to go deeply into this one. Because of the "Nature trick" and "hide the decline" controversy and the role of tree rings in the hockey stick story, there is much discussion out there way too polarised.
I see. As an up-coming scientist, I prefer to listen to scientists, not people who are only interested in hacking their emails and going conspiracy-theory-looney over things that were quoted without proper context. I'd thought the CRU scandal would have died down after the investigations, but evidently not. :/ Anyone who thinks there's some collusion among scientists with the divergence problem must not have read any of the scientific literature about it. It's an open discussion and there's a lot of interesting research being done.
But I digress. The important note for you here is that nobody claims that every proxy is a 100% correct reproduction of temperature. You have focused on tree rings in this discussion, and this isn't too surprising -- tree rings do not depend only on temperature, as you have correctly pointed out. You can find further failures of tree rings such as how they (don't) accurately reflect volcanic eruptions. Again this is why we use multiple proxies and check the agreement between them. Similar idea with hurricane forecasting -- you use multiple models and determine a consensus.
Quote Cue the instrumental record. A pity, then, if it doesn't quite line up. Then you need to solve that problem, look for things to throw into the equation that will give you what you look for without contradicting something else. But in order to do so, you also have to assume what you're trying to demonstrate, that the proxy follows temperature. See the problem?
Pity? No, it's actually very interesting and we can learn some new things from it. The divergence problem demonstrates that some tree rings stop correlating with temperature for the latter half of the 20th century (and curiously this is about the only time this problem occurs -- leading to the suggestions that either there's an anthropic effect going on, or perhaps it's a result of using young trees.) Until we figure out exactly what's going on we simply disregard those data that fail to maintain correlation. Why? Because they're not acting as proxies! (This is the cause of the ridiculous 'Hide the decline!' stuff that you see from people who aren't involved in climate science). So no, I see any problem. As I explained in my last post, there are additional ways of being sure if the proxy/temperature relation we're using is accurate.
Quote And my point was that when you do this, line up your proxies with the instrumental record, you can't then just say, hey, look how the proxies reproduce the instrumental record! Because you did, in fact, make them do so.
Your point is incorrect. Your reasoning is good, but the premise is wrong.
We do not simply make them line them up. We use the instrumental record to determine if the proxy is correlating with temperature (a good scientist will also demonstrate the process by which they are related). And because the actual temperatures are known we can quantify the relationship and thus turn the proxy curve into a temperature curve. Ie, it is a two-step process which first requires that we understand how the proxy is behaving. Now, when I say the proxies reproduce the record, this is what I am talking about:
This is a proxy not reproducing the temperature record. (Both fictional, obviously). The proxy shows no apparent dependence on temperature.
This is a proxy reproducing the temperature record. In this example it correlates with the temperature in an inverse way.
Clear now?
Quote I do not deny the possibility that a 0.5-1 degree change in a century is the first that the world has seen in 12,000 years, but I question whether the given evidence for the claim is sound, and in the light of the obvious (perhaps more obvious here in high latitudes) that the earth is a dynamic place, the claim should also require extraordinary evidence.
I've already talked about how dynamic the Earth is. But the best science we have shows that the global temperatures during the Holocene up to the industrial age were relatively stable compared to the present period of human-caused warming. It is good to ask for evidence, but I think you're not recognizing that it exists. Indeed, there is extraordinary evidence:
-Temperature data
-Atmospheric composition data
-Physics
Edited by Watsisname - Friday, 01.02.2013, 02:29 |
| |
| |
| midtskogen | Date: Friday, 01.02.2013, 15:39 | Message # 488 |
|
Star Engineer
Group: Users
Norway
Messages: 1674
Status: Offline
| Quote (Watsisname) As an up-coming scientist, I prefer to listen to scientists
And some wanted to hear what scientists say so badly that they hacked their emails.
Quote (Watsisname) The divergence problem demonstrates that some tree rings stop correlating with temperature for the latter half of the 20th century (and curiously this is about the only time this problem occurs -- leading to the suggestions that either there's an anthropic effect going on, or perhaps it's a result of using young trees.)
How exactly do you know it's the only time when direct temperature measurements before the 20th century are scarce or non-existing without assuming indirectly or directly that the proxy is valid before that?
Quote (Watsisname) Until we figure out exactly what's going on we simply disregard those data that fail to maintain correlation. Why? Because they're not acting as proxies!
It's a dangerous path to cherrypick. And to use circular logic to defend it. They might not act like the proxy you wish, simply because there might be no such proxy, unless you decided there is, but then you've equally decided how the data should be and act.
However, if you mean that we simply should disregard the whole proxy, not just the data that don't fit, then, if you want to take the safe path, I agree. There are other proxies which might pass tests better. Don't throw the data, but leave all of them out for now. But that could be a problem "given the wide use of tree-ring data in temperature reconstructions of the past millennium, of high-resolution paleoclimatology as a whole" to quote one paper that came up through your link.
You'll probably say that other proxies will support the removal of recent tree rings. Well, until we know the physical explanation (and in this case more explanations isn't better), it's hard to guess if and how often they've diverged before and when, so the safest option still would be to remove the whole proxy, hopefully just temporarily.
This is similar to the solar cycle and temperature correlations that have been presented. Until we have independently verified physical explanations for the correlations and the deviations, we must be cautious.
Quote (Watsisname) We do not simply make them line them up. We use the instrumental record to determine if the proxy is correlating with temperature (a good scientist will also demonstrate the process by which they are related).
I think we're talking a bit past each other here. It's a matter of how straight forward it is to interpret the proxy, how to tell fudge factors and physical properties apart. If you find the text 13.4 °C, tea time, 14th June 1984 embedded in your tree ring, you're pretty fine. If you can't get your proxy to match until you factor in the phase of Venus and how much the monarch of England had for breakfast, you're in deep water even if it then perfectly matches the instrument record and everything else. And now the crucial part: When you try to figure out how to interpret your proxy, you have the answer, you know what you must arrive at, and if you think that wont influence how you interpret, you should rethink. Proxies don't come with an intrinsic instruction manual.
If you're determined that it must be possible to deduce the temperature from tree rings, you can become quite willing to produce solutions to any problems. It's not necessarily a good sign. Remember the geocentric view and the epicycles.
Quote (Watsisname) It is good to ask for evidence, but I think you're not recognizing that it exists.
We were talking about variations like ±0.2 degrees or less. I think we might have different opinions on how little that is. For instance, it's about the absolute accuracy of modern professional thermometers excluding shielding and siting issues. But also consider that we're talking about surface temperature, something limited to 2D in a 3D world. What if there is a slight change in air or ocean circulation, and suddenly something cold or warm passes through our 2D observation plane and we'll detect a shift even if there was none overall in 3D. Intuitively it seems strange not to expect such phenomena of any significance.
I wish climate science were more self critical about its uncertainties and rush a bit less to publish papers. Look how often you encounter phrases like "could lead to" in climate research (but medical research probably beats them). When we discussed assumptations, you referred to Einstein and what he postulated. But did he ever say that this could (or possibly not) make clocks appear to have slowed down if sent through space and returned? Did Newton say that his work could explain Kepler's laws? Hypotheses weigh no more than number of ways to contradict them, and coulds are hard to disprove, since it implies that you don't only have to provide contradicting data, but also a better, alternative hypothesis. It somewhat shifts the burden of proof. I wish papers on climate could be more specific on how to they can be disproven. If they can't because climate is inherently uncertain, the scientists must come to terms with practising a lightweight science. Added (01.02.2013, 18:39)
---------------------------------------------
Quote (Watsisname) I'd thought the CRU scandal would have died down after the investigations, but evidently not.
Some additional thoughts. I'm not sure what the mandate of these investigations was, but I've no reason no believe that any of the scientists involved ever have intentionally presented scientific claims for which there was no firm scientific basis. On the contrary, I have an impression from some of the e-mails that I've seen that they've taken much care to make sure that they don't go further than that. But the science also offer some slack because of the unknowns. So it's no doubt possible to use that and tell the public a somewhat different story if you simply highlight different things, still within what you have a scientific basis for. When something has gone political, this is to be expected.
NIL DIFFICILE VOLENTI
|
| |
| |
| Watsisname | Date: Friday, 01.02.2013, 22:23 | Message # 489 |
|
Galaxy Architect
Group: Global Moderators
United States
Messages: 2613
Status: Offline
| Hacking a scientist's emails and then quoting things without the appropriate context does not seem like the actions of a person who is genuinely interested or skeptical. 
Quote (midtskogen) How exactly do you know it's the only time when direct temperature measurements before the 20th century are scarce or non-existing without assuming indirectly or directly that the proxy is valid before that?
Careful, I said it was about the only time, and I gave you another example of where tree rings can be faulty. But how do we know when they work and when they don't? Because:
-we can compare with other proxies
-we can examine how they handle known climate changes
Quote (midtskogen) It's a dangerous path to cherrypick. And to use circular logic to defend it. They might not act like the proxy you wish, simply because there might be no such proxy, unless you decided there is, but then you've equally decided how the data should be and act.
It is not cherrypicking, circular logic, or deciding how the proxy should act, if you can check it through multiple independent methods. If we had only one proxy, no understanding of the underlying processes, or no instrumental temperature record, then your statement would be valid. (But thanks for suggesting that the people who do this kind of work for a living are so misguided on basic scientific practices; that's very nice of you and I'm glad you understand the subject better than they do.) 
As for your suggestion that we disregard all tree ring data, which by and large agrees very well with the other data, simply because it fails under certain conditions which can be determined independently? Absolutely not. That would be completely ridiculous.
Quote I think we're talking a bit past each other here. It's a matter of how straight forward it is to interpret the proxy, how to tell fudge factors and physical properties apart. If you find the text 13.4 °C, tea time, 14th June 1984 embedded in your tree ring, you're pretty fine. If you can't get your proxy to match until you factor in the phase of Venus and how much the monarch of England had for breakfast, you're in deep water even if it then perfectly matches the instrument record and everything else. And now the crucial part: When you try to figure out how to interpret your proxy, you have the answer, you know what you must arrive at, and if you think that wont influence how you interpret, you should rethink. Proxies don't come with an intrinsic instruction manual.
Lol, ok, so you're concerned about the fine details. In that case I simply cannot help you. I think you might be in need of actually studying how this type of work is done before you assume it's all a bunch of guesswork, fudging, and foolishness.
Quote If you're determined that it must be possible to deduce the temperature from tree rings, you can become quite willing to produce solutions to any problems. It's not necessarily a good sign. Remember the geocentric view and the epicycles.
We do not believe it is possible to deduce temperature from tree rings because we want to. We believe it because that's what the evidence tells us. They are not perfect, and nobody claims otherwise, but they are by and large extremely good.
Quote We were talking about variations like ±0.2 degrees or less. I think we might have different opinions on how little that is. [etc]
There have been much greater than ±0.2°C variations during the pre-industrial / post-glacial period. A single volcanic eruption can change temperature by a full degree C. Also, the 3D nature of the atmosphere and relationships between it and the other earth systems are understood better than you might think. I'd recommend a course or textbook on atmospheric science; I am currently taking one and it's a great subject.
Quote I wish climate science were more self critical about its uncertainties and rush a bit less to publish papers. Look how often you encounter phrases like "could lead to" in climate research (but medical research probably beats them). [etc]
Climate scientists are generally very self critical of their own work and of others' work in the field. It is no different than any other science. When they are uncertain about something, they will say so. When they are confident about something, they will say so. Have you read the language of the IPCC reports? They are extremely diligent about what they mean with regards to uncertainty.
It would be wonderful if we could have a serious discussion about the real uncertainties in climate change with the public, because they are much more interesting and relevant than the misconceptions that keep flying around the internet and the news reports. 
|
| |
| |
| Disasterpiece | Date: Friday, 01.02.2013, 23:25 | Message # 490 |
 World Builder
Group: Users
United States
Messages: 640
Status: Offline
| Wasn't 0.97 supposed to release today?
I play teh spase engien
|
| |
| |
| Watsisname | Date: Saturday, 02.02.2013, 00:07 | Message # 491 |
|
Galaxy Architect
Group: Global Moderators
United States
Messages: 2613
Status: Offline
| No, Space Engineer has not set a release date. It may be soon though, so stay tuned.
|
| |
| |
| Disasterpiece | Date: Saturday, 02.02.2013, 04:27 | Message # 492 |
|
World Builder
Group: Users
United States
Messages: 640
Status: Offline
| Quote (Watsisname) No, Space Engineer has not set a release date.
Werdnaforever has dared fool the great Disasterpiece. He has made himself an enemy, an enemy that stands around every corner he turns, watches him where he thinks he is safe and strikes when he is least expecting it. Werdnaforever will regret this day, and he will pay dearly, in fact I may not even subject him to physical pain, I might blindfold him in my torture room and cut my other prisoner to pieces, and I'll let Werdna's imagination tell him what's happening, tell him all the horrible fates that could await him, and then I'll disappear. For the rest of his life he'll check behind him at night, flinch at every sound and grimace every time his door opens. You will regret your actions Werdna, I know you will.Added (02.02.2013, 07:27)
---------------------------------------------
Do NOT be alarmed, that was a joke. Or most of it at least.
I play teh spase engien
|
| |
| |
| midtskogen | Date: Saturday, 02.02.2013, 09:04 | Message # 493 |
|
Star Engineer
Group: Users
Norway
Messages: 1674
Status: Offline
| Quote (DoctorOfSpace) But how do we know when they work and when they don't? Because:
-we can compare with other proxies
-we can examine how they handle known climate changes
I commented on the other proxies in the last post, arguing that we also need to know what the physical explanation for the divergence is.
Your link on known climate changes marks large volcanic eruptions. So we know what those look like in tree rings, which doesn't necessarily mean that we know what other types of cooling and warming events look like. Rather, the signal of volcanic events might differ somewhat from other temperature changes.
Quote (Watsisname) (But thanks for suggesting that the people who do this kind of work for a living are so misguided on basic scientific practices; that's very nice of you and I'm glad you understand the subject better than they do.)
I meant that your statement was suggesting cherrypicking and the logic was circular, because you said that tree ring data that didn't fit the story could be thrown away because they don't act as a proxy. It may be what you do for other reasons, but it's by no means a sufficient reason. Ultimately all reasoning is circular with non-contradicting statements within, but we aim for the big circles. I know you meant that there are other reasons, other data and proxies.
Quote (Watsisname) Lol, ok, so you're concerned about the fine details. In that case I simply cannot help you. I think you might be in need of actually studying how this type of work is done before you assume it's all a bunch of guesswork, fudging, and foolishness.
No, not the fine detail. I'm trying to bring in some philosophy of science here. If the provocation of questioning something that could support the mainstream view distracts you, I'm happy to use an example from the opposing view instead.
We know that cold periods in at least some parts of the world have coincided somewhat with periods of few or no sunspots, and we have a fairly long record of sunspot counting. So it's a natural thing to investigate if we can use the sunspot data as a proxy for temperature. At least in Europe we have a reasonable temperature coverage back to around 1850, so we have several solar cycles to work with.
We have sunspot numbers, and we know what to arrive at (temperature). So we start the search. Does the sunspot number correlate with the concurrent temperature? No. Does the solar max level correspond to temperature? Nope. Does the solar min level correspond to temperature? Still no. Does the length of the solar cycle correspond to temperature? No yet again. Does the length of the previous solar cycle correspond to temperature? Eureka, we have a proxy. Then we need to find the physical explanation for this. Cosmic ray modulation and cloud formation has been proposed, and many are working on the details. But there is a slight problem. If we use temperature records prior to 1850, the correlation gets weak for some cycles. We have a divergence problem. Perhaps it can be explained by the scarcity of records before that, or perhaps other drivers happened to dominate in these periods. But since that can be looked at independently, and what we have after 1850 agrees well with other proxies, we'll simply ignore the data that don't act as a proxy for now.
Do you see the problem with this methodology that I see? What I'm most concerned about is that you know what you want to arrive at the whole time, and that is going to give a bias.
If your response is that scientists like Mann are fine scientists not subject to bias like Humlum et al. who are known for flawed research practices and dishonesty, fine, but that's totally beside the point and not only an argumentum ad verecundiam, but also ad hominem.
NIL DIFFICILE VOLENTI
Edited by midtskogen - Saturday, 02.02.2013, 09:45 |
| |
| |
| Spyro | Date: Sunday, 03.02.2013, 02:18 | Message # 494 |
 Explorer
Group: Users
United States
Messages: 162
Status: Offline
| Quote When the robot apocalypse happens, at least they'll do it in style
BTW how do you post videos? Sorry if this has been asked before, i'm a total noob.
I'm back, but am I here to stay? Hopefully!
|
| |
| |
| werdnaforever | Date: Sunday, 03.02.2013, 04:23 | Message # 495 |
 World Builder
Group: Users
Pirate
Messages: 897
Status: Offline
| Well, I finally did it. I got a .org URL for my website. 10 dollars a year- it's literally dirt cheap (can't think of a better expression). I can afford that. Why I haven't done this before is a question I cannot answer. I guess I never considered how cheap it was.
So, maybe I could have gotten it for a few dollars less somewhere else, but whatever.
Having a .org URL vs. a stupid .blogspot.com (it's so long!) URL is just... amazing. It's so cool! It's like it's a website now, and not a silly blog. A website! W - E - B - S - I- T - E ! A whebbe-sight? Not a blog... blog rhymes with smog for a reason.
EDIT:
Despite it's flaws, I love this movie.
Edited by werdnaforever - Sunday, 03.02.2013, 06:34 |
| |
| |
ENG
New site
United States 
Pirate 
Norway