# Talk:Proxy (climate)

WikiProject Environment / Climate change
This environment-related article is part of the WikiProject Environment to improve Wikipedia's coverage of the environment. The aim is to write neutral and well-referenced articles on environment-related topics, as well as to ensure that environment articles are properly categorized.
Read Wikipedia:Contributing FAQ and leave any messages at the project talk page.

Shouldn't the O-18/O-16 ratio be

O-18 to O-16 is 2005 * 10-6 rather than 106? --Vsmith 15:50, 16 Sep 2004 (UTC)

(William M. Connolley 17:36, 16 Sep 2004 (UTC)) I think you're right. I've changed it.

## Merge note

Don't merge - see discussion on Talk:Proxy (statistics). Vsmith 00:34, 21 March 2007 (UTC)

## 1000% or 100%?

The aticle states 1000% in the formula for delta. Shouldn't it be 100%? Dan Gluck (talk) 08:43, 23 October 2009 (UTC)

It is ugly and I don't have a ref handy but it says "1000‰" not "1000%" William M. Connolley (talk) 16:45, 23 October 2009 (UTC)

## Borehole question

I don't follow the explanation of how past temperatures can be discovered from boreholes. The article states that temperatures on the surface send "waves" into the earth. Is the reader to understand that at a certain depth, the temperature ar that depth is correlated with the surface temperature from a certain time period in the past, and that a different depth, the temperature would be related to a different time period in the past? What sort of depth per hundred, or per thousand years, is involved? What sorts of temperatures are involved? What effects from different types of rock?

This sounds really interesting, so I would like to know quite a bit more about it. --Janice Vian, Ph.D. (talk) 23:06, 6 December 2009 (UTC)

If you pretend the temperature of the interior is fixed, and then suppose that the sfc T was too, then you would get a linear variation of T with depth. Now, suppose you instantaneously change the sfc T: you get a slow propagation of a temperature signal into the interior, as the system adjusts to a new linear profile. The response is -exponential with depth, so can be very slow. But since the system doesn't adjust immeadiately it does mean that different depths to some extent preserve a surface T signal from different times, but there is no 1-to-1 correlation of depth A representing time B. Think of it like the mixing that occurs in going from space to time domains in fourier analysis, if you know that. So, with a given starting point and a given sfc T history, the borehole T profile can be predicted moderately accurately. Unfortunately, we have the reverse problem: a given borehole T profile, which needs to be "inverted" to produce a sfc T history. In theory, the two items of information if measured to infinite resolution are equivalent. Since there are uncertainties in the borehole T's (small; ~1/00th's of oC from memory) it is not possible to exactly reconstruct the sfc T history, and the further you go back in time the worse the temporal resolution gets (for example, it may just about be possible to see the last ice age, but only very broadly). I don't have a good ref to hand, but [1] (yes, I know, its by me) brings out some of the issues William M. Connolley (talk) 00:12, 7 December 2009 (UTC)

## Sourcing; self-published vs a secondary book

This edit removed a secondary source and replaced it with a link to a self-published blog. Per WP's guidelines, reliable secondary sources are preferable. Cla68 (talk) 11:50, 19 July 2010 (UTC)

Please stop your disruption here and elsewhere. It does no good, and your replacing a RS with a non-RS is obviously unhelpful William M. Connolley (talk) 13:05, 19 July 2010 (UTC)
Plimer's book is a terrible source (fringe, non-expert, riddled with inaccuracies) but I think we can do better than RC. I've rewritten the intro with a (IMO) more informative description and a source that should be uncontroversial. -- ChrisO (talk) 19:44, 19 July 2010 (UTC)

## comparison?

which proxies have less confounding factors? More closely follow recorded local temperatures. They can't all be equal. Some must be better than others... —Preceding unsigned comment added by 173.67.242.4 (talk) 17:41, 18 August 2010 (UTC)

Fair point. And some reflect different seasons, have different temporal resolution, etc William M. Connolley (talk) 17:39, 19 August 2010 (UTC)

## General problems of paleotemperature proxies

We need to add a section re this, which in essence is the difficulty in separating the temp signal from the noisy background. This affects all the temp. proxies to varying degrees. A good place to start, when it's actually published, will be McShane and Wyner 2010 , the most recent review. --Pete Tillman (talk) 20:37, 20 August 2010 (UTC)

Problems should be mentioned. Your preference for M+W is perhaps indicative of the problems we might have in doing one William M. Connolley (talk) 18:47, 3 October 2010 (UTC)

## Boreholes

Not sure that they should be in here, as they aren't a proxy per se: you are measuring actual temperature, through a filter of thermal diffusivity. Awickert (talk) 13:42, 3 October 2010 (UTC)

Aha, but they are a spatial proxy for a temporal signal :-) William M. Connolley (talk) 18:44, 3 October 2010 (UTC)
Very cute, but seriously isn't that a bit of a stretch? Your logic makes a chart recorder a "spatial proxy".
The 2003 book "Borehole Climatology" by Bodri and Cermak says "Borehole paleoclimatology enables climate reconstruction of the past several millennia, unlike proxy methods provides direct past temperature assessment, and can well broaden the areal range to the remote regions poorly covered with meteorological observations."
Boreholes aren't proxies, they're even more direct recorders of temperature than chart recorders. Vaughan Pratt (talk) 05:31, 5 August 2014 (UTC)

## Boreholes - Limits for Deep Holes

I think you should mention that temperature reconstructions are usually only made upto a depth of about 400 m. In 1997 Shaopeng Huang and Henry N. Pollack published a reconstruction using measurements from a 2 000 m borehole http://www-personal.umich.edu/~shaopeng/97GL01846.pdf. Later, realizing that due to lack of knowledge concerning porosity, rock type, geothermal flow, etc. a proper temperature reconstruction for such a long time range 20 000 years(!) was not possible, they retracted the paper. — Preceding unsigned comment added by 84.176.205.240 (talk) 16:45, 8 January 2012 (UTC)

## Oldest known records

Can we add the oldest known records? Prokaryotes (talk) 20:22, 9 April 2014 (UTC) Suggestion(see editor for sources):
The oldest known climate records are from the Barberton Greenstone Belt South Africa, and Pilbara Craton in Australia which helps to understand Earth's early atmospheric oxygen content 2.3 billion years ago.[1][2] The first glaciation as evidenced in the Pongola Supergroup records from South Africa.[3]

The NYT doesn't even mention the word "climate" William M. Connolley (talk) 20:31, 9 April 2014 (UTC)
See this book tinyurl com/qxdscdx and this recent article http://www.sciencedaily.com/releases/2014/02/140219133331.htm (doi:10.1038/nature13068) Prokaryotes (talk) 20:52, 9 April 2014 (UTC)
Which barely mentions the word. This isn't really a climate proxy in the usual sense. Its so unimportant that our article doesn't discuss it. Why should it be in the lede? William M. Connolley (talk) 07:05, 10 April 2014 (UTC)
Notice that paleoclimatology lists rocks as climate proxy, and ofc it can be mentioned somewhere else other than in the lede. Prokaryotes (talk) 14:18, 10 April 2014 (UTC)
For a broader discussion of earliest climate proxies, see http://arxiv.org/abs/1204.4449 (PDF download link on the right sidebar) Prokaryotes (talk) 00:58, 12 April 2014 (UTC)