Comments on: Droughts and Antarctica http://landshape.org/enm/droughts-and-antarctica/ The Power of Numeracy Tue, 07 Feb 2012 17:57:00 +0000 hourly 1 http://wordpress.org/?v=3.3.1 By: davids99us http://landshape.org/enm/droughts-and-antarctica/#comment-1233 davids99us Wed, 10 Feb 2010 17:02:29 +0000 http://landshape.org/enm/?p=3617#comment-1233 Just a note to point out that just because I don't respond to something doesn't mean I agree. I just want to focus on the main thing, which is getting some data and crunching it. Eg I dont think that reversal of claims by subsequent data is normal scientific method, in the sense that as more data is accumulated, certainty, or the strength of claims should increase, not decrease. If they decrease, then inappropriate significance has been given to it, usually because the distribution was assumed to be Gaussian and it wasn't, there was autocorrelation in the data and it wasn't handled, and so on.False certainty is important. In 1997, Niederhoffer Investments lost 100 million in one day because he assumed the probability of the market falling more that 5% in one day was astronomically low. It fell 7.2% and he was wiped out. LTCM same story. The 'long tail' of probability got them. Its the same with climate. You have to be very careful arguing that something is outside the normal and therefore manmade when you have 1. limited data and 2. non-normal distributions. It's virtually impossible to estimate the true variance of climate data, and so usually very suspect to argue that something is 3-sigma, 1:1000 chance and so on, with any reliability. Just a note to point out that just because I don't respond to something doesn't mean I agree. I just want to focus on the main thing, which is getting some data and crunching it. Eg I dont think that reversal of claims by subsequent data is normal scientific method, in the sense that as more data is accumulated, certainty, or the strength of claims should increase, not decrease. If they decrease, then inappropriate significance has been given to it, usually because the distribution was assumed to be Gaussian and it wasn't, there was autocorrelation in the data and it wasn't handled, and so on.False certainty is important. In 1997, Niederhoffer Investments lost 100 million in one day because he assumed the probability of the market falling more that 5% in one day was astronomically low. It fell 7.2% and he was wiped out. LTCM same story. The 'long tail' of probability got them. Its the same with climate. You have to be very careful arguing that something is outside the normal and therefore manmade when you have 1. limited data and 2. non-normal distributions. It's virtually impossible to estimate the true variance of climate data, and so usually very suspect to argue that something is 3-sigma, 1:1000 chance and so on, with any reliability.

]]> By: davids99us http://landshape.org/enm/droughts-and-antarctica/#comment-1236 davids99us Wed, 10 Feb 2010 14:37:10 +0000 http://landshape.org/enm/?p=3617#comment-1236 Just an update. Tas has been more than helpful in supplying material for replicating his study, and I will be working on this over the next few days. Just an update. Tas has been more than helpful in supplying material for replicating his study, and I will be working on this over the next few days.

]]> By: Anonymous http://landshape.org/enm/droughts-and-antarctica/#comment-12204 Anonymous Wed, 10 Feb 2010 12:02:00 +0000 http://landshape.org/enm/?p=3617#comment-12204 Just a note to point out that just because I don't respond to something doesn't mean I agree. I just want to focus on the main thing, which is getting some data and crunching it. Eg I dont think that reversal of claims by subsequent data is normal scientific method, in the sense that as more data is accumulated, certainty, or the strength of claims should increase, not decrease. If they decrease, then inappropriate significance has been given to it, usually because the distribution was assumed to be Gaussian and it wasn't, there was autocorrelation in the data and it wasn't handled, and so on. False certainty is important. In 1997, Niederhoffer Investments lost 100 million in one day because he assumed the probability of the market falling more that 5% in one day was astronomically low. It fell 7.2% and he was wiped out. LTCM same story. The 'long tail' of probability got them. Its the same with climate. You have to be very careful arguing that something is outside the normal and therefore manmade when you have 1. limited data and 2. non-normal distributions. It's virtually impossible to estimate the true variance of climate data, and so usually very suspect to argue that something is 3-sigma, 1:1000 chance and so on, with any reliability. Just a note to point out that just because I don’t respond to something doesn’t mean I agree. I just want to focus on the main thing, which is getting some data and crunching it.

Eg I dont think that reversal of claims by subsequent data is normal scientific method, in the sense that as more data is accumulated, certainty, or the strength of claims should increase, not decrease. If they decrease, then inappropriate significance has been given to it, usually because the distribution was assumed to be Gaussian and it wasn’t, there was autocorrelation in the data and it wasn’t handled, and so on.

False certainty is important. In 1997, Niederhoffer Investments lost 100 million in one day because he assumed the probability of the market falling more that 5% in one day was astronomically low. It fell 7.2% and he was wiped out. LTCM same story. The ‘long tail’ of probability got them.

Its the same with climate. You have to be very careful arguing that something is outside the normal and therefore manmade when you have 1. limited data and 2. non-normal distributions. It’s virtually impossible to estimate the true variance of climate data, and so usually very suspect to argue that something is 3-sigma, 1:1000 chance and so on, with any reliability.

]]> By: tasvanommen http://landshape.org/enm/droughts-and-antarctica/#comment-1235 tasvanommen Wed, 10 Feb 2010 12:01:17 +0000 http://landshape.org/enm/?p=3617#comment-1235 Hi David,The paper is at Nature Geoscience (Advance Online), and I'd encourage you to take a look (and at the online supplementary information). Contact me by email if you have trouble getting the paper.The data are available online at the Australian Antarctic Data Centre (<a href="http://data.aad.gov.au/aadc/metadata/citation.cfm?entry_id=AAS_757_LDAccu2005" rel="nofollow">http://data.aad.gov.au/aadc/metadata/citation.c...</a>)and will be indexed too by the NASA Global Change Master Directory, which appears to be offline at present.Again contact me (email is better than on-blog) for specifics of data and paper access.On the fact that claims can be reversed by subsequent data, this sounds like a description of the advance of science via the normal method. All any of us can do is honestly report our findings and uncertainties, try to use our heads and add some interpretation (or else it is largely just a technological exercise) to guide followup studies. If they find new things out that make our work obsolete then so be it - we have helped push things forward.Cheers,Tas Hi David,The paper is at Nature Geoscience (Advance Online), and I'd encourage you to take a look (and at the online supplementary information). Contact me by email if you have trouble getting the paper.The data are available online at the Australian Antarctic Data Centre (http://data.aad.gov.au/aadc/metadata/citation.c…)and will be indexed too by the NASA Global Change Master Directory, which appears to be offline at present.Again contact me (email is better than on-blog) for specifics of data and paper access.On the fact that claims can be reversed by subsequent data, this sounds like a description of the advance of science via the normal method. All any of us can do is honestly report our findings and uncertainties, try to use our heads and add some interpretation (or else it is largely just a technological exercise) to guide followup studies. If they find new things out that make our work obsolete then so be it – we have helped push things forward.Cheers,Tas

]]> By: davids99us http://landshape.org/enm/droughts-and-antarctica/#comment-1234 davids99us Wed, 10 Feb 2010 11:36:37 +0000 http://landshape.org/enm/?p=3617#comment-1234 Tas, I appreciate you commenting as I have no interest to preaching to the choir, and a one-sided blog. It seems we are agreed that the reporting was, and still is, overblown. I think there is a need to speak out about this as it is damaging the reputation of science.The substantive issue on the significant change in circulation is something I have looked into with the Walker Circulation (<a href="http://landshape.org/enm/no-weakening-of-the-walker/" rel="nofollow">http://landshape.org/enm/no-weakening-of-the-wa...</a>). As I am sure you know, Power and Smith and Vecchi reported a weakening of the WC as evidence of AGW. However, since a run of strong El Ninos has eased the data indicate a return of the indices to normal and no longer support their claim to any degree, suggesting, what? We have seen before where these sorts of claims are reversed by subsequent data (eg <a href="http://landshape.org/enm/wp-content/uploads/2009/06/ee-20-4_7-stockwell2.pdf" rel="nofollow">http://landshape.org/enm/wp-content/uploads/200...</a>).I wonder if this case might be similar? I would be prepared to look into it . Would you post the data for your graph showing the inverse link between rainfall in south-west Western Australia and snowfall in Antarctica (<a href="http://www.aad.gov.au/MediaLibrary/asset/MediaItems/ml_402173825925926_03_RainSnowComparisonGraph.pdf" rel="nofollow">http://www.aad.gov.au/MediaLibrary/asset/MediaI...</a>) so we can check the statistics of the relationship? I admit I haven't read your paper as I didn't see a reference to it anywhere, but if you have a reference and a link I would appreciate it. Tas, I appreciate you commenting as I have no interest to preaching to the choir, and a one-sided blog. It seems we are agreed that the reporting was, and still is, overblown. I think there is a need to speak out about this as it is damaging the reputation of science.The substantive issue on the significant change in circulation is something I have looked into with the Walker Circulation (http://landshape.org/enm/no-weakening-of-the-wa…). As I am sure you know, Power and Smith and Vecchi reported a weakening of the WC as evidence of AGW. However, since a run of strong El Ninos has eased the data indicate a return of the indices to normal and no longer support their claim to any degree, suggesting, what? We have seen before where these sorts of claims are reversed by subsequent data (eg http://landshape.org/enm/wp-content/uploads/200…).I wonder if this case might be similar? I would be prepared to look into it . Would you post the data for your graph showing the inverse link between rainfall in south-west Western Australia and snowfall in Antarctica (http://www.aad.gov.au/MediaLibrary/asset/MediaI…) so we can check the statistics of the relationship? I admit I haven't read your paper as I didn't see a reference to it anywhere, but if you have a reference and a link I would appreciate it.

]]> By: davids99us http://landshape.org/enm/droughts-and-antarctica/#comment-1232 davids99us Wed, 10 Feb 2010 11:02:29 +0000 http://landshape.org/enm/?p=3617#comment-1232 Just a note to point out that just because I don't respond to something doesn't mean I agree. I just want to focus on the main thing, which is getting some data and crunching it. Eg I dont think that reversal of claims by subsequent data is normal scientific method, in the sense that as more data is accumulated, certainty, or the strength of claims should increase, not decrease. If they decrease, then inappropriate significance has been given to it, usually because the distribution was assumed to be Gaussian and it wasn't, there was autocorrelation in the data and it wasn't handled, and so on.False certainty is important. In 1997, Niederhoffer Investments lost 100 million in one day because he assumed the probability of the market falling more that 5% in one day was astronomically low. It fell 7.2% and he was wiped out. LTCM same story. The 'long tail' of probability got them. Its the same with climate. You have to be very careful arguing that something is outside the normal and therefore manmade when you have 1. limited data and 2. non-normal distributions. It's virtually impossible to estimate the true variance of climate data, and so usually very suspect to argue that something is 3-sigma, 1:1000 chance and so on, with any reliability. Just a note to point out that just because I don't respond to something doesn't mean I agree. I just want to focus on the main thing, which is getting some data and crunching it. Eg I dont think that reversal of claims by subsequent data is normal scientific method, in the sense that as more data is accumulated, certainty, or the strength of claims should increase, not decrease. If they decrease, then inappropriate significance has been given to it, usually because the distribution was assumed to be Gaussian and it wasn't, there was autocorrelation in the data and it wasn't handled, and so on.False certainty is important. In 1997, Niederhoffer Investments lost 100 million in one day because he assumed the probability of the market falling more that 5% in one day was astronomically low. It fell 7.2% and he was wiped out. LTCM same story. The 'long tail' of probability got them. Its the same with climate. You have to be very careful arguing that something is outside the normal and therefore manmade when you have 1. limited data and 2. non-normal distributions. It's virtually impossible to estimate the true variance of climate data, and so usually very suspect to argue that something is 3-sigma, 1:1000 chance and so on, with any reliability.

]]> By: Anonymous http://landshape.org/enm/droughts-and-antarctica/#comment-12203 Anonymous Wed, 10 Feb 2010 09:37:00 +0000 http://landshape.org/enm/?p=3617#comment-12203 Just an update. Tas has been more than helpful in supplying material for replicating his study, and I will be working on this over the next few days. Just an update. Tas has been more than helpful in supplying material for replicating his study, and I will be working on this over the next few days.

]]> By: davids99us http://landshape.org/enm/droughts-and-antarctica/#comment-1231 davids99us Wed, 10 Feb 2010 08:37:10 +0000 http://landshape.org/enm/?p=3617#comment-1231 Just an update. Tas has been more than helpful in supplying material for replicating his study, and I will be working on this over the next few days. Just an update. Tas has been more than helpful in supplying material for replicating his study, and I will be working on this over the next few days.

]]> By: Anonymous http://landshape.org/enm/droughts-and-antarctica/#comment-12202 Anonymous Wed, 10 Feb 2010 07:01:00 +0000 http://landshape.org/enm/?p=3617#comment-12202 Hi David, The paper is at Nature Geoscience (Advance Online), and I'd encourage you to take a look (and at the online supplementary information). Contact me by email if you have trouble getting the paper. The data are available online at the Australian Antarctic Data Centre (http://data.aad.gov.au/aadc/metadata/citation.cfm?entry_id=AAS_757_LDAccu2005) and will be indexed too by the NASA Global Change Master Directory, which appears to be offline at present. Again contact me (email is better than on-blog) for specifics of data and paper access. On the fact that claims can be reversed by subsequent data, this sounds like a description of the advance of science via the normal method. All any of us can do is honestly report our findings and uncertainties, try to use our heads and add some interpretation (or else it is largely just a technological exercise) to guide followup studies. If they find new things out that make our work obsolete then so be it - we have helped push things forward. Cheers, Tas Hi David,
The paper is at Nature Geoscience (Advance Online), and I’d encourage you to take a look (and at the online supplementary information). Contact me by email if you have trouble getting the paper.

The data are available online at the Australian Antarctic Data Centre (http://data.aad.gov.au/aadc/metadata/citation.cfm?entry_id=AAS_757_LDAccu2005)
and will be indexed too by the NASA Global Change Master Directory, which appears to be offline at present.

Again contact me (email is better than on-blog) for specifics of data and paper access.

On the fact that claims can be reversed by subsequent data, this sounds like a description of the advance of science via the normal method. All any of us can do is honestly report our findings and uncertainties, try to use our heads and add some interpretation (or else it is largely just a technological exercise) to guide followup studies. If they find new things out that make our work obsolete then so be it – we have helped push things forward.

Cheers,
Tas

]]> By: Anonymous http://landshape.org/enm/droughts-and-antarctica/#comment-12201 Anonymous Wed, 10 Feb 2010 06:36:00 +0000 http://landshape.org/enm/?p=3617#comment-12201 Tas, I appreciate you commenting as I have no interest to preaching to the choir, and a one-sided blog. It seems we are agreed that the reporting was, and still is, overblown. I think there is a need to speak out about this as it is damaging the reputation of science. The substantive issue on the significant change in circulation is something I have looked into with the Walker Circulation (http://landshape.org/enm/no-weakening-of-the-walker/). As I am sure you know, Power and Smith and Vecchi reported a weakening of the WC as evidence of AGW. However, since a run of strong El Ninos has eased the data indicate a return of the indices to normal and no longer support their claim to any degree, suggesting, what? We have seen before where these sorts of claims are reversed by subsequent data (eg http://landshape.org/enm/wp-content/uploads/2009/06/ee-20-4_7-stockwell2.pdf). I wonder if this case might be similar? I would be prepared to look into it . Would you post the data for your graph showing the inverse link between rainfall in south-west Western Australia and snowfall in Antarctica (http://www.aad.gov.au/MediaLibrary/asset/MediaItems/ml_402173825925926_03_RainSnowComparisonGraph.pdf) so we can check the statistics of the relationship? I admit I haven't read your paper as I didn't see a reference to it anywhere, but if you have a reference and a link I would appreciate it. Tas, I appreciate you commenting as I have no interest to preaching to the choir, and a one-sided blog. It seems we are agreed that the reporting was, and still is, overblown. I think there is a need to speak out about this as it is damaging the reputation of science.

The substantive issue on the significant change in circulation is something I have looked into with the Walker Circulation (http://landshape.org/enm/no-weakening-of-the-walker/). As I am sure you know, Power and Smith and Vecchi reported a weakening of the WC as evidence of AGW. However, since a run of strong El Ninos has eased the data indicate a return of the indices to normal and no longer support their claim to any degree, suggesting, what?

We have seen before where these sorts of claims are reversed by subsequent data (eg http://landshape.org/enm/wp-content/uploads/2009/06/ee-20-4_7-stockwell2.pdf).

I wonder if this case might be similar? I would be prepared to look into it . Would you post the data for your graph showing the inverse link between rainfall in south-west Western Australia and snowfall in Antarctica (http://www.aad.gov.au/MediaLibrary/asset/MediaItems/ml_402173825925926_03_RainSnowComparisonGraph.pdf) so we can check the statistics of the relationship?

I admit I haven’t read your paper as I didn’t see a reference to it anywhere, but if you have a reference and a link I would appreciate it.

]]>