3 ms·

After climategate I have a hard time giving much creedence to studies on either side of the anthropomorphic global warming issue. It seems like the right answer

After climategate I have a hard time giving much creedence to studies on either side of the anthropomorphic global warming issue. It seems like the right answer is "we don't know." That having been said, articles and books like this seem like not much more than a desperate attempt at staying relevant.

Retric 17y ago
I suspect that "climategate" did not change your stance from agreeing with climate change to disagreeing with it. However, if honestly changed your mind I must ask you why?
- yummyfajitas 17y ago
  I changed my stance on historical climate data as a result of climategate [1]. I previously accepted it as more or less correct, but now I'm very skeptical. The fact is, a big chunk of it cannot be reproduced. A big chunk of historical climate data was was created by the CRU group. The CRU folks are revealed to have an agenda which they are willing to push by subverting the scientific process. Perhaps other groups are better, or perhaps they just haven't had their email hacked. But the field of historical climate is small enough that there isn't much redundancy (only 4 or 5 major groups, AFAIK). In an experimental field this would be less of an issue, since experiments provide a reality check. But in a purely analytical field, it's murder. Furthermore, even works produced by completely honest and accurate groups may depend on flawed results coming out of the CRU group. Until we carefully audit the field, it's hard to be sure anything is legitimate. Furthermore, in the wake of climategate, I've become more educated on their data analysis techniques. Consider the "trick" used to "hide the decline": tree ring data agrees with temperature for the first half of the calibration period, but not the second. Therefore, we discard the second half, and assume the correlations in the first half of the training data always applied. This isn't fraud, since it's well documented what is being done. This is worse. This means that there are so few good proxy measurements that they are forced to use proxies which only match half the training data! I'm not saying this is necessarily bad science; it might be the best that can be done with the available data. Best possible != good, and in this case it looks like best possible might be very poor. [1] Full disclosure: I was previously skeptical of the claims made by climate modeling. If you wish to claim this is equivalent to "did not change your stance", feel free.
  - Retric 17y ago
    Ahh, ok. Yea, tree rings are a horrible proxy for temperature. "Global cooling"/"Global dimming" (small particulates suspended in the upper atmosphere) and an increase in pollution etc. has messed with them. I have long been more skeptical estimates of past global temperatures than our ability to predict future temperatures. Most of the other proxy’s for historic temperature are highly dependent on ocean currents and we don’t have a good proxy for figuring those out prior to accurate temperature measurements. Pluss we are trying to look further into the past than the future so the number man made changes to the enviernment (deforistation, urbon areas, the sahara etc) are even more extream IMO, the major issue dealing with global temperatures is actually how ocean currents effected and affect temperatures. You can model changes in gas concentrations but ocean currents estimates are closer to predicting weather than climate. Also, we have high quality mesurments over a range of CO2 consentrations, but so far global ocean currents have been fairly steady. PS: Global Cooling is actually a well accepted part of current clement change models; the assumption is the effect of particulates suspended in the atmosphere is not going to increase so while they offset a significant amount of global warming they are not going to increase anytime soon. However, looking at the past we need get an accurate estimate of the effects which is hard.
    - yummyfajitas 17y ago
      "Global cooling"/"Global dimming" (small particulates suspended in the upper atmosphere) and an increase in pollution etc. has messed with them [tree rings]. Actually, we don't know what factor messed up tree rings. If we knew that tree rings were f(warmth, x), we would simply keep tree ring data for the entire calibration period and include x in the calculations. The fact is, factor x might apply in the past. We simply don't know. I'm curious though; if you are skeptical of our knowledge of past temperatures, how can you not be skeptical of predictions of future temperatures? All the climate models are calibrated to past temperature records, so how can the models give good results if the past records are bad? A hypothetical: I come to you with a proposed algorithmic trading strategy which I've validated it against historical data. But oops I'm not that great at keeping records, and I actually threw away the market data. I also emailed my employees and told them to delete the data if the SEC ever came asking for it. And I also ignored some historical periods where the numbers didn't work. But I kept some stock charts showing exponential growth! So, you want to invest in my fund?
gjm11 17y ago
If you think that "climategate" throws much doubt on the main claims of anthropogenic global warming, then either you've been misinformed about what's in the emails or I have. (I think the "climategate" emails establish that some climate scientists aren't very nice people sometimes, that some software written by research scientists is a bit of a mess, and that if you cherry-pick the worst-sounding bits from hundreds of emails you can make things sound worse than they really are. Not much more than that.) I don't understand. Friel hasn't been particularly associated with global warming in the past; why would he be writing a book like this as "a desperate attempt at staying relevant"? Lomborg gets plenty of attention for his work. If, as Friel claims, it's full of dishonest pseudo-scholarship, isn't that worth drawing attention to?

Retric 17y ago

I suspect that "climategate" did not change your stance from agreeing with climate change to disagreeing with it. However, if honestly changed your mind I must ask you why?

yummyfajitas 17y ago
I changed my stance on historical climate data as a result of climategate [1]. I previously accepted it as more or less correct, but now I'm very skeptical. The fact is, a big chunk of it cannot be reproduced. A big chunk of historical climate data was was created by the CRU group. The CRU folks are revealed to have an agenda which they are willing to push by subverting the scientific process. Perhaps other groups are better, or perhaps they just haven't had their email hacked. But the field of historical climate is small enough that there isn't much redundancy (only 4 or 5 major groups, AFAIK). In an experimental field this would be less of an issue, since experiments provide a reality check. But in a purely analytical field, it's murder. Furthermore, even works produced by completely honest and accurate groups may depend on flawed results coming out of the CRU group. Until we carefully audit the field, it's hard to be sure anything is legitimate. Furthermore, in the wake of climategate, I've become more educated on their data analysis techniques. Consider the "trick" used to "hide the decline": tree ring data agrees with temperature for the first half of the calibration period, but not the second. Therefore, we discard the second half, and assume the correlations in the first half of the training data always applied. This isn't fraud, since it's well documented what is being done. This is worse. This means that there are so few good proxy measurements that they are forced to use proxies which only match half the training data! I'm not saying this is necessarily bad science; it might be the best that can be done with the available data. Best possible != good, and in this case it looks like best possible might be very poor. [1] Full disclosure: I was previously skeptical of the claims made by climate modeling. If you wish to claim this is equivalent to "did not change your stance", feel free.
- Retric 17y ago
  Ahh, ok. Yea, tree rings are a horrible proxy for temperature. "Global cooling"/"Global dimming" (small particulates suspended in the upper atmosphere) and an increase in pollution etc. has messed with them. I have long been more skeptical estimates of past global temperatures than our ability to predict future temperatures. Most of the other proxy’s for historic temperature are highly dependent on ocean currents and we don’t have a good proxy for figuring those out prior to accurate temperature measurements. Pluss we are trying to look further into the past than the future so the number man made changes to the enviernment (deforistation, urbon areas, the sahara etc) are even more extream IMO, the major issue dealing with global temperatures is actually how ocean currents effected and affect temperatures. You can model changes in gas concentrations but ocean currents estimates are closer to predicting weather than climate. Also, we have high quality mesurments over a range of CO2 consentrations, but so far global ocean currents have been fairly steady. PS: Global Cooling is actually a well accepted part of current clement change models; the assumption is the effect of particulates suspended in the atmosphere is not going to increase so while they offset a significant amount of global warming they are not going to increase anytime soon. However, looking at the past we need get an accurate estimate of the effects which is hard.
  - yummyfajitas 17y ago
    "Global cooling"/"Global dimming" (small particulates suspended in the upper atmosphere) and an increase in pollution etc. has messed with them [tree rings]. Actually, we don't know what factor messed up tree rings. If we knew that tree rings were f(warmth, x), we would simply keep tree ring data for the entire calibration period and include x in the calculations. The fact is, factor x might apply in the past. We simply don't know. I'm curious though; if you are skeptical of our knowledge of past temperatures, how can you not be skeptical of predictions of future temperatures? All the climate models are calibrated to past temperature records, so how can the models give good results if the past records are bad? A hypothetical: I come to you with a proposed algorithmic trading strategy which I've validated it against historical data. But oops I'm not that great at keeping records, and I actually threw away the market data. I also emailed my employees and told them to delete the data if the SEC ever came asking for it. And I also ignored some historical periods where the numbers didn't work. But I kept some stock charts showing exponential growth! So, you want to invest in my fund?

gjm11 17y ago

If you think that "climategate" throws much doubt on the main claims of anthropogenic global warming, then either you've been misinformed about what's in the emails or I have. (I think the "climategate" emails establish that some climate scientists aren't very nice people sometimes, that some software written by research scientists is a bit of a mess, and that if you cherry-pick the worst-sounding bits from hundreds of emails you can make things sound worse than they really are. Not much more than that.) I don't understand. Friel hasn't been particularly associated with global warming in the past; why would he be writing a book like this as "a desperate attempt at staying relevant"? Lomborg gets plenty of attention for his work. If, as Friel claims, it's full of dishonest pseudo-scholarship, isn't that worth drawing attention to?