3 ms·

I am not a computational physicist but I had the privilege of being lectured to by one who worked on multiple environmental models. The key takeaway was that t

by iaw 12y ago

I am not a computational physicist but I had the privilege of being lectured to by one who worked on multiple environmental models. The key takeaway was that the models aren't truly predictive, they had to do a substantial amount of tweaking to get the model to line with historic in sample data but could never expand the model to out sample data. A larger point that was raised by a different lecturer is that solar output has increased drastically over the last 100 years and that the global warming trends we see correlate perfectly with this increase in output. He actually believed that the industrial revolution lining up with this increase in solar output was entirely coincidental. So to confound the issue further, we may not be responsible for global warming but our production of greenhouse gasses undoubtedly has an impact on our environment/climate that seems unlikely to be positive. Climate science is a nasty beast.

quink 12y ago
http://www.skepticalscience.com/solar-activity-sunspots-global-warming-advanced.htm http://www.skepticalscience.com/solar-activity-sunspots-glob... In a nutshell, CO2 is near enough on the path to increasing by 50% and it's pretty important for the greenhouse effect. Solar variation over millenia is, peak-to-peak, 0.1%.
- iaw 12y ago
  That lambda value in the equation under Figure 2 is their 'fudge factor'. How do you quantify that value when we only have one complex system with multiple inputs to study?
a8da6b0c91d 12y ago
Do you know much about the models they use? It all seems kind of silly to me because as I understand it they use neither monte-carlo simulation nor stochastic calculus. The models don't spit out anything about probabilities. They only describe the one outcome your starting assumptions produce. It's a glorified excel spreadsheet.
- quink 12y ago
  It's kind of like that. In three ways that I can think of. The models are run hundreds or thousands of times across different countries. The climate, modelled on a day by day basis, is effectively random for any particular instant. And the input variables, where they do change even minutely, act like seeds in a random number generator. It's not so much a glorified Excel spreadsheet as it is having thousands of Excel spreadsheets, each dependent on the one before and a slightly different scenario that's different for every set of these Excel sheets. They are running these on super-computers many many times over and would have no interest in doing exactly the same thing over and over. http://www.climatechange2013.org/images/report/WG1AR5_Chapter09_FINAL.pdf http://www.climatechange2013.org/images/report/WG1AR5_Chapte... And yes, they even mention the 15 year hiatus - they don't sweep this under the rug like so many have taken to believing they do. In fact, it's mentioned throughout the AR5. In fact, you can get a sense of them being just eager to tackle it in substantial detail - it's wonderful.
  - a8da6b0c91d 12y ago
    That pdf is a great link, thanks. I've gone looking for this information before and failed to find it.
jjoonathan 12y ago
Solar forcing as an explanation for AGW has been tested and rejected by the scientific consensus. The scientific consensus could certainly be wrong, but it's a huge red flag if anyone implies that it hasn't been given serious consideration by mainstream climate science, that it has been swept under the rug, etc, because that couldn't be further from the truth. Pages 188-193 summarize why the scientific consensus believes what it does: http://www.ipcc.ch/pdf/assessment-report/ar4/wg1/ar4-wg1-chapter2.pdf http://www.ipcc.ch/pdf/assessment-report/ar4/wg1/ar4-wg1-cha...
- iaw 12y ago
  I just did a back of the envelope calculation. A change of 0.5 W/m^2 would yield (in equilibrium) an increase of 0.4 degrees C based on basic heat transfer. I see them dismissing the significance of solar irradiance but no where do I see them explain their methodology for determining the relationship of solar irradiance to temperature. 1% of 1300 W/m^2 is still a significant amount of extra energy that the planet needs to dissipate. Also, while the lead author of that paper is a physicist, looking at his published history he seems firmly biased into the realm of particulate/aerosol driven climate impacts. It would be against his academic interests to publish a paper that at all discounts his historic work. If you have any knowledge of papers that try to quantify and link solar irradiance changes to global climate change I'd be interested in reading over their methodology. I just didn't see that in here (correct me if I'm wrong).
  - jjoonathan 12y ago
    Yeah, they don't discuss methodology in the summaries, and it's a PITA to chase down their references. It looks slightly easier in AR5 so I'll switch to that one. First of all, did you multiply by the geometric and albedo factors (0.25 and 0.7 respectively)? Because 0.5W/m^2 * 0.25 * 0.7 is 0.9W/m^2 which is within the range given in 5AR in the context of this energy-balance comparison (made in units of radiative forcing): Effective RF due to all anthropogenic effects: 1.1-3.3W/m^2 Effective RF due to greenhouse gasses: 2.54-3.12W/m^2 Effective RF due to non-CO2 GHGs: 0.43-0.53 W/m^2 RF due to Solar Forcing: 0.05-0.10W/m^2 These numbers indirectly answer your question (solar forcing is estimated to only be accountable for a fraction of the ~1ºC warming we have seen thus far) but I'm afraid that the detailed methodology is buried in a completely different section of the report (probably because it involves models of steady-state equilibrium between energy reservoirs now). I don't have the time to dig it up today and I'm not a subject expert, I am just loath to let people get away with the "scientists haven't considered..." type arguments when they're so blatantly wrong. 5AR (solar forcing on page 688): http://www.climatechange2013.org/images/report/WG1AR5_Chapter08_FINAL.pdf http://www.climatechange2013.org/images/report/WG1AR5_Chapte...

by iaw 12y ago

quink 12y ago

http://www.skepticalscience.com/solar-activity-sunspots-global-warming-advanced.htm http://www.skepticalscience.com/solar-activity-sunspots-glob... In a nutshell, CO2 is near enough on the path to increasing by 50% and it's pretty important for the greenhouse effect. Solar variation over millenia is, peak-to-peak, 0.1%.

iaw 12y ago
That lambda value in the equation under Figure 2 is their 'fudge factor'. How do you quantify that value when we only have one complex system with multiple inputs to study?

a8da6b0c91d 12y ago

Do you know much about the models they use? It all seems kind of silly to me because as I understand it they use neither monte-carlo simulation nor stochastic calculus. The models don't spit out anything about probabilities. They only describe the one outcome your starting assumptions produce. It's a glorified excel spreadsheet.

quink 12y ago
It's kind of like that. In three ways that I can think of. The models are run hundreds or thousands of times across different countries. The climate, modelled on a day by day basis, is effectively random for any particular instant. And the input variables, where they do change even minutely, act like seeds in a random number generator. It's not so much a glorified Excel spreadsheet as it is having thousands of Excel spreadsheets, each dependent on the one before and a slightly different scenario that's different for every set of these Excel sheets. They are running these on super-computers many many times over and would have no interest in doing exactly the same thing over and over. http://www.climatechange2013.org/images/report/WG1AR5_Chapter09_FINAL.pdf http://www.climatechange2013.org/images/report/WG1AR5_Chapte... And yes, they even mention the 15 year hiatus - they don't sweep this under the rug like so many have taken to believing they do. In fact, it's mentioned throughout the AR5. In fact, you can get a sense of them being just eager to tackle it in substantial detail - it's wonderful.
- a8da6b0c91d 12y ago
  That pdf is a great link, thanks. I've gone looking for this information before and failed to find it.

jjoonathan 12y ago

Solar forcing as an explanation for AGW has been tested and rejected by the scientific consensus. The scientific consensus could certainly be wrong, but it's a huge red flag if anyone implies that it hasn't been given serious consideration by mainstream climate science, that it has been swept under the rug, etc, because that couldn't be further from the truth. Pages 188-193 summarize why the scientific consensus believes what it does: http://www.ipcc.ch/pdf/assessment-report/ar4/wg1/ar4-wg1-chapter2.pdf http://www.ipcc.ch/pdf/assessment-report/ar4/wg1/ar4-wg1-cha...

iaw 12y ago
I just did a back of the envelope calculation. A change of 0.5 W/m^2 would yield (in equilibrium) an increase of 0.4 degrees C based on basic heat transfer. I see them dismissing the significance of solar irradiance but no where do I see them explain their methodology for determining the relationship of solar irradiance to temperature. 1% of 1300 W/m^2 is still a significant amount of extra energy that the planet needs to dissipate. Also, while the lead author of that paper is a physicist, looking at his published history he seems firmly biased into the realm of particulate/aerosol driven climate impacts. It would be against his academic interests to publish a paper that at all discounts his historic work. If you have any knowledge of papers that try to quantify and link solar irradiance changes to global climate change I'd be interested in reading over their methodology. I just didn't see that in here (correct me if I'm wrong).
- jjoonathan 12y ago
  Yeah, they don't discuss methodology in the summaries, and it's a PITA to chase down their references. It looks slightly easier in AR5 so I'll switch to that one. First of all, did you multiply by the geometric and albedo factors (0.25 and 0.7 respectively)? Because 0.5W/m^2 * 0.25 * 0.7 is 0.9W/m^2 which is within the range given in 5AR in the context of this energy-balance comparison (made in units of radiative forcing): Effective RF due to all anthropogenic effects: 1.1-3.3W/m^2 Effective RF due to greenhouse gasses: 2.54-3.12W/m^2 Effective RF due to non-CO2 GHGs: 0.43-0.53 W/m^2 RF due to Solar Forcing: 0.05-0.10W/m^2 These numbers indirectly answer your question (solar forcing is estimated to only be accountable for a fraction of the ~1ºC warming we have seen thus far) but I'm afraid that the detailed methodology is buried in a completely different section of the report (probably because it involves models of steady-state equilibrium between energy reservoirs now). I don't have the time to dig it up today and I'm not a subject expert, I am just loath to let people get away with the "scientists haven't considered..." type arguments when they're so blatantly wrong. 5AR (solar forcing on page 688): http://www.climatechange2013.org/images/report/WG1AR5_Chapter08_FINAL.pdf http://www.climatechange2013.org/images/report/WG1AR5_Chapte...