Regression Conclusion

Is Kirchhoff’s rule in atmosphere proven by the Cabauw measurements, or not? I had earlier received a note on calculations by Noor van Andel claiming, yes, Kirchoff’s relationship as used in Miskolczi’s theory was confirmed by the linear regression of Ed (longwave down radiation) and Su(1-Ta) (longwave surface up, without transmitted longwave). Miskolczi also confirmed this result, using older results in a the previous post in this series.

However, I just received an email from Rob van Dorland who took the actual measurements claiming this is not the case. In Rob’s opinion:

The calculations show “Kirchhoff’s rule” cannot be obtained using my measurements. These measurements consist of downward radiation at the surface and the mixing ratio’s of water vapor at the surface (that is what I sent to Noor van Andel). From this you cannot compute the atmospheric LW transparency (Ta), like Noor van Andel did, because you need information on profiles of temperature, water vapor and of other greenhouse gases. Of course, I used this profile information to compute the downward radiation at the surface and compare these computed values with the observed ones (see attachment).

From Thesis Rob van Dorland, 1999.

I think the computation of Kirchhoff’s rule by Noor illustrates very well that assumptions (which are not a standard part of radiative transfer calculations) are put into the model of Noor (and probably also in Ferenc’s model). In turn these assumptions result to my opinion in wrong conclusions such as the optimum optical depth for the present Earth’s atmosphere and the very low sensitivity to CO2 changes, stated in Ferenc paper.

Rob mentions for background information you can download the pdf of chapter 3 of his thesis. Click on “deel 2″ at the end of this page after the reference to the thesis.

I must admit I thought Noor’s results were based on profile measurements at the 200m high Cabauw tower, near Lopik, the Netherlands. Unfortunately I have been unable to contact Noor to gain his approval to publish the findings on the web.

Update: Rob has sent details of the data and the data itself he sent to Noor. I think it clears up the profile issue, as I (mis)understood Noor’s note to read that he calculated the profiles. The data sent by Rob is here, and email below.

Dear David
See my brief response below
regards
Rob

David Stockwell wrote:

>Dear Rob,

>Thanks very much. It takes a disagreement to provoke discussion on the web, you are right!
>Still I think we learn more in an interesting way when that happens.
>I am interested in your statement that you did not send Noor temperature and water
>vapor profile information, as he says clearly in his note that the calculations of
>Ta are based on profile information sent by you. I am just a bit confused.

Noor only got the surface values. please ask him. I send you the same file as I did in May to Noor.
column 3: 2m temperature
column 4: 2m humidity
column 5: Longwave downward measured at the surface
column 6: Longwave downward calculated at the surface

lwdmeas_calc_van_dorland

>”The 200 m high radio broadcast transmitter in Cabauw, near Lopik, the Netherlands,
>can be used, like a weather balloon, to measure atmosphere profiles,
>albeit only until 200 m high. Rob van Dorland has measured these profiles. ”

I actually combined the 200 m tower measurements with the balloon measurements. So I got profiles of temperature and humidity up til a height of 25 to 30 km. Further details of my comparison can be read in paragraph 3.5 of my thesis.

About these ads

0 thoughts on “Regression Conclusion

  1. David S

    “I must admit I thought Noor’s results were based on profile measurements at the 200m high Cabauw tower, near Lopik, the Netherlands. ”

    I got the same impression and see no reason just yet to change this, from the abstract:

    “We evaluate this scheme with surface radiation measurements at Cabauw in the Netherlands for 253 clear sky hours during the period 1995-1996. As input for the scheme we combine the atmospheric profiles of temperature and humidity from the Cabauw tower (200m) with rawind sonde measurements at De Bilt.”

    I also don’t quite understand what he means by this:

    ” From this you cannot compute the atmospheric LW transparency (Ta), like Noor van Andel did, because you need information on profiles of temperature, water vapor and of other greenhouse gases.”

    If Noor used programs similar to HARTCODE then he used information Rob tries to imply he did not have. Whether he did or not I don’t know, but H2O vapour is there in such large quantity, and is a strong absorber that overlaps with CO2, so just using that will be a fairly close approximation any way.

    Nevertheless Ferenc used the TIGR profiles that include mixing ratios for H2O, CO2, O3, N2O, CO, CH4, and O2 in ppmv and M7M 2004 used 11 absorbing species.

    In neither case are assumptions made, as far as I can see, about radiative transfer that are non standard though I have seen numerous attempts to cast them as such.

  2. David S

    “I must admit I thought Noor’s results were based on profile measurements at the 200m high Cabauw tower, near Lopik, the Netherlands. ”

    I got the same impression and see no reason just yet to change this, from the abstract:

    “We evaluate this scheme with surface radiation measurements at Cabauw in the Netherlands for 253 clear sky hours during the period 1995-1996. As input for the scheme we combine the atmospheric profiles of temperature and humidity from the Cabauw tower (200m) with rawind sonde measurements at De Bilt.”

    I also don’t quite understand what he means by this:

    ” From this you cannot compute the atmospheric LW transparency (Ta), like Noor van Andel did, because you need information on profiles of temperature, water vapor and of other greenhouse gases.”

    If Noor used programs similar to HARTCODE then he used information Rob tries to imply he did not have. Whether he did or not I don’t know, but H2O vapour is there in such large quantity, and is a strong absorber that overlaps with CO2, so just using that will be a fairly close approximation any way.

    Nevertheless Ferenc used the TIGR profiles that include mixing ratios for H2O, CO2, O3, N2O, CO, CH4, and O2 in ppmv and M7M 2004 used 11 absorbing species.

    In neither case are assumptions made, as far as I can see, about radiative transfer that are non standard though I have seen numerous attempts to cast them as such.

  3. Thanks Jan. Fixed. One thing they both agree on; KL is crucial to Miskolczi’s important global warming results. I would like to understand it all a lot better. Right now, I am not sure what it is about Rob and Noor’s graphs that prove or disprove KL. Is it becuase the correlation is lower in Noor’s? In that case, what is the critical correlation that would disprove?

    That is apart from the actual provenance issue. I don’t want to embarrass anybody. I just want to understand and get closer to the truth.

  4. Thanks Jan. Fixed. One thing they both agree on; KL is crucial to Miskolczi’s important global warming results. I would like to understand it all a lot better. Right now, I am not sure what it is about Rob and Noor’s graphs that prove or disprove KL. Is it becuase the correlation is lower in Noor’s? In that case, what is the critical correlation that would disprove?

    That is apart from the actual provenance issue. I don’t want to embarrass anybody. I just want to understand and get closer to the truth.

  5. David S,

    “I just want to understand and get closer to the truth.”

    I think the same could be said for most of us including Ferenc.

    It’s not the poorer correlation of Noor’s graph which is quite comprehensible if Noor water vapour as the only absorber as it will mean each “measurement” is less precise. Neither graph disproves KL. Rob’s graph does not address it as it compares computed downward radiation with that measured by pyrgeometer. It just confirms the validity of the computed values.

    It’s the computed values from the profiles (validated by Rob) that Noor and Ferenc use and plot against computed values of absorbed radiation from the surface (or near the surface). This does address and tends to prove KL for lower atmosphere.

    KL is just a common sense restatement of conservation of energy in a purely radiative setting. If you have local thermodynamic equilibrium all heat fluxes some to zero this includes radiative fluxes and if the radiativc flux sums to zero the temperature remains steady (thermal equilibrium) and the absorptivity = emissivity.

    In Ferenc’s case he uses the data available in the TIGR profiles which does contain data of the 7 major atmospheric absorbers so I’d expect greater precision hence better correlation.

    The upshot is that I don’t really undersand Rob’s objection.

    The only thing that bothers me is that the “surface” is 1.5 m above the surface.

  6. David S,

    “I just want to understand and get closer to the truth.”

    I think the same could be said for most of us including Ferenc.

    It’s not the poorer correlation of Noor’s graph which is quite comprehensible if Noor water vapour as the only absorber as it will mean each “measurement” is less precise. Neither graph disproves KL. Rob’s graph does not address it as it compares computed downward radiation with that measured by pyrgeometer. It just confirms the validity of the computed values.

    It’s the computed values from the profiles (validated by Rob) that Noor and Ferenc use and plot against computed values of absorbed radiation from the surface (or near the surface). This does address and tends to prove KL for lower atmosphere.

    KL is just a common sense restatement of conservation of energy in a purely radiative setting. If you have local thermodynamic equilibrium all heat fluxes some to zero this includes radiative fluxes and if the radiativc flux sums to zero the temperature remains steady (thermal equilibrium) and the absorptivity = emissivity.

    In Ferenc’s case he uses the data available in the TIGR profiles which does contain data of the 7 major atmospheric absorbers so I’d expect greater precision hence better correlation.

    The upshot is that I don’t really undersand Rob’s objection.

    The only thing that bothers me is that the “surface” is 1.5 m above the surface.

  7. Jan, Yes that follows from the axes of the graphs. Perhaps Rob will clarify.

    As to KL, it makes sense to me too, if the atmosphere is sufficiently ‘opaque’ to IR then it would be like a Kirchhoff cavity. If the semi-infinite theory held, there would be a discontinuity with the surface. Can you answer, what should the graph of DLW vs Su(1-Ta) look like for the semi-infinite case? My lack of background in the area stops me from taking any steps on my own.

    In an earlier post I got from KL (Su=Ed in the cloudy case) and M’s eqn 7 to Su=3F/2 in one step. Alternatively, from Ed=Eu, a top of atmosphere constraint, and eqn 7 you get to Su=2F. (http://landshape.org/enm/models-of-greenhouse-effect/)

    In the first case, with the surface constraint, the maximum greenhouse effect is no greater that the present day. In the second case, it is much greater (possibly a Venus atmosphere, which is opaque to SW as well, I don’t know). Is this a valid interpretation of the importance of the KL assumption, Jan?

    Sorry, I am a simple minded kind of guy.

  8. Jan, Yes that follows from the axes of the graphs. Perhaps Rob will clarify.

    As to KL, it makes sense to me too, if the atmosphere is sufficiently ‘opaque’ to IR then it would be like a Kirchhoff cavity. If the semi-infinite theory held, there would be a discontinuity with the surface. Can you answer, what should the graph of DLW vs Su(1-Ta) look like for the semi-infinite case? My lack of background in the area stops me from taking any steps on my own.

    In an earlier post I got from KL (Su=Ed in the cloudy case) and M’s eqn 7 to Su=3F/2 in one step. Alternatively, from Ed=Eu, a top of atmosphere constraint, and eqn 7 you get to Su=2F. (http://landshape.org/enm/models-of-greenhouse-effect/)

    In the first case, with the surface constraint, the maximum greenhouse effect is no greater that the present day. In the second case, it is much greater (possibly a Venus atmosphere, which is opaque to SW as well, I don’t know). Is this a valid interpretation of the importance of the KL assumption, Jan?

    Sorry, I am a simple minded kind of guy.

  9. Dave, Jan:
    There are standard equations to estimate the h2o from surface observations, and also there are standard relationships to estimate the flux transmittance from the total h2o amount. Jan is right, the scatter in Noor’s plot is probably from the h2o (prcm) conversion to flux transmittance Ta. See Fig. 7 in the M04 paper and the related (first) equation in Table 4.

  10. Dave, Jan:
    There are standard equations to estimate the h2o from surface observations, and also there are standard relationships to estimate the flux transmittance from the total h2o amount. Jan is right, the scatter in Noor’s plot is probably from the h2o (prcm) conversion to flux transmittance Ta. See Fig. 7 in the M04 paper and the related (first) equation in Table 4.

  11. David

    “Sorry, I am a simple minded kind of guy.”

    LOL Yeah right! If I had a fraction of your ability with statistics I’d be delighted.

    “Can you answer, what should the graph of DLW vs Su(1-Ta) look like for the semi-infinite case?”

    With a vertical (downward) slope of temperature to atmosphere I would expect that the most important feature will be a line that does not pass through the origin, at some point there will be zero LWD for finite positive SU(1-Ta) i.e. Aa = Su(1 -Ta) > LWD.

    “Is this a valid interpretation of the importance of the KL assumption, Jan?”

    I think so, your diagram for Ed=Eu (constrained at the top) has much in common the “the higher, the cooler” scenario often seen and invoked in Held & Soden Annual Reviews 2000 “WATER VAPOR FEEDBACK AND GLOBAL WARMING”

    It’s behind a money wall but I have a copy.

    Of greater importance however is I believe that it’s what is observed as opposed to a temperature discontinuity which is not.

  12. David

    “Sorry, I am a simple minded kind of guy.”

    LOL Yeah right! If I had a fraction of your ability with statistics I’d be delighted.

    “Can you answer, what should the graph of DLW vs Su(1-Ta) look like for the semi-infinite case?”

    With a vertical (downward) slope of temperature to atmosphere I would expect that the most important feature will be a line that does not pass through the origin, at some point there will be zero LWD for finite positive SU(1-Ta) i.e. Aa = Su(1 -Ta) > LWD.

    “Is this a valid interpretation of the importance of the KL assumption, Jan?”

    I think so, your diagram for Ed=Eu (constrained at the top) has much in common the “the higher, the cooler” scenario often seen and invoked in Held & Soden Annual Reviews 2000 “WATER VAPOR FEEDBACK AND GLOBAL WARMING”

    It’s behind a money wall but I have a copy.

    Of greater importance however is I believe that it’s what is observed as opposed to a temperature discontinuity which is not.

  13. “the most important feature will be a line that does not pass through the origin,”
    There does seem to be a deflection upward at the low end of the LWD. I would think this would be a crucial experiment to perform, and distinguish between the two with a test of non-zero intercept.

    Yes, “higher the cooler” seems like a way of expressing a fundamental constraint acting at top of atmosphere. Is this why the GCMs show elevated upper tropospheres.

  14. “the most important feature will be a line that does not pass through the origin,”
    There does seem to be a deflection upward at the low end of the LWD. I would think this would be a crucial experiment to perform, and distinguish between the two with a test of non-zero intercept.

    Yes, “higher the cooler” seems like a way of expressing a fundamental constraint acting at top of atmosphere. Is this why the GCMs show elevated upper tropospheres.

  15. “There does seem to be a deflection upward at the low end of the LWD. I would think this would be a crucial experiment to perform,”

    I’ll see what I can do. I am getting bits and pieces of hardware together to do some experimenting and Ferenc has kindly offered to run my results through HARTCODE. I’m not rich so it will take a little time. I agree it’s important.

    While I’m not absolutely sure that linearity will be retained at the low end I do expect an offset to be obvious over the whole range. The reason I don’t expect strict linearity is that I suspect results will be somewhat skewed at the low end because of lowering water vapour concentration.

    It’s good to get the clarification from Rob: so while the profiles he took should be adequate to calculate the absorbed vs emitted, but it’s difficult to see how what he sent to Noor is.

    “Is this why the GCMs show elevated upper tropospheres?”

    I suspect so but have never seen it spelt out in any detail.

    FYI my background is mainly instrumentation, feedback and control.

  16. “There does seem to be a deflection upward at the low end of the LWD. I would think this would be a crucial experiment to perform,”

    I’ll see what I can do. I am getting bits and pieces of hardware together to do some experimenting and Ferenc has kindly offered to run my results through HARTCODE. I’m not rich so it will take a little time. I agree it’s important.

    While I’m not absolutely sure that linearity will be retained at the low end I do expect an offset to be obvious over the whole range. The reason I don’t expect strict linearity is that I suspect results will be somewhat skewed at the low end because of lowering water vapour concentration.

    It’s good to get the clarification from Rob: so while the profiles he took should be adequate to calculate the absorbed vs emitted, but it’s difficult to see how what he sent to Noor is.

    “Is this why the GCMs show elevated upper tropospheres?”

    I suspect so but have never seen it spelt out in any detail.

    FYI my background is mainly instrumentation, feedback and control.

  17. That sounds great. Instead of all this endless theorizing just design the crucial experiment and be done with it.

    It would be good to get an idea of the accuracy and number of points you need to collect to get a result either way.

  18. That sounds great. Instead of all this endless theorizing just design the crucial experiment and be done with it.

    It would be good to get an idea of the accuracy and number of points you need to collect to get a result either way.

  19. “It would be good to get an idea of the accuracy and number of points you need to collect to get a result either way.”

    Yes, I’ll have to do more preliminary work, and will definitely be open to suggestions. I wasn’t planning to but Ferenc has suggested to take humidity readings that he will need so that’s what I’ll do. Presently then I’m planning a low level (2m but might be too low) profile over day/night with surface temperature measurement.

    For the surface temperature I’m hovering between Pt100 probe that I dug up in my shed, and non-contact thermometer that I will have to purchase or make .

  20. “It would be good to get an idea of the accuracy and number of points you need to collect to get a result either way.”

    Yes, I’ll have to do more preliminary work, and will definitely be open to suggestions. I wasn’t planning to but Ferenc has suggested to take humidity readings that he will need so that’s what I’ll do. Presently then I’m planning a low level (2m but might be too low) profile over day/night with surface temperature measurement.

    For the surface temperature I’m hovering between Pt100 probe that I dug up in my shed, and non-contact thermometer that I will have to purchase or make .

  21. Well, I could try to work out the power of test needed. If you write a one page description of what you are trying to do, I will try to work out the experimental parameters.

  22. Well, I could try to work out the power of test needed. If you write a one page description of what you are trying to do, I will try to work out the experimental parameters.

  23. Pingback: website
  24. Pingback: escort vienna
  25. Pingback: XXX Videos
  26. Pingback: wynajem samochodow
  27. Pingback: strona firmy
  28. Pingback: livecooking
  29. Pingback: Nue Design
  30. Pingback: kliknij link
  31. Pingback: seo
  32. Pingback: carlos siderman
  33. Pingback: investir ehpad
  34. Pingback: link
  35. Pingback: blood testing
  36. Pingback: strona www
  37. Pingback: XXX
  38. Pingback: Moyou london
  39. Pingback: Maria Divine Mercy
  40. Pingback: pinganillo
  41. Pingback: lucrare licenta
  42. Pingback: tomelloso
  43. Pingback: tomelloso
  44. Pingback: birmingham escorts
  45. Pingback: birmingham escorts
  46. Pingback: Great Clips Prices
  47. Pingback: alcudia
  48. Pingback: steroid
  49. Pingback: ????? ????
  50. Pingback: ????? ???? ???????
  51. Pingback: penapis air murah
  52. Pingback: Kayak Fishing
  53. Pingback: Camilo Concha
  54. Pingback: arganolie
  55. Pingback: baby expo

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s