Saturn model (+ a bit of Uranus)
Converted Jovian interactive interface to Kronian model use. Split up cross-section common block so it can be increased in size without contravening fortran limits, as happened with the Uranus model. If successful, the expended cross-section block will be transferred to the Uranus model too, as it is the same species at the same energies, doing the same things.
First run shows altering the cross-sections didn't effect the final result, which is good. A little confusion when comparing new 12keV rates with old 12keV rates explained when I realised I was comparing a new monochromatic run with an old maxwellian. Such is life...
Can't upload images to blogger - it doesn't like tiffs. Don't blame it.
Just checked Uranus, I must've sorted out the cross-section problem earlier on. Strange. This was prelog days. Uranian results show a lower rate of ionisation spread over a larger altitude range, but only slightly larger. Would be better if I could switch off the hydrocarbons again and see how they've affected things.
Oops, inserted zeros at wrong place - this model looks spectacular whilst crashing horribly...
Of course, it would help if I used an electron energy that penetrated into the hydrocarbon region... my mind is on Saturn!
Doesn't look good - 30keV creates a lower ionisation rate than 12keV - just found cross-section file, it wasn't sorted out!
Found it - problem in hit.f, when converting cross-section array from 3 to 6 species, I incremented the number of interactions, not the number of species, hence wrecking the common block. 100keV profile much improved... problem with H having a low but constant level of ionisation - including in places where there are no H's to be found... reminiscent of oxygen on Jupiter problem in the original RIDE. But that was sorted, wasn't it?
12keV more penetrating than 100keV - why did I go back to this program? I'll see the Hydrocarbon effect, then return to Saturn. Right, think I know what's going on - everything that hits the bottom must be bouncing back. This explains the deeply penetrating tail of the 30keV shot, why everything stops at a certain alt and why hydrocarbons have so little (ie no) effect... will deal with when the time comes... anyway. Saturn... after just... no that didn't work, H problem remains. But only above certain energies. This is the eternal problem of PDFs, whether the next probability up has any value if it is zero as well as the current probability. "Less than" on each probability sorts it out until the last probability, which loses one bin of bandwidth if less than is put on the decision making thing. How to be fair to all sides? Sorted it, did less than on all bins and cleared out H totally. Really should've kept some, but the atmosphere is discontinuous in the region where H is expected. Why don't the observationalists do a more thorough job with our blue friend up there? Perhaps an estimate of the missing proton column density could be helpful in some way... ie an output to all this? We shall see. Subject closed for now.
Saturn
Now worried about saturn. Testing for H problem
shitshitshitshitshit
resolving H problem...
who wrote this model?
H problem resolved
Testing for penetration problem
shitshitshitshitshit
I would expect a 100keV electron to be a little more penetrating than that! I would expect a 50keV electron to at least make the effort... Can't get away from it, these new cross-sections need to be in up and running. What on Earth possessed Achilleos to leave so much out of the JIM listings? Steve thinks there was a quality control exercise going on - which cross-sections are important, which aren't. Getting back to reality, probably true that the search for cross-sections was harder in the mid to late 1990's than in 2006... Especially since JIM had other bits to concentrate on. Did no-one think to check afterwards? I guess they did given that other JIM paper, but not hard enough, apparently...
The hydrocarbon cross-sections exist as a single spreadsheet. Alledgedly, so do the JIM ones, but since I've added and altered them, they now exist over several. Need to go back, put them together, consolidate in a single sheet all the RIDE cross-sections - yes again... must be the third time, and I still won't give it a name I remember.
emails keep coming from alan. Must be pleasant lying on a Turkish beach, just occasionally pricking a student's conscience with another part of their project whenever you feel like it... pastoral care...
uranushyrdoc2 set off, to see if I can improve the profile with a big run...
...and that's it for today
First run shows altering the cross-sections didn't effect the final result, which is good. A little confusion when comparing new 12keV rates with old 12keV rates explained when I realised I was comparing a new monochromatic run with an old maxwellian. Such is life...
Can't upload images to blogger - it doesn't like tiffs. Don't blame it.
Just checked Uranus, I must've sorted out the cross-section problem earlier on. Strange. This was prelog days. Uranian results show a lower rate of ionisation spread over a larger altitude range, but only slightly larger. Would be better if I could switch off the hydrocarbons again and see how they've affected things.
Oops, inserted zeros at wrong place - this model looks spectacular whilst crashing horribly...
Of course, it would help if I used an electron energy that penetrated into the hydrocarbon region... my mind is on Saturn!
Doesn't look good - 30keV creates a lower ionisation rate than 12keV - just found cross-section file, it wasn't sorted out!
Found it - problem in hit.f, when converting cross-section array from 3 to 6 species, I incremented the number of interactions, not the number of species, hence wrecking the common block. 100keV profile much improved... problem with H having a low but constant level of ionisation - including in places where there are no H's to be found... reminiscent of oxygen on Jupiter problem in the original RIDE. But that was sorted, wasn't it?
12keV more penetrating than 100keV - why did I go back to this program? I'll see the Hydrocarbon effect, then return to Saturn. Right, think I know what's going on - everything that hits the bottom must be bouncing back. This explains the deeply penetrating tail of the 30keV shot, why everything stops at a certain alt and why hydrocarbons have so little (ie no) effect... will deal with when the time comes... anyway. Saturn... after just... no that didn't work, H problem remains. But only above certain energies. This is the eternal problem of PDFs, whether the next probability up has any value if it is zero as well as the current probability. "Less than" on each probability sorts it out until the last probability, which loses one bin of bandwidth if less than is put on the decision making thing. How to be fair to all sides? Sorted it, did less than on all bins and cleared out H totally. Really should've kept some, but the atmosphere is discontinuous in the region where H is expected. Why don't the observationalists do a more thorough job with our blue friend up there? Perhaps an estimate of the missing proton column density could be helpful in some way... ie an output to all this? We shall see. Subject closed for now.
Saturn
Now worried about saturn. Testing for H problem
shitshitshitshitshit
resolving H problem...
who wrote this model?
H problem resolved
Testing for penetration problem
shitshitshitshitshit
I would expect a 100keV electron to be a little more penetrating than that! I would expect a 50keV electron to at least make the effort... Can't get away from it, these new cross-sections need to be in up and running. What on Earth possessed Achilleos to leave so much out of the JIM listings? Steve thinks there was a quality control exercise going on - which cross-sections are important, which aren't. Getting back to reality, probably true that the search for cross-sections was harder in the mid to late 1990's than in 2006... Especially since JIM had other bits to concentrate on. Did no-one think to check afterwards? I guess they did given that other JIM paper, but not hard enough, apparently...
The hydrocarbon cross-sections exist as a single spreadsheet. Alledgedly, so do the JIM ones, but since I've added and altered them, they now exist over several. Need to go back, put them together, consolidate in a single sheet all the RIDE cross-sections - yes again... must be the third time, and I still won't give it a name I remember.
emails keep coming from alan. Must be pleasant lying on a Turkish beach, just occasionally pricking a student's conscience with another part of their project whenever you feel like it... pastoral care...
uranushyrdoc2 set off, to see if I can improve the profile with a big run...
...and that's it for today
0 Comments:
Post a Comment
<< Home