Poster at Childhood Leukaemia conference (London 2004) and related stuff
[ICO]NameLast modifiedSizeDescription
[PARENTDIR]Parent Directory  -  
[ ]post_jh.zip2004-09-07 08:29 639K 
[IMG]action_spectrum_RussFoster.jpg2004-09-13 17:49 620K 
[ ]post_jh.pdf2004-09-13 17:39 351KChildh. Leuk. poster
[TXT]raw2lum.pas2008-02-08 18:43 113KPascal source code
[IMG]pilz_vin.jpg2003-06-19 17:49 103K 
[IMG]detail.jpg2004-09-05 21:20 102K 
[IMG]action_spectr_RussFoster.jpg2004-09-13 17:50 81K 
[ ]actsp.zip2004-09-18 00:34 62Kaction spectrum curve
[IMG]act_sp_detail.jpg2004-09-13 19:51 48K 
[ ]16spL.eps2004-09-13 17:39 40K 
[ ]15spL.eps2004-09-13 17:39 39K 
[IMG]16spIlan.jpg2004-09-05 21:22 21K 
[ ]15spT.eps2004-09-06 22:00 21K 
[ ]16spT.eps2004-09-06 22:21 21K 
[ ]16spL.pdf2004-09-13 17:39 19K 
[ ]15spL.pdf2004-09-13 17:39 17K 
[ ]calib_faint.eps2004-09-06 21:00 15K 
[TXT]post_jh.tex2004-09-06 22:38 11K 
[ ]15spT.pdf2004-09-06 22:03 5.7K 
[ ]16spT.pdf2004-09-06 22:21 5.6K 
[ ]actsp.pdf2004-09-18 00:30 5.5Kaction spectrum curve
[TXT]cz_light.htm2004-08-06 23:26 3.7Kposter's abstract
[ ]calib_faint.pdf2004-09-06 22:03 3.0K 
[TXT]actsp_fit1nm.txt2004-09-18 00:47 2.6Kaction spectrum curve
[TXT]mul_sp.pas2004-09-06 21:07 569  

Poster and related images, data and formulas

The poster itself is post_jh.pdf, its abstract made earlier is cz_light.htm. A lot of auxiliary files is here too, some of them needed for the poster, some made later. I keep them here as some links may exist to them.

From a letter to David Keith, Sep 18, 2004, on action spectrum:

I've measured the Hankins&Lucas data from the slide... the precision is much better than the uncertainty of the data themselves. Thapan data are measured from their original paper, it's available somewhere as a pdf, free. Brainard's are given in that ``Cologne'' issue of Neuroendocrin. Letters.

The fit follows the lens filtering curve for the violet up to 410 nm, as if the ganglion cell pigment sensitivity were 1 all the way from to the UV. This is probably an extreme possibility.

I've used a pair of cubic parables, touching 0 at 460 nm (just quadratic and cubic components are nonzero). The analytic expression is evident from the short Pascal programme I used to get a table per each nanometer. In the zip file, there are also the source data and scripts for gnuplot, difficult to understand for another people then me.

That table is actsp_fit1nm.txt (with Unix end of lines, i.e. just LF, not CR LF).

The real action spectrum might be a bit narrower.

Or, if Thapan is true, then the maximum sensitivity is more to the left and the violet part of the curve should begin to fall very steeply down just from 420 nm. For the time being, it might be a better precautionary approach. (I have no explanation why that Thapan point differs from the other two teams.)

I think we may use the current curve as a good working one, until new data are published.