Further errata for alg2001

[Jan Hollan <jhollan@amper....muni.cz>]

> Lunar illumination is an order of magnitude lower, 
> not 1 to 10 lux, 
> but up to perhaps 1 lux (the upper limit requires snow and suitable
> clouds with a hole for the full Moon itself, 
> without them, 0.3 lux is an exceptional maximum).

some more errata (is there a more suitable address for sending them?),

a simple consequence of the previously mentioned one, but maybe a more
serious one. Under moonlight, color perception begins, i.e., mesopic
perception really exists. Definitely at illumination levels of 0.1 lux
already. (I did not remember any literature on that, but I am often
checking that I can really discriminate colors a bit under moonlight.)
Consequently,

 the figure 2.7 on the
 p. 2-8, ref. 2.1.7,

should shift the mesopic/scotopic threshold at least an order of
magnitude left or even further, to perhaps 0.1 lux. A truly scotopic
vision would then disappear from the plotted range...

Further on this page, a sentence on scotopic vision should read
,,substantially less than 0.1 footcandle``. 


I am much afraid, that this order-of-magnitude misconcept of what is the
true level of natural nighttime illuminations, may lead to
``catastrophic'' wasteful tenfold illuminances for outdoor lighting
practices. Some codes (at least that which holds in Czechia) still contain
the lowest average illuminance limit 0.1 lux for footpaths etc.,
implicated long ago by the evident fact that moonlight is often enough,
but just the next category means a twentyfold leap to much higher
illumination level. Subsequent categories follow then smoothly by powers
of two. How old may be that blunder regarding moonlight? It's a pity that
lighting experts are so innocent as regards astronomical photometry (no
wonder, its language being so fuzzy and controversial, see my article
 http://astro.sci.muni.cz/pub/hollan/a_papers/english/ ).

regards,

jenik hollan