Re: Asphalt and Concrete surfaces

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

> D.K. wrote :
>
> >But still some
> >(http://eande.lbl.gov/HeatIsland/PUBS/2000/43442rep.pdf gives road albedos
> >for solar radiation, visual ones may be a bit lower)...
> >Asphalt roads are not black at all as a rule ...
> >as the real (in vast majority, more
> >than 3 years old) asphalt streets and roads are far brighter than those 7
> >or 10 per cent. The typical albedo is perhaps 12 per cent, 15 being quite
> >common as well, or even more. Specular reflectivity is high in general
> >and the more in the very polished paths of tyres.
>
> I am very interested in the source of this data - please let me know where
> this information can be referenced!  Of course the "tyre" tracks are only a
> small portion of the roadway - all of which gets weathered - which for
> asphalt that tends to *reduce* its reflectance - so this data is surprising
> to say the least!

The albedos (total=diffuse+specular reflectances) are from that HeatIsland
study. There is a graph showing albedo as a function of the years.

The claim the specular component rises as the surface is used by cars is
my guess, from observing streets in Brno. I know no research on it. It may
not hold for roads which are seldom used, where no physical polishing is
in action.

The tyre paths are bright against the low-lying sun and dark when viewing
them with the sun in your back, compared with less used parts of the road
surface -- so the specular component is much higher for tyre paths.
Non-technically, they are much more polished then the rest. They may
comprise half of the road width.

Their albedo compared to the less polished parts of the road can be
observed when the sky is overcast suitably so that it has a completely
uniform luminance (thick fog may do the same) (CIE skies III.1). The
albedo is different form the rest, if they are observable as albedo
features at such circumstances. They could be darker if black rubber
traces are abundant on them. I did not do such observations yet.

> A paper to be presented at this year's IESNA Conference by Knox and Keith
> (myself) discusses the photopic hemispherical reflectances of these asphalt
> and concrete surfaces under different sources and finds that the asphalt
> ranges from 7.3% to 7.6% and concrete from 18.1% to 33.7%.

How old are these asphalt surfaces, how many millions of cars passed them,
how do the properties differ transversally to the road? How it is possible
that the range of albedos (7.3 % to 7.6 %) is so narrow? It could
concern mean value of fresh asphalt roads but the main material are the
mineral grains (the road is asphalt concrete, asphalt being some 15 %,
``concrete'' roads differ just by using cement instead of asphalt as a
glue).  After the first thousands of cars pass the road, the mineral
grains start to become visible and the albedo is much dependent on them,
even before substantial bleaching by sun takes place. (Not to speak about
technology, where mineral grains are spread onto a fresh sticky asphalt
surface, so that it's blackness is hindered from the very beginning.)

> This data does
> support what I have been saying for years - that using LUMINANCE can lead to
> substantial reductions in lighting energy and reduce ambient light
> pollution - skyglow - much more than any mandate for FC distributions.

Reductions in needed luminous fluxes, yes. (For less educated in
photometry, luminance means how bright the scene appears. Illuminance
means how much light falls on it. For a given illuminance, you get more
luminance if the scene is rather white than black.)

But for reduction of light pollution, including skyglow, using FS (fully
shielded) luminaires (no light above horizon, simply speaking) helps much
more.

Skyglow is caused mostly by the almost horizontal light going upwards, and
vast majority of such light is produced by the non-FS luminaires nowadays.
Another types of light pollution IS simply their direct visibility. FS
ones are not visible from above, hence they don't pollute in this respect
(they do no harm to birds, people and insects). And, in the average, they
are much fainter then the most common non-FS luminaires even when viewed
slightly from below (they have to be, having zero cd /klm horizontally).
Again, attraction of insects, illumination of distant windows and glare is
reduced a lot, in most cases.

Of course, demanding an utilance far over fifty per cent (say, whole 0.9
of the emitted light going just down to the road) additionally to a simple
FS rule would be possible and should happen in the future, when at least
the basic rule (NO light upwards) will be obeyed.

> Using visibility metrics and complete directional reflectance data should
> lead to further reductions - as stated in my 1999 paper which looked

Yes, and there is another approach available, namely evaluating the real
scenes visually, and now by digital imaging photometry as well, using
common digital cameras offering raw data from their A/D converters. This
is what I plan to do (I made some minor steps using Canon EOS D60, there
is no obstacle to get scientific data of the similar quality as with
16-bit cameras with filter wheels). The David Coffin's work enables this:
             Raw Digital Photo Decoding in Linux
        URL: http://www2.primushost.com/~dcoffin/powershot/
 (the programmes can be compiled under M$windows as well).

jenik