using turbidity in gensky

Hello Radiance users,

I am trying to generate sky distribution during sunny day for my model to
compare the results with the measured data of illuminance. It seems that the
day when the measurements were obtained was very clear and I would like to
decrease the turbidity in gensky to get higher global and vertical
illuminance.
When I decreased the turbidity from the default of 2.45 to 1 then the
illuminance decrease, while when I increase -t option then the illuminance
increase too. I would actually expect opposite, and that the illuminance
will increase with lower turbidity since the turbidity should scattered
light more and thus block the direct contribution and increase the diffuse
contribution form the sky.

Here is the gensky command I am using:
gensky 9 5 $(printf "%02d:%02d" $h $m)CEST -a 55.964875 -o -12.4934 -m -15
+s -t 1

Does someone have had this problem before? Or is this as gensky should
works?

Thank you
David

Hi David,

It may seem counterintuitive, but the way gensky works is that the solar intensity is unaffected by turbidity. Decreasing turbidity actually reduces the sky zenith brightness using a simple model, which makes sense since no atmosphere would look like a night sky even in the daytime. Thus, since we don't properly account for the loss in solar intensity as turbidity is reduced, the total illuminance goes up.

Generally speaking, you shouldn't bother with the turbidity setting at all. Since it only affects the default sky brightness and not the sky distribution, it makes more sense to control these values directly using any of the -b, -B, -r, or -R options for setting radiance or irradiance.

Best,
-Greg

···

From: David Appelfeld <[email protected]>
Date: May 5, 2011 4:56:01 PM PDT

Hello Radiance users,

I am trying to generate sky distribution during sunny day for my model to compare the results with the measured data of illuminance. It seems that the day when the measurements were obtained was very clear and I would like to decrease the turbidity in gensky to get higher global and vertical illuminance.
When I decreased the turbidity from the default of 2.45 to 1 then the illuminance decrease, while when I increase -t option then the illuminance increase too. I would actually expect opposite, and that the illuminance will increase with lower turbidity since the turbidity should scattered light more and thus block the direct contribution and increase the diffuse contribution form the sky.

Here is the gensky command I am using:
gensky 9 5 $(printf "%02d:%02d" $h $m)CEST -a 55.964875 -o -12.4934 -m -15 +s -t 1

Does someone have had this problem before? Or is this as gensky should works?

Thank you
David

David

If you want to match measured values you should use the -R or -B options.
See the man page for details.

Regards,
Thomas

···

On Thu, May 5, 2011 at 6:56 PM, David Appelfeld <[email protected]>wrote:

Hello Radiance users,

I am trying to generate sky distribution during sunny day for my model to
compare the results with the measured data of illuminance. It seems that the
day when the measurements were obtained was very clear and I would like to
decrease the turbidity in gensky to get higher global and vertical
illuminance.
When I decreased the turbidity from the default of 2.45 to 1 then the
illuminance decrease, while when I increase -t option then the illuminance
increase too. I would actually expect opposite, and that the illuminance
will increase with lower turbidity since the turbidity should scattered
light more and thus block the direct contribution and increase the diffuse
contribution form the sky.

Here is the gensky command I am using:
gensky 9 5 $(printf "%02d:%02d" $h $m)CEST -a 55.964875 -o -12.4934 -m -15
+s -t 1

Does someone have had this problem before? Or is this as gensky should
works?

Thank you
David

_______________________________________________
Radiance-general mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/radiance-general

Hi Greg and Thomas,

thank you for the answer. I was not sure if I can use -R option, because I
have available only global horizontal irradiation and illuminance and it
seems from gensky manual that I need to have direct irradiance measured
horizontally without the diffuse contribution.
In the case I can use -R option, will the gensky looks simple like this
gensky 9 5 12:30CEST -a 55.964875 -o -12.4934 -m -15 +s -R 600

thank you
David

···

On 5 May 2011 17:02, Greg Ward <[email protected]> wrote:

Hi David,

It may seem counterintuitive, but the way gensky works is that the solar
intensity is unaffected by turbidity. Decreasing turbidity actually reduces
the sky zenith brightness using a simple model, which makes sense since no
atmosphere would look like a night sky even in the daytime. Thus, since we
don't properly account for the loss in solar intensity as turbidity is
reduced, the total illuminance goes up.

Generally speaking, you shouldn't bother with the turbidity setting at all.
Since it only affects the default sky brightness and not the sky
distribution, it makes more sense to control these values directly using any
of the -b, -B, -r, or -R options for setting radiance or irradiance.

Best,
-Greg

> From: David Appelfeld <[email protected]>
> Date: May 5, 2011 4:56:01 PM PDT
>
> Hello Radiance users,
>
> I am trying to generate sky distribution during sunny day for my model to
compare the results with the measured data of illuminance. It seems that the
day when the measurements were obtained was very clear and I would like to
decrease the turbidity in gensky to get higher global and vertical
illuminance.
> When I decreased the turbidity from the default of 2.45 to 1 then the
illuminance decrease, while when I increase -t option then the illuminance
increase too. I would actually expect opposite, and that the illuminance
will increase with lower turbidity since the turbidity should scattered
light more and thus block the direct contribution and increase the diffuse
contribution form the sky.
>
> Here is the gensky command I am using:
> gensky 9 5 $(printf "%02d:%02d" $h $m)CEST -a 55.964875 -o -12.4934 -m
-15 +s -t 1
>
> Does someone have had this problem before? Or is this as gensky should
works?
>
> Thank you
> David

_______________________________________________
Radiance-general mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/radiance-general

E(global-h) = E(diffuse-h) + E(direct-h)

or

E(global-h) = (-B value) + (-R value)

Note that this is different than the values used for gendaylit -W or -L which require direct *normal* and diffuse horizontal. But for gensky it seems the horizontal value is used for both diffuse and direct components, so take advantage of the additive nature of light.

I was not sure if I can use -R option, because I have available only global horizontal irradiation and illuminance and it seems from gensky manual that I need to have direct irradiance measured horizontally without the diffuse contribution.

···

From: David Appelfeld [mailto:[email protected]]
Sent: Thursday, May 05, 2011 8:25 PM

____________________________________________________________
Electronic mail messages entering and leaving Arup business
systems are scanned for acceptability of content and viruses