The confusion about the definition of view azimuth angle.

[JZH]

From Figure 67 in DART_User_Manual.pdf, the azimuth angle is the anti clockwise angle relative to the horizontal axis beginning from the east. However, I found it is wrong from Figure 124 because the view anzimuth angle is the clockwise angle relative to the horizontal axis beginning from the east.

[Jean-Philippe Gastellu-Etchegorry]

In DART, in the 2D graph of reflectance, radiance and temperature, the origin of the azimuth angle is the horizontal Ox axis, anticlockwise.

In figure 67 (sorry, it is no more 67 in my DART User Manual), it is stated: "Anti clockwise angle relative to the horizontal axis = azimuth angle Phi."

I do not understand the problem of figure 124. Figure 124.b displays the sun direction with Phi=135°, which is consistent with the definition of the azimuth angle.

Possible source of confusion: in the displays of DART mock-ups, the Ox axis is "vertical downward"...

[JZH]

 

In Table 51, it is said that Sun azimuth angle is 225°, not 135°.

In Figure 124, as you said that the sun azimuth angle is 135°, and the view azimuth angle is 30°, then I should see one sunlit wall and one shadow wall, not two shadow walls as shown in Figure 124.

[Jean-Philippe Gastellu-Etchegorry]

The simulation is conducted with an azimuth angle that is equal to 225°. Do not forget that the origin of the azimuth angle is:

- the horizontal Ox asis in 2D graphs. In figure 124 (Reflectance r(q,f). a) 1D: sun (f = 225°) and perpendicular (f = 135°) planes. b,c) 2D & 3D graphs)": the directions of the hot spot configuration appear at 225°, in figure 124.b. In figure 124.b, the graph shows 2 curves: the reflectanc in the sun plane (azimuth = 225°) and in the perpendicular palne (azimuth = 135°)

- the vertical downward axis in DART images and mok-up. This is verified when looking at the nadir image: the shadows of the houses appear along a direction "North West - South -East".

Note that the direction along which you see the sun (upward direction) differs from the sun direction (downward direction): in order to switch between these 2 directions (labelled 1 and 2), the mathematical transformation is:
       Zenith_1 = Pi - Zenith_2     and   Azimuth_1 = Azimuth_2 + Pi  (modulo 2.Pi)

[JZH]

I think we refer to different versions of DART_User_Manual. In my pdf file, the figure 124 is like this:

[Jean-Philippe Gastellu-Etchegorry]

Exact. In the most recent DART user manual, this figure is Number 129. Sorry for the confusion. The DART user manual changes quite often, with the improvements of DART, the exercises that are improved, etc.  and also remarks of users like you. Thank you. 

In Figure 129 (124, for you), shadows appear correctly according to the angles that are specified. Do you agree?

[JZH]

I am sorry that I don't totally understand your explanations. The following figure shows what I want to say.

[Tiangang Yin]

Dear JZH,

Apparently there is a difference in the definition of angles.

Please see the figure enclosed.

Regards,

Tiangang

[JZH]

Thank you. So the following figure gives the wrong definition of the azimuth angle.

[Tiangang Yin]

Actually not, they are both correct.

We use different coordinate systems to define images and 2-D BRF.

It might sound confusing, but under certain circumstances, they work quite conveniently. E.g, when you read animage in matlab, the row index 0 is always at the top of the image instead of the bottom of the image.

Regards,

[Jean-Philippe Gastellu-Etchegorry]

Hi,

I join the figure (with addition of axes and sun direction), in addition to the explanation that is provided by Tiangang. The present DART User Manual has these axes in some figures of DART exercises: thank you for your remarks, JZH.

Cheers

Jean Philippe

[JZH]

From my point of view, in one system, like DART, one definition is enough. In my field, the azimuth angle is in 0 -360 in a clockwise direction beginning from the north(Upward direction). Now, is the azimuth from the south (downward direction) in a anti-clockwise direction used for the image, not the 2D BRF?

Thank you.

Zhong-Hu

[JZH]

You are welcome. I learn more from you. Thank you too.

[Jean-Philippe Gastellu-Etchegorry]

Yes, in the DART mock-up and in the DART images, the Ox axis is vertical downward. This Ox axis is the origin of the azimuth angle for the view direction and for the direction that looks up to the sun.

In the 2D BRF graph, the Ox axis is horizontal (West to East). It is the origin of the azimuth angle.

In DART, azimuth angles use the trigonometric system (anticlockwise)

[JZH]

Thank you very much. I understand now.