[GRASS-dev] [GRASS GIS] #2014: r.sun using EPSG:3031 projection gives strange results

[GRASS GIS]

#2014: r.sun using EPSG:3031 projection gives strange results
---------------------------+------------------------------------------------
 Reporter:  pierreroudier  |       Owner:  grass-dev@…              
     Type:  defect         |      Status:  new                      
 Priority:  normal         |   Milestone:  7.0.0                    
Component:  Raster         |     Version:  svn-trunk                
 Keywords:  r.sun          |    Platform:  Linux                    
      Cpu:  x86-64         |  
---------------------------+------------------------------------------------

Comment(by hamish):

 Replying to [comment:6 hamish]:
 > yeah, looking at the results for day=1 and 30 there's significantly more
 > light showing on the true-south side of Erebus at that time of year. To
 > test if that's a counter-intuitive cumulative time-spent effect or not,
 > next things to try I think are to plot the sun's position through the
 day,
 > and try it in another Antarctic stereographic projection that puts
 > grid-north and true-north in alignment and see if we get the same
 result.

 Just tried with epsg:3286 (ant. polar stereo +lon_0=165) and the results
 are the same. So I think what we're seeing is that at in mid-summer the
 low sun angle from the south at midnight is shadowing the north side of
 Erebus, while at midday the sun is high enough in the sky to get a bit of
 incident light onto the southern slope of the mountain. Net result: in
 mid-summer there seems to be more sunlight reaching the southern side of
 the mountains in Antarctica.

 Near first/last light in spring/autumn with the sun just peeking over the
 northern horizon the shadow effects are as you'd expect.

 It would be good to try with more realistic values for albedo (since snow)
 and Linke turbidity (since the beams are passing through a lot of
 atmosphere at these angles), and also to do the fine resolution sum for
 January 1st by hour or finer to confirm that the tortoise beats the hare.

 Day 045 seems to be where the crossover happens from more sun reaching the
 south to more reaching the north side.


 regards,
 Hamish

 ps- glad to see that 'g.region -n' is working in both projections, but
 d.grid is only working in epsg:3286 so had to use v.mkgrid + v.proj to get
 the lat/lon (1 deg x 20 minutes) overlay grid there.

-- 
Ticket URL: <https://trac.osgeo.org/grass/ticket/2014#comment:8>
GRASS GIS <http://grass.osgeo.org>