4 replies to this topic

[drogers]

Hi,

I have a map projection that shows cloud cover data over eastern Canada (Link: http://www.weatheroffice.gc.ca/astro/image...TRO_nt_003.png)
What I would like to do is determine what pixel is equivillent to a given lat/long. I have e-mailed Environment Canada to ask for the parameters but recieved the following response:

"For the lat-long, these are forecast image and not actual satellite images. We utilize a polar stereographic
projection for all the astro-forecast images. Iguess you will have to create the interface to convert lat and
long for each pixel."

So I am left to determine the parameters myself. I believe that this map projection is zoomed in on a larger map projection similar to the one in this link: http://www.weatherof...STRO_uv_003.png

Does anyone have any good ideas of how I might be able to determine what pixil is equivilent to a particular lat/long?

Thanks for any information/help you can provide.

Regards,
Dan Rogers

*Edit- The above link does not work anymore (My mistake) The map projections can be accessed from here: http://www.weatherof...clds_vis_e.html
I am interested in the "Regional model, satellite simulation, Eastern Canada window"

Edited by drogers, 11 February 2011 - 01:33 PM.

[rudy]

The first link doesn't leads to a map; the second one does. Without seeing the map ion question, it is difficult to answer your question.

[drogers]

The first link doesn't leads to a map; the second one does. Without seeing the map ion question, it is difficult to answer your question.

Sorry about the bad link. They update these on a 24 hour basis and I didn't realize the address changes. Anyway, here is a link to the main page where I access the maps: http://www.weatherof...clds_vis_e.html
I am particularly interested in the images from the "Regional model, satellite simulation, Eastern Canada window". If you click on any of the hours (i.e. "o3h") it will bring up the image.

[Melita Kennedy]

Your correspondent says it's polar stereographic. I get an image in ArcMap that looks pretty close by choosing a North pole-centered stereographic projection. Standard parallel is set to 75N, central meridian to 110W. You might ask them for the projection parameters and the sphere or ellipsoid they're using.

Oh look, I think the central meridian is 111W:

Weatheroffice Technical Grid specification

If you can't get any information on the sphere/ellipsoid or parameter values, you may be able to use the lat-lon corner values to georeference the raster. (Or project the lat-lon values into a polar stereographic and then georeference).

Melita

[drogers]

Melita,

Thank you for your help. That Weather Office link was very useful. I wish I had been able to find it on my own. I think that the "Global GEM on full-resolution 30 km Polar Stereographic grid" matches the closest to my map of eastern Canada.

Now I am attempting to create a spreadsheet that will output X/Y coordinates for a given lat long. I found some formulas on wolfram (here: http://mathworld.wolfram.com/StereographicProjection.html). Any idea what the conformal latitude means? I don't see this parameter listed. Is it the centre latitude?

Edited @ 3:22 Tuesday FEB 15th 2011
-I have created a spreadsheet with the required caclulations and get an output of X and Y
-I noticed the X & Y coordinates are realted to the value I use for my radius.
-How do I realte these XY coordinates with the map projection? The parameters state:

ni 247
nj 200
resolution at 60° N 30 km
coordinate of first grid point 32.549° N
134.614° W
(i,j) coordinate of North Pole (88.0, 200.0)
grid orientation
(with respect to j axis) -111.0°