Re: [gdal-dev] gdaldem aspect is grid-oriented

2021-06-28 Thread Michael Sumner
True, we don't care where the needle points - we need polar north, it's entirely dependent on longitude, and in our usual case, stere centred on south pole, a simple special case. Cheers, Mike On Tue, 29 Jun 2021, 01:06 Andrew C Aitchison, wrote: > On Fri, 25 Jun 2021, Michael Sumner wrote: > >

Re: [gdal-dev] gdaldem aspect is grid-oriented

2021-06-28 Thread Andrew C Aitchison
On Fri, 25 Jun 2021, Michael Sumner wrote: Hello, we're using `gdaldem aspect` on polar grid products and we need to convert the grid-oriented directions to compass oriented ones, location-specific directions, where "compass north" is dependent on longitude. I am curious; in the terminology of

Re: [gdal-dev] gdaldem aspect is grid-oriented

2021-06-28 Thread Thomas Knudsen
The proj_factors(...) function, which a.o. returns the meridian convergence in the PJ_FACTORS typedef, may be useful in this work. Den fre. 25. jun. 2021 kl. 12.26 skrev Even Rouault < even.roua...@spatialys.com>: > Michael, > > A general solution would have more appeal, although I suspect it wil

Re: [gdal-dev] gdaldem aspect is grid-oriented

2021-06-25 Thread Even Rouault
Michael, A general solution would have more appeal, although I suspect it will be a bit more computation intensive as you'll probably have to transform to invoke PROJ multiple times per grid point to compute the direction to the North Even Le 25/06/2021 à 06:29, Michael Sumner a écrit : Hel

[gdal-dev] gdaldem aspect is grid-oriented

2021-06-24 Thread Michael Sumner
Hello, we're using `gdaldem aspect` on polar grid products and we need to convert the grid-oriented directions to compass oriented ones, location-specific directions, where "compass north" is dependent on longitude. It's easy but inefficient for us to do this after running gdal dem for our polar g