r.stream.extract: lat-long vs projected location

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

r.stream.extract: lat-long vs projected location

Moritz Lennert
Hi,

Does anyone have any experience, or theoretical knowledge based on the
code, on whether the results of r.stream.extract (and also r.watershed)
will be influenced by working in a lat-long location instead of a
projected location ?

Moritz
_______________________________________________
grass-user mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|

Re: r.stream.extract: lat-long vs projected location

Markus Metz-3


On Fri, May 3, 2019 at 4:17 PM Moritz Lennert <[hidden email]> wrote:
>
> Hi,
>
> Does anyone have any experience, or theoretical knowledge based on the
> code, on whether the results of r.stream.extract (and also r.watershed)
> will be influenced by working in a lat-long location instead of a
> projected location ?

The A* search in both r.stream.extract and r.watershed distinguishes only between direct and diagonal neighbor cells. This determines flow D8 direction. MFD flow direcition also
distinguishes only between direct and diagonal neighbor cells. Flow accumulation is reported as number of cells, i.e. independent of he CRS. In r.watershed. slope length and slope steepness use map units for horizontal length and units of the elevation map for vertical units, which will cause nonsense output for slope steepness. TCI and SPI use distance in meters for contour length in ll, but they use the average nsres and ewres for the current region in ll which is not accurate for areas away from the equator.

In both modules, threshold refers to the unit of flow accumulation (usually number of cells). r.stream.extract accepts any flow accumulation of any units as input, the threshold option must match the units of flow accumulation. Montgomery adjustment distinguishes only between direct and diagonal neighbors.

HTH,

Markus M

_______________________________________________
grass-user mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/grass-user
Reply | Threaded
Open this post in threaded view
|

Re: r.stream.extract: lat-long vs projected location

Moritz Lennert
On 3/05/19 22:07, Markus Metz wrote:

>
>
> On Fri, May 3, 2019 at 4:17 PM Moritz Lennert
> <[hidden email] <mailto:[hidden email]>> wrote:
>  >
>  > Hi,
>  >
>  > Does anyone have any experience, or theoretical knowledge based on the
>  > code, on whether the results of r.stream.extract (and also r.watershed)
>  > will be influenced by working in a lat-long location instead of a
>  > projected location ?
>
> The A* search in both r.stream.extract and r.watershed distinguishes
> only between direct and diagonal neighbor cells. This determines flow D8
> direction. MFD flow direcition also
> distinguishes only between direct and diagonal neighbor cells. Flow
> accumulation is reported as number of cells, i.e. independent of he CRS.
> In r.watershed. slope length and slope steepness use map units for
> horizontal length and units of the elevation map for vertical units,
> which will cause nonsense output for slope steepness. TCI and SPI use
> distance in meters for contour length in ll, but they use the average
> nsres and ewres for the current region in ll which is not accurate for
> areas away from the equator.
>
> In both modules, threshold refers to the unit of flow accumulation
> (usually number of cells). r.stream.extract accepts any flow
> accumulation of any units as input, the threshold option must match the
> units of flow accumulation. Montgomery adjustment distinguishes only
> between direct and diagonal neighbors.

Thanks for the clarification !

Moritz

_______________________________________________
grass-user mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/grass-user