[GRASS-dev] r.slope.aspect with -n flag giving aspect of flat areas as -9998

Markus Metz markus.metz.giswork at gmail.com
Thu Jan 30 00:26:47 PST 2020


On Tue, Jan 28, 2020 at 12:50 PM Pedro VenĂ¢ncio <pedrongvenancio at gmail.com>
wrote:
>
> Hi all,
>
> I was testing r.slope.aspect with the "new" -n flag, also to include -e
and -n flags in QGIS Processing, but -n flag is giving aspect values of
flat areas as -9998, instead of -9999, when the precision choosed is CELL,
and the output data type is CELL.

that seems to be a rounding error, even though float or double should be
able to represent -9999 exactly. Should be easy to fix.

Markus M
>
> https://grass.osgeo.org/grass78/manuals/r.slope.aspect.html
>
> I've tested with GRASS 7.6 (on Linux) and GRASS 7.8.2 (on Windows), both
in the GUI and CLI.
>
> (Tue Jan 28 11:19:40 2020)

> r.slope.aspect -e -n --verbose elevation=SRTM_PT_25m at PERMANENT
slope=SRTM_PT_25m_Slope_Deg_GRASS aspect=SRTM_PT_25m_Aspect_Std_GRASS
precision=CELL
> Percent complete...
> Elevation products for mapset <PERMANENT> in <SRTM_PT_25m>
> Min computed aspect 0.0000, max computed aspect 360.0000
> Aspect raster map <SRTM_PT_25m_Aspect_Std_GRASS> complete
> Min computed slope 0.0000, max computed slope 75.2970
> Slope raster map <SRTM_PT_25m_Slope_Deg_GRASS> complete
> (Tue Jan 28 11:20:24 2020) Comando terminado (43 sec)
>
> r.info map=SRTM_PT_25m_Aspect_Std_GRASS at PERMANENT

>
 +----------------------------------------------------------------------------+
>  | Map:      SRTM_PT_25m_Aspect_Std_GRASS@  Date: Tue Jan 28 11:20:24
2020    |
>  | Mapset:   PERMANENT                      Login of Creator:
PedroVenancio   |
>  | Location: SRTM_PT_25m
     |
>  | DataBase: D:\ICNF\Modelos_Combustivel_2018\Landscape_File\grass78
     |
>  | Title:    Aspect counterclockwise in degrees from east
    |
>  | Timestamp: none
     |
>
 |----------------------------------------------------------------------------|
>  |
     |
>  |   Type of Map:  raster               Number of Categories: 360
    |
>  |   Data Type:    CELL
    |
>  |   Rows:         23200
     |
>  |   Columns:      11360
     |
>  |   Total Cells:  263552000
     |
>  |        Projection: ETRS89 / Portugal TM06
     |
>  |            N:     278000    S:    -302000   Res:    25
    |
>  |            E:     164000    W:    -120000   Res:    25
    |
>  |   Range of data:    min = -9998  max = 360
    |
>  |
     |
>  |   Data Source:
    |
>  |    raster elevation file SRTM_PT_25m at PERMANENT
    |
>  |
     |
>  |
     |
>  |   Data Description:
     |
>  |    generated by r.slope.aspect
    |
>  |
     |
>  |   Comments:
     |
>  |    aspect map elev = SRTM_PT_25m at PERMANENT
    |
>  |    zfactor = 1.00
     |
>  |    min_slope = 0.000000
     |
>  |
     |
>  |    r.slope.aspect --verbose -e -n elevation="SRTM_PT_25m at PERMANENT"
slo\   |
>  |    pe="SRTM_PT_25m_Slope_Deg_GRASS"
aspect="SRTM_PT_25m_Aspect_Std_GRAS\   |
>  |    S" format="degrees" precision="CELL" zscale=1.0 min_slope=0.0
    |
>  |
     |
>
 +----------------------------------------------------------------------------+
> (Tue Jan 28 11:26:09 2020) Comando terminado (0 sec)

>
>
> Using the default precision as FCELL, all goes ok:
>
> (Tue Jan 28 11:33:08 2020)

> r.slope.aspect -e -n --verbose elevation=SRTM_PT_25m at PERMANENT
slope=SRTM_PT_25m_Slope_Deg_GRASS_2 aspect=SRTM_PT_25m_Aspect_Std_GRASS_2
> Percent complete...
> Elevation products for mapset <PERMANENT> in <SRTM_PT_25m>
> Min computed aspect 0.0000, max computed aspect 359.8096
> Aspect raster map <SRTM_PT_25m_Aspect_Std_GRASS_2> complete
> Min computed slope 0.0000, max computed slope 75.2970
> Slope raster map <SRTM_PT_25m_Slope_Deg_GRASS_2> complete
> (Tue Jan 28 11:34:01 2020) Comando terminado (53 sec)
>
> (Tue Jan 28 11:34:27 2020)

> r.info map=SRTM_PT_25m_Aspect_Std_GRASS_2 at PERMANENT

>
 +----------------------------------------------------------------------------+
>  | Map:      SRTM_PT_25m_Aspect_Std_GRASS_  Date: Tue Jan 28 11:34:01
2020    |
>  | Mapset:   PERMANENT                      Login of Creator:
PedroVenancio   |
>  | Location: SRTM_PT_25m
     |
>  | DataBase: D:\ICNF\Modelos_Combustivel_2018\Landscape_File\grass78
     |
>  | Title:    Aspect counterclockwise in degrees from east
    |
>  | Timestamp: none
     |
>
 |----------------------------------------------------------------------------|
>  |
     |
>  |   Type of Map:  raster               Number of Categories: 360
    |
>  |   Data Type:    FCELL
     |
>  |   Rows:         23200
     |
>  |   Columns:      11360
     |
>  |   Total Cells:  263552000
     |
>  |        Projection: ETRS89 / Portugal TM06
     |
>  |            N:     278000    S:    -302000   Res:    25
    |
>  |            E:     164000    W:    -120000   Res:    25
    |
>  |   Range of data:    min = -9999  max = 359.8097
     |
>  |
     |
>  |   Data Source:
    |
>  |    raster elevation file SRTM_PT_25m at PERMANENT
    |
>  |
     |
>  |
     |
>  |   Data Description:
     |
>  |    generated by r.slope.aspect
    |
>  |
     |
>  |   Comments:
     |
>  |    aspect map elev = SRTM_PT_25m at PERMANENT
    |
>  |    zfactor = 1.00
     |
>  |    min_slope = 0.000000
     |
>  |
     |
>  |    r.slope.aspect --verbose -e -n elevation="SRTM_PT_25m at PERMANENT"
slo\   |
>  |    pe="SRTM_PT_25m_Slope_Deg_GRASS_2"
aspect="SRTM_PT_25m_Aspect_Std_GR\   |
>  |    ASS_2" format="degrees" precision="FCELL" zscale=1.0 min_slope=0.0
     |
>  |
     |
>
 +----------------------------------------------------------------------------+
> (Tue Jan 28 11:34:28 2020) Comando terminado (0 sec)
>
>
> The elevation map I'm using is this one:
> https://www.fc.up.pt/pessoas/jagoncal/srtm/SRTM_PT_25m.tif
>
> As CELL is a 32 bit integer with a range from -2,147,483,647 to
+2,147,483,647, -9999 should be fine to conform inside it.
>
> Another issue I see is in r.info, that is showing the title as
>
> | Title:    Aspect counterclockwise in degrees from east
>
> which is not correct in this case, as the -n flag was used.
>
> Could these be bugs?
>
> Thank you very much.
>
> Best regards,
> Pedro VenĂ¢ncio
> _______________________________________________
> grass-dev mailing list
> grass-dev at lists.osgeo.org
> https://lists.osgeo.org/mailman/listinfo/grass-dev
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/grass-dev/attachments/20200130/d5ce7ed2/attachment.html>


More information about the grass-dev mailing list