[GRASS-dev] r.sim.water buffer overflow

[Hanlie Pretorius]

2010/8/19, Hanlie Pretorius <hanlie.pretorius at gmail.com>:
>
> I will get the Spearfish dataset into a LatLong location and see what
> happens when I run r.sim.water on that. I could place the Spearfish
> latlong location on a temporary web site if needed.
>

I couldn't convert the spearfish60 location into a WGS84 location:
-----
GRASS 6.4.0RC6 (spearfish_wgs84):~ > r.proj input=elevation.10m
location=spearfish60 mapset=PERMANENT output=elevation.10m
method=cubic --overwrite
Input Projection Parameters:  +proj=utm +zone=13 +a=6378206.4
+rf=294.9786982 +no_defs +nadgrids=/usr/lib/grass64/etc/nad/conus
Input Unit Factor: 1
Output Projection Parameters:  +proj=longlat +no_defs +a=6378137
+rf=298.257223563 +towgs84=0.000,0.000,0.000
Output Unit Factor: 1
WARNING: pj_transform() failed: latitude or longitude exceeded limits
WARNING: pj_transform() failed: latitude or longitude exceeded limits
WARNING: pj_transform() failed: latitude or longitude exceeded limits
.
. (lots of these...)
.
WARNING: pj_transform() failed: latitude or longitude exceeded limits
WARNING: pj_transform() failed: latitude or longitude exceeded limits
WARNING: pj_transform() failed: latitude or longitude exceeded limits

Input:
Cols: 1899 (1899)
Rows: 1398 (1398)
North: 4928000.000000 (4928000.000000)
South: 4914020.000000 (4914020.000000)
West: 590010.000000 (590010.000000)
East: 609000.000000 (609000.000000)
EW-res: 10.000000
NS-res: 10.000000

Output:
Cols: 1 (360)
Rows: 1 (180)
North: 45.000000 (90.000000)
South: 44.000000 (-90.000000)
West: -104.000000 (-180.000000)
East: -103.000000 (180.000000)
EW-res: 1.000000
NS-res: 1.000000

Allocating memory and reading input map...
 100%
Projecting...
 100%
r.proj complete.
-----

r.info for the result:
-----
 |   Type of Map:  raster               Number of Categories: 255
 |   Data Type:    FCELL
 |   Rows:         1
 |   Columns:      1
 |   Total Cells:  1
 |        Projection: Latitude-Longitude
 |            N:        45N    S:        44N   Res:     1
 |            E:       103W    W:       104W   Res:     1
 |   Range of data:    min = -nan  max = -nan
-----

g.region was set to:
-----
g.region -p n=90 s=-90 e=180 w=-180
projection: 3 (Latitude-Longitude)
zone:       0
datum:      wgs84
ellipsoid:  wgs84
north:      90N
south:      90S
west:       180W
east:       180E
nsres:      1
ewres:      1
rows:       180
cols:       360
cells:      64800
-----

However, I do have my dataset in a Transverse Mercator projected
coordinate system, so I ran r.sim.water using that data (66MB, posted
on http://www.nedbib.za.net/rsimwater/ if you want to download the
location) and I didn't get the buffer overflow, but I now get strange
results.

At a resolution of 100m (822120 cells), I get an output map that looks
like something ('lieb_depth_100m_res' raster in my downloadable
location). However, at a resolution of 25m - the original resolution
of the DEM - (13153920 cells) I get 0 depth over the whole region
('lieb_depth_25m_res' raster in my downloadable location)

To set the region, I use a command such as:
g.region -p vect=lieb_border res=25

Also, r.sim.water runs implausibly fast and I don't get all the output
that I used to get, like the time it took for the command to complete.
For example:
-----
GRASS 6.4.0RC6 (sa_tm_29deg_E_extract):~ > r.sim.water
elevin=lieb_dem_25m dxin=lieb_dem_25m_dx dyin=lieb_dem_25m_dy
depth=lieb_depth_25m_res --overwrite
default nwalk=26307840, rwalk=26307840.000000

Min elevation = 1498.27 m
Max elevation = 2445.12 m
Mean Source Rate (rainf. excess or sediment) = 0.000006 m/s or kg/m2s
Mean flow velocity = 1.166379 m/s
Mean Mannings = 0.170640
Number of iterations = 27 cells
Time step = 5.36 s
GRASS 6.4.0RC6 (sa_tm_29deg_E_extract):~ >
-----

So, it seems the LatLong coordinate system was part of the problem,
but I'm not out of the woods yet. Should I still try to compile GRASS
myself with the debugging symbols?

Regards
Hanlie