[GRASS-user] Re: understanding r.watershed

[Markus Metz]

Georg Kaspar wrote:
>> If these lakes have an outflow, i.e. water is leaving these lakes, the
>> results will be more realistic when you omit the depression input to
>> r.watershed and only use the (not filled) DEM.
>>
>> If you are talking about the basins output having NULL values around
>> these depression, this is because your basin threshold value was too
>> high, set it to a lower value. Check the current flow accumulation
>> output for a reasonable threshold.
>>     
>
> so, my accumulation map contains values from -144714 to 58920 with a 
> majority of cells between 0-100. what would be an appropriate threshold 
> value? 
You can try standard deviation of flow accumulation (r.univar -g).
> I already tried 5000 and it looked similar to the output I 
> received from r.terraflow, 
The basins of r.watershed look similar to the basins of r.terraflow?
> but when running r.watershed with depression 
> input I still receive those null()-areas...
>   
... in the basins output I assume. If you really want to treat lakes as 
real depressions, i.e. water is not supposed to leave these lakes, then 
there will always be NULL areas around the lakes unless you set the 
threshold to something much smaller than 100, but then the stream 
segments and basins become meaningless because there are too many...
> by the way, this is what my region looks like:
>
> GRASS 6.2.3 
Rather use 6.4.0RC3 instead of 6.2.3, you should get a couple of nice 
surprises...
> [...]
> cells:      466400
>   
Not that many cells, should take just a few seconds to get the basins 
(with 6.4.0RC3).