[GRASS-user] r. walk and sites which can be reached in a certain time

[Anna Petrášová]

Please, keep the conversation on the mailing list.

On Mon, Oct 24, 2016 at 12:31 PM, Phil Burgdorf <P_Burgdorf at web.de> wrote:
> I use Grass together with QGIS.
> So i donÄt know how to give you the command. I'll send you the r. info for
> may input data and wrote down what I#ve done, I hope that'll give yo the
> information you need.
>
> First i calculated the speed km/h with this formula (6e ^-3,5Islope + 0,05)
> (layername:Test_Neo_Geschw)
> r. info for that:
>
> Map: tmp1477325189748 Date: Mon Oct 24 18:06:30 2016 |
> | Mapset: PERMANENT Login of Creator: Phil |
> | Location: temp_location |
> | DataBase: C:\Users\Phil\AppData\Local\Temp\processing99ff3938206843eb8f58e
> |
> | Title: ( tmp1477325189748 ) |
> | Timestamp: none |
> |----------------------------------------------------------------------------|
> | |
> | Type of Map: raster Number of Categories: 0 |
> | Data Type: FCELL |
> | Rows: 7832 |
> | Columns: 9000 |
> | Total Cells: 70488000
> | Projection: unnamed
> | N: 5598800 S: 5403000 Res: 25
> | E: 4585500 W: 4360500 Res: 25
> | Range of data: min = 0.05126764 max = 0.1423827
>
> After that the cost with c = 60 /v (layer name: Test_Neo_Kosten)
>
> r. info for that:
> Map: tmp14773254289710 Date: Mon Oct 24 18:10:29 2016 |
> | Mapset: PERMANENT Login of Creator: Phil |
> | Location: temp_location |
> | DataBase: C:\Users\Phil\AppData\Local\Temp\processing99ff3938206843eb8f58e
> |
> | Title: ( tmp14773254289710 ) |
> | Timestamp: none |
> |----------------------------------------------------------------------------|
> | |
> | Type of Map: raster Number of Categories: 0 |
> | Data Type: FCELL |
> | Rows: 7832 |
> | Columns: 9000 |
> | Total Cells: 70488000
> | Projection: unnamed
> | N: 5598800 S: 5403000 Res: 25
> | E: 4585500 W: 4360500 Res: 25
> | Range of data: min = 421.3996 max = 1170.329
>
> In r. walk I used the slope of my test region (layername: Test_Neo_Slope) as
> elevation raster and the layer " Test_Neo_Kosten" as input raster for the

why do you use slope as elevation raster? It should be elevation,
slope is computed internally. Also, I don't really understand how you
define the cost. You should define friction as additional time to
cross a cell depending on landcover for example. I suggest looking at
r.walk manual, also look at:
http://www.bioone.org/doi/full/10.1659/MRD-JOURNAL-D-13-00041.1
where r.walk is used.


> friction cost. my starting ponit is a simple shape-point in this region. The
> first result is fine. The algorithm creates a map in my test region. I
> have'nt changed all the other values. As i told you the algorithm fails when
> I change the Max. cost.
>
>
> r.info for r.walk:
> | Map: tmp14773265793514 Date: Mon Oct 24 18:29:43 2016 |
> | Mapset: PERMANENT Login of Creator: Phil |
> | Location: temp_location |
> | DataBase: C:\Users\Phil\AppData\Local\Temp\processing99ff3938206843eb8f58e
> |
> | Title: ( tmp14773265793514 ) |
> | Timestamp: none |
> |----------------------------------------------------------------------------|
> | |
> | Type of Map: raster Number of Categories: 0 |
> | Data Type: DCELL |
> | Rows: 7832 |
> | Columns: 9000 |
> | Total Cells: 70488000 |
> | Projection: unnamed |
> | N: 5598800 S: 5403000 Res: 25 |
> | E: 4585500 W: 4360500 Res: 25 |
> | Range of data: min = 0 max = 775143.701578344 |
>
> I hope these informations are more helpfull.
>
> Gesendet: Montag, 24. Oktober 2016 um 17:46 Uhr
> Von: "Anna Petrášová" <kratochanna at gmail.com>
> An: "Phil Burgdorf" <P_Burgdorf at web.de>
> Betreff: Re: Re: Re: [GRASS-user] r. walk and sites which can be reached in
> a certain time
> On Mon, Oct 24, 2016 at 11:41 AM, Phil Burgdorf <P_Burgdorf at web.de> wrote:
>> Thank you for the formula. I replaced "time to walk" by my layer named
>> "acumulated_cost".
>> So it's now: "accumulated_cost_limited = if(acumulated_cost < 3600,
>> acumulated_cost,null())" Is that right?
>> After running the caclculator I didn't get a new raster map althogh it
>> says
>> "algorithm finished"
>
>
> You have to provide more information. What does r.info acumulated_cost
> show? What's the r.walk command you used and what are the input data?
> Did you set the computational region to match your input data (see
> g.region)?
>
>>
>> Gesendet: Montag, 24. Oktober 2016 um 16:34 Uhr
>> Von: "Anna Petrášová" <kratochanna at gmail.com>
>> An: "Phil Burgdorf" <P_Burgdorf at web.de>, "GRASS user list"
>> <grass-user at lists.osgeo.org>
>> Betreff: Re: Re: [GRASS-user] r. walk and sites which can be reached in a
>> certain time
>> On Mon, Oct 24, 2016 at 10:22 AM, Phil Burgdorf <P_Burgdorf at web.de> wrote:
>>> Hi Anna,
>>> thanks for your answer. I've tried to use 3600 as max. Cost but the
>>> algorithm failed.
>>
>> That's weird.
>>
>>> The values for min and max for the cummulative cost are "nan". But I
>>> don't
>>> know why!
>>> How can I do this in r. mapcalc. I haven't used it before
>>
>>
>> r.mapcalc "time_to_walk_limited = if(time_to_walk < 3600, time_to_walk,
>> null())"
>>
>> Please keep the discussion on the mailing list.
>>
>>
>>> Phil
>>>
>>> Gesendet: Montag, 24. Oktober 2016 um 16:02 Uhr
>>> Von: "Anna Petrášová" <kratochanna at gmail.com>
>>> An: "Phil Burgdorf" <P_Burgdorf at web.de>
>>> Cc: "GRASS user list" <grass-user at lists.osgeo.org>
>>> Betreff: Re: [GRASS-user] r. walk and sites which can be reached in a
>>> certain time
>>> On Mon, Oct 24, 2016 at 7:52 AM, Phil Burgdorf <P_Burgdorf at web.de> wrote:
>>>> Hello there,
>>>> my name is Phil and I'm a PhD Student in Archaeology and I'm analiesing
>>>> a
>>>> small area in bavaria. I'm trying to locate archaeological sites round
>>>> different strating points. Therefor I created a cost map using r. walk.
>>>> The
>>>> result is fine but how can I adapted my values. Actualy I'm interrested
>>>> in
>>>> sites which can be reached in, let's say, one hour. I've tried to change
>>>> the values for max costs but I didn't get any results. Do i have to use
>>>> a
>>>> nother algorithm? here must be an easy way to define the max cost for
>>>> that
>>>> analysis.
>>>>
>>>> I couldn''t find any help in the mailing list's archive and I#m totally
>>>> lost. Can you help me?
>>>
>>> max_cost will do the trick, don't forget that the time is in seconds.
>>> You can also use raster algebra (r.mapcalc) to extract the area within
>>> certain time.
>>>
>>> Anna
>>>
>>>>
>>>> Cheers
>>>> Phil
>>>>
>>>>
>>>> _______________________________________________
>>>> grass-user mailing list
>>>> grass-user at lists.osgeo.org
>>>> http://lists.osgeo.org/mailman/listinfo/grass-user
>>>
>>
>