Hi everyone,
I’ve run in to a problem that I hope to get some help with.
I’m preparing a set of rasters to perform species distribution modelling. For one
of my predictors, I have chosen to use r.walk in order to record how far away my occurrence
records are to the nearest shoreline. Here’s the catch though… I will be generating my walk-distance
rasters by using the GEBCO bathymetry data set. Here in lies the problem. Seeing as it is bathymetry, my raster will have negative values. My initial thought was simply to run the bathymetry raster through r.mapcalc in this way:
r.mapcalc “expression=prehist_dem = bathymetry + 866”
In this example, “866” references the lowest depth recorded (i.e. -866m). Using the expression above, I have
now removed any negative value in the raster so that min = 0. However, while doing this I have now also added
“866” to every other cell in the raster, and not only will this be incorrect for places inland where a cell that originally
was 5 m.a.s.l. now is 871m, but it will probably affect the cells close to the sea in the same way (which in most cases should probalby be close to, or slightly above, 0m). More importantly, if I would just stick to this approach, I can’t help but imagine that it won’t produce inaccurate results for r.walk further down the line as well.
I can’t help but think that there is some clever work-around to this using the mapcalculator, but I’m simply stuck
and don’t know how to proceed. Hope any of you can provide some advice!
Best,
Victor
Hi Victor,
I am not sure I understand your concern. Have you tried running r.walk with the original raster with negative values? Theoretically, I don’t see why r.walk couldn’t work with negative elevation, although I haven’t tried it. It should work the same if you add a constant value as you suggest. Perhaps you want to e.g. use an absolute value of the elevation?
Anna
On Wed, Dec 20, 2023 at 5:59 PM Victor Lundström via grass-user <grass-user@lists.osgeo.org> wrote:
Hi everyone,
I’ve run in to a problem that I hope to get some help with.
I’m preparing a set of rasters to perform species distribution modelling. For one
of my predictors, I have chosen to use r.walk in order to record how far away my occurrence
records are to the nearest shoreline. Here’s the catch though… I will be generating my walk-distance
rasters by using the GEBCO bathymetry data set. Here in lies the problem. Seeing as it is bathymetry, my raster will have negative values. My initial thought was simply to run the bathymetry raster through r.mapcalc in this way:
r.mapcalc “expression=prehist_dem = bathymetry + 866”
In this example, “866” references the lowest depth recorded (i.e. -866m). Using the expression above, I have
now removed any negative value in the raster so that min = 0. However, while doing this I have now also added
“866” to every other cell in the raster, and not only will this be incorrect for places inland where a cell that originally
was 5 m.a.s.l. now is 871m, but it will probably affect the cells close to the sea in the same way (which in most cases should probalby be close to, or slightly above, 0m). More importantly, if I would just stick to this approach, I can’t help but imagine that it won’t produce inaccurate results for r.walk further down the line as well.
I can’t help but think that there is some clever work-around to this using the mapcalculator, but I’m simply stuck
and don’t know how to proceed. Hope any of you can provide some advice!
Best,
Victor
grass-user mailing list
grass-user@lists.osgeo.org
https://lists.osgeo.org/mailman/listinfo/grass-user
Hi Anna, thank you for responding!
No I haven’t tried running r.walk using negative values, I guess I could give it a try. But, if I remember correctly, didn’t Michael Barton inquired about the use of negative values and r.walk previously. I can’t remember the outcome of that discussion, and I don’t seem to have kept the emails unfortunately. I’ll give the negative values a try, if it works, then that will at least be less distorted as opposed to removing negative values by addition.
/Victor
Skickat från Outlook för iOS
···
Från: Anna Petrášová kratochanna@gmail.com
Skickat: Thursday, December 21, 2023 5:18:03 AM
Till: Victor Lundström Victor.Lundstrom@uib.no
Kopia: grass-user@lists.osgeo.org grass-user@lists.osgeo.org
Ämne: Re: [GRASS-user] re-setting bathymetry data using the raster calculator.
Hi Victor,
I am not sure I understand your concern. Have you tried running r.walk with the original raster with negative values? Theoretically, I don’t see why r.walk couldn’t work with negative elevation, although I haven’t tried it. It should work the same if you add a constant value as you suggest. Perhaps you want to e.g. use an absolute value of the elevation?
Anna
On Wed, Dec 20, 2023 at 5:59 PM Victor Lundström via grass-user <grass-user@lists.osgeo.org> wrote:
Hi everyone,
I’ve run in to a problem that I hope to get some help with.
I’m preparing a set of rasters to perform species distribution modelling. For one
of my predictors, I have chosen to use r.walk in order to record how far away my occurrence
records are to the nearest shoreline. Here’s the catch though… I will be generating my walk-distance
rasters by using the GEBCO bathymetry data set. Here in lies the problem. Seeing as it is bathymetry, my raster will have negative values. My initial thought was simply to run the bathymetry raster through r.mapcalc in this way:
r.mapcalc “expression=prehist_dem = bathymetry + 866”
In this example, “866” references the lowest depth recorded (i.e. -866m). Using the expression above, I have
now removed any negative value in the raster so that min = 0. However, while doing this I have now also added
“866” to every other cell in the raster, and not only will this be incorrect for places inland where a cell that originally
was 5 m.a.s.l. now is 871m, but it will probably affect the cells close to the sea in the same way (which in most cases should probalby be close to, or slightly above, 0m). More importantly, if I would just stick to this approach, I can’t help but imagine that it won’t produce inaccurate results for r.walk further down the line as well.
I can’t help but think that there is some clever work-around to this using the mapcalculator, but I’m simply stuck
and don’t know how to proceed. Hope any of you can provide some advice!
Best,
Victor
grass-user mailing list
grass-user@lists.osgeo.org
https://lists.osgeo.org/mailman/listinfo/grass-user
|
[External email] Make sure you recognize the sender’s email address before you click links, open attachments, or get involved in financial transactions. Contact IT-support BRITA if you have any questions.
|
On Thu, Dec 21, 2023 at 4:45 AM Victor Lundström <Victor.Lundstrom@uib.no> wrote:
Hi Anna, thank you for responding!
No I haven’t tried running r.walk using negative values, I guess I could give it a try. But, if I remember correctly, didn’t Michael Barton inquired about the use of negative values and r.walk previously. I can’t remember the outcome of that discussion, and I don’t seem to have kept the emails unfortunately. I’ll give the negative values a try, if it works, then that will at least be less distorted as opposed to removing negative values by addition.
He was testing negative values in a friction map, but that’s a different case. Negative values in elevation (that’s your case, right?) should not matter I would hope.
/Victor
Skickat från Outlook för iOS
Från: Anna Petrášová <kratochanna@gmail.com>
Skickat: Thursday, December 21, 2023 5:18:03 AM
Till: Victor Lundström <Victor.Lundstrom@uib.no>
Kopia: grass-user@lists.osgeo.org <grass-user@lists.osgeo.org>
Ämne: Re: [GRASS-user] re-setting bathymetry data using the raster calculator.
Hi Victor,
I am not sure I understand your concern. Have you tried running r.walk with the original raster with negative values? Theoretically, I don’t see why r.walk couldn’t work with negative elevation, although I haven’t tried it. It should work the same if you add a constant value as you suggest. Perhaps you want to e.g. use an absolute value of the elevation?
Anna
On Wed, Dec 20, 2023 at 5:59 PM Victor Lundström via grass-user <grass-user@lists.osgeo.org> wrote:
Hi everyone,
I’ve run in to a problem that I hope to get some help with.
I’m preparing a set of rasters to perform species distribution modelling. For one
of my predictors, I have chosen to use r.walk in order to record how far away my occurrence
records are to the nearest shoreline. Here’s the catch though… I will be generating my walk-distance
rasters by using the GEBCO bathymetry data set. Here in lies the problem. Seeing as it is bathymetry, my raster will have negative values. My initial thought was simply to run the bathymetry raster through r.mapcalc in this way:
r.mapcalc “expression=prehist_dem = bathymetry + 866”
In this example, “866” references the lowest depth recorded (i.e. -866m). Using the expression above, I have
now removed any negative value in the raster so that min = 0. However, while doing this I have now also added
“866” to every other cell in the raster, and not only will this be incorrect for places inland where a cell that originally
was 5 m.a.s.l. now is 871m, but it will probably affect the cells close to the sea in the same way (which in most cases should probalby be close to, or slightly above, 0m). More importantly, if I would just stick to this approach, I can’t help but imagine that it won’t produce inaccurate results for r.walk further down the line as well.
I can’t help but think that there is some clever work-around to this using the mapcalculator, but I’m simply stuck
and don’t know how to proceed. Hope any of you can provide some advice!
Best,
Victor
grass-user mailing list
grass-user@lists.osgeo.org
https://lists.osgeo.org/mailman/listinfo/grass-user
|
[External email] Make sure you recognize the sender’s email address before you click links, open attachments, or get involved in financial transactions. Contact IT-support BRITA if you have any questions.
|