The output maps from r.slope.aspect. r.resamp.rst, and even r.topidx the
output maps have values for each cell rather than trends. This is very
difficult to interpret by non-technical viewers.
I've seen maps of slopes, profile and planar curvature, and topographic
wetness indices that are more smoothly colored. If one wants to determine
the curvature of slopes adjacent to a specific cell at its base it would be
easier to see this if the values do not change with each cell. When we stand
at the foot of a hill we tend to not focus on each detail of the terrain but
the overall slope, profile curvature, and planar curvature. It's this type
of overview that I'm trying to display.
Is there a way to get smoother output maps that are visually more
informative to non-specialists?
The output maps from r.slope.aspect. r.resamp.rst, and even r.topidx the
output maps have values for each cell rather than trends. This is very
difficult to interpret by non-technical viewers.
I've seen maps of slopes, profile and planar curvature, and topographic
wetness indices that are more smoothly colored. If one wants to determine
the curvature of slopes adjacent to a specific cell at its base it would be
easier to see this if the values do not change with each cell. When we stand
at the foot of a hill we tend to not focus on each detail of the terrain but
the overall slope, profile curvature, and planar curvature. It's this type
of overview that I'm trying to display.
Is there a way to get smoother output maps that are visually more
informative to non-specialists?
Filter the input data with r.neighbors[1], r.mfilter, etc.
[1] I wouldn't use method=mode if you're going to calculate
derivatives. I'm not so sure about method=median either.
The output maps from r.slope.aspect. r.resamp.rst, and even r.topidx the
output maps have values for each cell rather than trends. This is very
difficult to interpret by non-technical viewers.
I've seen maps of slopes, profile and planar curvature, and topographic
wetness indices that are more smoothly colored. If one wants to determine
the curvature of slopes adjacent to a specific cell at its base it would be
easier to see this if the values do not change with each cell. When we
stand at the foot of a hill we tend to not focus on each detail of the
terrain but the overall slope, profile curvature, and planar curvature.
It's this type of overview that I'm trying to display.
Is there a way to get smoother output maps that are visually more
informative to non-specialists?
I have found that with a certain amount of smoothing (r.param.scale ,
r.neighbors) and the right color scheme (r.colors [-e]) one can generate
pretty nice looking maps that display overall trends.
One thing to remember-- most GRASS modules are geared towards analysis--
making pretty maps comes second.
I have found that with a certain amount of smoothing (r.param.scale ,
r.neighbors) and the right color scheme (r.colors [-e]) one can generate
pretty nice looking maps that display overall trends.
Dylan,
I used r.param.scale extensively; will learn r.neighbors.
One thing to remember-- most GRASS modules are geared towards analysis--
making pretty maps comes second.
That's very true. However, when the analytical results need to be
presented to non-technical decision-makers (lawyers, regulators,
politicians) it's vital that the presentations be visually attractive. It
adds credibility to the analytical results and, like typeset docuements
rather than word processed ones, makes them more easily accepted.
On Tue, 2 Mar 2010, Dylan Beaudette wrote:
> I have found that with a certain amount of smoothing (r.param.scale ,
> r.neighbors) and the right color scheme (r.colors [-e]) one can generate
> pretty nice looking maps that display overall trends.
Dylan,
I used r.param.scale extensively; will learn r.neighbors.
> One thing to remember-- most GRASS modules are geared towards analysis--
> making pretty maps comes second.
That's very true. However, when the analytical results need to be
presented to non-technical decision-makers (lawyers, regulators,
politicians) it's vital that the presentations be visually attractive. It
adds credibility to the analytical results and, like typeset docuements
rather than word processed ones, makes them more easily accepted.
I agree 100% -- All I was getting at was that the nice map-making requires
some creativity. If you have some time check out GMT-- it can generate some
very nice looking maps in PS/EPS format. There are some scripts on the GRASS
add-ons wiki page with some ideas.