Steep Slope analysis accuracy?

[C3D_RickGraham]

Hi all,

I was doing a quick steep slope analysis on a surface. My criteria was that
I have one range that goes from 15% to Vertical (I want to see all slopes
that are 15% and higher).

I'm now spot checking the surface. When I add a surface label (two point)
for slope on a section. I get results that are different than what I told
the analysis. I'nm now not trusting either the slope analysis or the surface
label.

I also ran a quick profile through an area and hand-calced it out. The label
is correct. So I'm REALLY not trusting the slope analysis! Anyone have any
similar experiences? I can provide a drawing to those who want to take a
gander at it. (rgraham AT jrholley DOT com)

--
Rick

C3D 2007 SP3
Dell DuoCore 2.66GHz 2GB Ram 256 Dual DVI Dual monitors! 🙂

[jwedding]

Two point slope labels do a simple rise over run across the entire distance.
Slope analysis looks at the centroid of each triangular face. You're
comparing oranges and tangerines. Sort of the same, but different once you
get at it.

--
James Wedding, P.E.
Engineered Efficiency, Inc.
Civil 3D 2007
XP Tablet, SP2, 2GHz, 2G
www.eng-eff.com
www.civil3d.com

[BrianHailey]

My first thoughts as well. Try using the one point slope method. This will label the slope of whatever triangle it happens to be in.

Brian

[C3D_RickGraham]

But if I do a distance from one contour perpendicular to the next contour
and do my slope by hand it calcs out correctly. My engineers are screaming
because of the extraneous little trianglulated spots, and in some cases I
have a large area that is shaded in except for a small triangle sliver that
is not. It makes no sense

Which leads to my next questions - IN THEORY - Slope analysis should be
measure perpendicularly to the contour (water running down hill). How come
it isn't doing this? Perhaps I'm just not grasping the concept here. I WISH
there was a way tha tI could take this steep slope and make it somthing
thatI could do a BPOLY on and hatch myself instead of have millions of
little triangles on the drawing. Another nice wishlist would be the ability
to say "I want to only see the steep slopes on areas that are laarger than
2000 square feet (or whatever number the municipality says they need to
see).

Thanks for the reply James.

Rick
wrote in message news:5526436@discussion.autodesk.com...
Two point slope labels do a simple rise over run across the entire distance.
Slope analysis looks at the centroid of each triangular face. You're
comparing oranges and tangerines. Sort of the same, but different once you
get at it.

--
James Wedding, P.E.
Engineered Efficiency, Inc.
Civil 3D 2007
XP Tablet, SP2, 2GHz, 2G
www.eng-eff.com
www.civil3d.com

[dana.probert]

it's tin based instead of contour based.

you might get more smoothy PA friendly looking results if you copy your surface, then explode it while it looks like contours, then build a surface out of that and then run the analysis cause then it is more of a contour steep slope instead of a tin steep slope.

it does best when you have that copied temp surface have a tight contour interval before you explode it.

PA agencies tend to want a contour based steep slope analysis.

[Anonymous]

Hi Rick,

Slope analysis deals with the maximum slope within a triangle in the DTM and
these are always irregular in location in any area with artificial
construction. A typical kerb face might have slope between 500% and 1000%
Because of it narrow width, if the survey picks up the toe and top closer
than they actually are, you may end up with a reported value of 2000%

Reported slope between contours in general will be correct, because it
ignores the irregularities between the contours.

If you run a water drop analysis, you will see that the flow path always
crosses a contour at right angles, but depending on the surface may flow a
very irregular path between the contours. In general the flow path length
will be longer than the distance between the contours.

The type of algorithms required to amalgamate triangles into larger areas
are impossibly complex if they are to have meaningful output, So although it
would be nice to say only provide maximum slope for areas > some value, a
human is far better at judging which data items to omit than a computer.

For purposes of keeping your engineers happy, one approach is to create a
grid of points on the surface and then build a new surface from those
points.

--

Laurie Comerford
CADApps
www.cadapps.com.au
www.civil3Dtools.com


wrote in message news:5526494@discussion.autodesk.com...
But if I do a distance from one contour perpendicular to the next contour
and do my slope by hand it calcs out correctly. My engineers are screaming
because of the extraneous little trianglulated spots, and in some cases I
have a large area that is shaded in except for a small triangle sliver that
is not. It makes no sense

Which leads to my next questions - IN THEORY - Slope analysis should be
measure perpendicularly to the contour (water running down hill). How come
it isn't doing this? Perhaps I'm just not grasping the concept here. I WISH
there was a way tha tI could take this steep slope and make it somthing
thatI could do a BPOLY on and hatch myself instead of have millions of
little triangles on the drawing. Another nice wishlist would be the ability
to say "I want to only see the steep slopes on areas that are laarger than
2000 square feet (or whatever number the municipality says they need to
see).

Thanks for the reply James.

Rick
wrote in message news:5526436@discussion.autodesk.com...
Two point slope labels do a simple rise over run across the entire distance.
Slope analysis looks at the centroid of each triangular face. You're
comparing oranges and tangerines. Sort of the same, but different once you
get at it.

--
James Wedding, P.E.
Engineered Efficiency, Inc.
Civil 3D 2007
XP Tablet, SP2, 2GHz, 2G
www.eng-eff.com
www.civil3d.com