topic Re: Why are Modelbuilder "Bearing Distance To Line" results different from manually drawn direction/distance function in ModelBuilder Questions

Why are Modelbuilder "Bearing Distance To Line" results different from manually drawn direction/distance function

RobertReyes — Sun, 24 Jun 2018 22:06:33 GMT

I created a model to plot a line for specified distances along a specified azimuth, then plot another line from the end vertice of that line for a specified distance in a perpendicular direction, then generate a point from end point of the second line. I used the “Bearing Distance To Line” tool to pull from a table that lists degree values to four decimal points (for example, 132.2458, 45.3789, etc.). The model seems to run fine, but the orientation of the automatically generated line is different from a manually generated line using the distance/direction function, even though the inputs are exactly the same. I thought maybe the projections were different for the two feature classes, but they are the same as well. I cannot figure out why this is happening or which result is more accurate, and was hoping someone might be able to help.

Re: Why are Modelbuilder "Bearing Distance To Line" results different from manually drawn direction/distance function

DanPatterson_Retired — Mon, 25 Jun 2018 00:33:14 GMT

When you say 'manual' are your referring to using the tool itself outside of your model?

Bearing Distance To Line—Data Management toolbox | ArcGIS Desktop (there is one for arcmap)

I wouldn't compare the model output to anything other than the tool output.

Re: Why are Modelbuilder "Bearing Distance To Line" results different from manually drawn direction/distance function

RobertReyes — Mon, 25 Jun 2018 15:08:00 GMT

Thank you, Dan. Sorry that I wasn't more clear. When I said "manual" I meant using edit to create a new line feature, right-clicking as it is being drawn, selecting the direction/distance function, and then inputting the bearing in the "Horizontal/NAz" field, and the distance in the "Distance/ftUS" field. I left the "Pitch/dd" set at it's default value of 0.

Re: Why are Modelbuilder "Bearing Distance To Line" results different from manually drawn direction/distance function

DanPatterson_Retired — Mon, 25 Jun 2018 15:55:49 GMT

pardon the pun... but that is a 'sketchy' method and I don't know if it has the same controls over it that creating a featureclass has. I would perform the comparison as I suggested first

Re: Why are Modelbuilder "Bearing Distance To Line" results different from manually drawn direction/distance function

RobertReyes — Mon, 25 Jun 2018 16:52:36 GMT

I did follow your suggestion and used the "Bearing Distance to Line" from the data management toolbox, and that line drew the same as the one derived from the model. I'm still trying to figure out why the "sketch" method is different and which one is more accurate. I assume from your suggestion that you think the one from the data management toolbox is more accurate; my concern was whether it was rounding the azimuth value (which has four decimal points) before applying it, while the "sketchy" method was not. Thanks.

Re: Why are Modelbuilder "Bearing Distance To Line" results different from manually drawn direction/distance function

DanPatterson_Retired — Mon, 25 Jun 2018 19:06:08 GMT

Robert... you would have to do the "maths" to see whether angle 132.2458 versus 132 would produce the deviation at your distance.

Re: Why are Modelbuilder "Bearing Distance To Line" results different from manually drawn direction/distance function

RobertReyes — Mon, 23 Jul 2018 15:27:34 GMT

Dan, I'm not sure what you meant when you suggested doing the "maths." However, I dropped all decimals, then again ran the "bearing distance to line" tool and compared the results with a line created using the editor. I drew "L-shaped" lines using the same input for both tools: Line 1 was an azimuth of 312 degrees for a distance of 150 ft followed by an azimuth of 42 degrees for 50 feet; Line 2 was on an azimuth of 312 degrees for a distance of 200 ft followed by an azimuth of 42 degrees for 80 feet. I used the same coordinate system (NAD 1983 StatePlane Virginia North FIPS 4501 Feet) for both feature classes, but the results were again different (see attached graphic), with a difference of 3.14 feet on the longest line and 2.31 feet on the shorter line. I did not use any decimals, so it can't be a rounding error. This is very frustrating to me because I don't know why they wouldn't be the same, nor do I know which output to trust.

Re: Why are Modelbuilder "Bearing Distance To Line" results different from manually drawn direction/distance function

DanPatterson_Retired — Mon, 23 Jul 2018 17:09:23 GMT

I would file a Tech Support issue since you should be able to use both methods.

If the dataframe and the output featureclasses are in the same coordinate systems, then you have ruled that out.

I would trust the tool but... who knows.

I am more concerned that the angles are slightly different from the same origin. If the angle specification was identical and the coordinates systems were identical, then they should be the same. The distance issue could be due to a datum difference or some other specification. Just ship it off for testing with your data would be your best bet.

Re: Why are Modelbuilder "Bearing Distance To Line" results different from manually drawn direction/distance function

RobertReyes — Mon, 23 Jul 2018 17:59:39 GMT

Thank you, Dan. I used exactly the same parameters for each method and I agree that the angles appear to be more of a problem than the distances. I will give tech support a try.

Re: Why are Modelbuilder "Bearing Distance To Line" results different from manually drawn direction/distance function

kmsmikrud — Wed, 17 Oct 2018 16:36:23 GMT

Hi Robert,

I was searching and found your post regarding similar results I'm getting for drawing a line with the same bearing using the editor direction/distance compared to the Bearing Distance to Line tool. I'm really curious about the difference in angles between the lines. Did you get a response from tech support on this?

Thanks,

Kathy

Re: Why are Modelbuilder "Bearing Distance To Line" results different from manually drawn direction/distance function

DanPatterson_Retired — Wed, 17 Oct 2018 16:47:37 GMT

are you using a coordinate system with feet as well Kathy?

Re: Why are Modelbuilder "Bearing Distance To Line" results different from manually drawn direction/distance function

kmsmikrud — Wed, 17 Oct 2018 16:50:04 GMT

I was using WGS84 for both the feature class line I was editing and also the Bearing Distance to Line tool. I know the distance is hard to measure with the editor directions/length since the length units but its the difference in angles that is really quite different. I have the editor set to Direction Type: North Azimuth so they should be comparable right for the angles between the two methods?

Re: Why are Modelbuilder "Bearing Distance To Line" results different from manually drawn direction/distance function

DanPatterson_Retired — Wed, 17 Oct 2018 16:58:03 GMT

you would need geodesic values for both if you are working with a Geographic Coordinate System. I think the tool works best (only?) with projected data.

Re: Why are Modelbuilder "Bearing Distance To Line" results different from manually drawn direction/distance function

kmsmikrud — Wed, 17 Oct 2018 17:09:18 GMT

The geodesic values apply to the length correct, but shouldn't the angle of the lines be the same between the two tools? The Bearing Distance to Line says "The angles are measured clockwise from North" and in the Editor if I choose North Azimuth, the help says "the azimuth of a line is the horizontal angle measured from a meridian to the line, measured in the clockwise direction from north" and the schematic looks like a compass. Below the two lines are both originating from the same point with a bearing of 297,the cyan line is delineated based on the Editor -Direction and the green line is the Bearing Distance to Line tool using that same starting point bearing, then specifying distance in miles with a rhumb line. Like I said I know the distance isn't comparable since WGS84 doesn't have measurable distance units when editing but I feel like there must be something obvious in my methods for such a big difference?

Re: Why are Modelbuilder "Bearing Distance To Line" results different from manually drawn direction/distance function

DanPatterson_Retired — Wed, 17 Oct 2018 17:25:43 GMT

if the cyan line is starting from the peninsula, it's bearing looks about right. No clue what the other line is

Re: Why are Modelbuilder "Bearing Distance To Line" results different from manually drawn direction/distance function

kmsmikrud — Wed, 17 Oct 2018 17:31:23 GMT

The cyan line is the 297 bearing using the editor direction function for editing/adding a line. The greenish line was created using that Bearing Distance to Line tool. Actually someone else put in that line and then I just tried the tool and got the very same results. If the bearing doesn't look correct in the Bearing Distance to Line tool output then why is that? The benefit of the tool when using WGS84 is being able to have it draw a line in miles whereas I'm not able to do that editing a line using direction/distance with WGS84 since decimal degrees is the units or is there a workaround. Shouldn't both of these line angles be the same? I'm working on a project that I know someone else used the Bearing Distance to Line tool quite a bit and it doesn't make me feel very confident in our lines if the angles are this far off in cases where the tool was used.

Re: Why are Modelbuilder "Bearing Distance To Line" results different from manually drawn direction/distance function

RobertReyes — Wed, 17 Oct 2018 17:33:21 GMT

Kathy and Dan, I did speak at length with ESRI technical support about this issue. I went through three different levels of support and at the end the tech support analyst said that both outputs were correct. The difference, according to tech support, is that the editor tool draws lines using "planar" geometry, while the bearing distance to line tool draws lines in a "geodesic" geometry. The tools cannot at this time draw lines in the same geometry, so they will appear different. Tech Support said they would submit a request to the programmers for an "enhancement" to address this problem, which would be incorporated in some future update to the program. I just downloaded ArcGIS Pro 2.2.3, and it doesn't appear that they have fixed the problem in that version yet. I personally think this is a fairly major discrepancy; maybe if you reach out to ESRI we can get some momentum on the issue?

Re: Why are Modelbuilder "Bearing Distance To Line" results different from manually drawn direction/distance function

DanPatterson_Retired — Wed, 17 Oct 2018 17:42:14 GMT

azimuth cases you should try from a point relative to north

22.5, 45. , 67.5, 90. , 112.5, 135. , 157.5, 180. , 202.5, 225. , 247.5, 270. , 292.5, 315. , 337.5

you should be able to ascertain where things are breaking down since this is nothing more than a circle increment by 22.5 degrees clockwise from North.

Re: Why are Modelbuilder "Bearing Distance To Line" results different from manually drawn direction/distance function

DanPatterson_Retired — Wed, 17 Oct 2018 17:45:15 GMT

That's what I figured... It needs clearer documentation to that effect. In fact drawing a 'clock/circle' to show the distortion would help assess the angles relative to it

Re: Why are Modelbuilder "Bearing Distance To Line" results different from manually drawn direction/distance function

RobertReyes — Wed, 17 Oct 2018 17:58:54 GMT

Until ESRI programmers fix the "problem," though, it doesn't seem like it's a good practice to mix lines drawn with the different tools.