Latest Contributions by ChrisDonohue__GISP

Some general comments/visual: If your data includes whole states, using only UTM 10 may be a bit problematic as Caliornia is split between UTM 10 and 11 pretty equally. Source:  https://www.google.com/search?q=Teale+Albers&ie=utf-8&oe=utf-8  The common solution at the California state level is to use the Teale Albers projections.  Also, as mentioned in the picture above, some folks use a custom "UTM 10.5". http://spatialreference.org/ref/sr-org/california-teale-albers-nad27-projection/prettywkt/  http://spatialreference.org/ref/sr-org/10/prettywkt/  I'd consider what Melita Kennedy suggested. Chris Donohue, GISP ... View more

Not sure if the answer is here (I'm not a statistician), but take a peek if you have not already: Geographically Weighted Regression (GWR)—Help | ArcGIS for Desktop  Adding https://community.esri.com/community/gis/analysis/spatial-statistics for greater exposure. Chris Donohue, GISP ... View more

Just a quick thought on a workaround - add a process after the first Extract to Points output to rename the offending field to what is needed for the input before doing the next iteration.  For example, Alter Field (Data Managment) could be used to rename the field. Alter Field—Data Management toolbox | ArcGIS Desktop  Outside of that, it would probably be helpful to see what the model looks like to get a sense of how things are currently structured.  There possibly may also be some options to use In-Line Variable Substitution or other methods to resolve this, but it depends on how it all is currently set up. Examples of inline model variable substitution—Help | ArcGIS for Desktop  Ultimately, if you will be doing this process often and are comfortable with Python, I'd go with the suggestion by Dan_Patterson and use Python instead.  Modelbuilder can be useful but it is quirky in some ways. Chris Donohue, GISP ... View more

Probably what is going on is that the field characteristics of the same named fields are not exactly the same in each table (even though they have the same names).  For example, if in one table a named field is text type and in a second table the same named field is integer.  The Merge process will treat these as different fields, even though they have the same name. I would investigate the field properties of each layer to see if this is what is tripping up the Merge. Also, another option besides Merge is to use Append (Data Management).  Start with one dataset and Append the others to it.  Be sure to use the Test option under "Schema type (optional)" when running the tool. ArcGIS Help (10.2, 10.2.1, and 10.2.2)  Chris Donohue, GISP ... View more

If the intent is to create the fillet for display purposes, it may be possible to do this with Representations.  I'm not absolutely sure offhand if there is a way to represent a fillet with Representations, but it potentially could do that. Representations allow cartographic display of data without altering the underlying data.  So if the intent is purely display, they can be a solution. They are very handy in resolving cartographic issues without having to constantly edit or copy data to so as to make changes purely for display reasons.  However, if one needs to actually alter the data as the alteration needs to be permanent, then another solution will be needed. What are representations?—Help | ArcGIS for Desktop  Caveat - from a pragmatic point of view, if the project is a short one-shot effort, Representations are probably not worth doing, as they take time to learn and then set up.  However, if you find you will be doing this type of mapping often, they can be well worth it. Chris Donohue, GISP ... View more

There are several methods to produce polygons from points.  It may be that one may be more appropriate for what you are after than another. Points to Line Feature to Polygon Aggregate Points Points to Regular Polygons (ET Geowizards) How To: Convert a point feature class to a polygon feature class  One caveat, though, is that some of the methods require an advanced license and in one case the tool is part of a 3rd party software (ET GeoWizards). Chris Donohue, GISP ... View more

It is most likely the lack of connectivity. I'd run topology to see what is going on. Also, another potential issue is it appears there are spaces in some of the folders and file names.  Use underscores or get rid of the spaces, otherwise processes may fail (and unfortunately in these cases there is no error message that directly refers to the spaces). "Network Analysis = 1% Analysis + 99% Data Cleanup" Chris Donohue, GISP ... View more

It might make a good research thesis for an aspiring graduate student. Chris Donohue, GISP ... View more

Here's a workflow for creating your network dataset: Creating a network dataset—Help | ArcGIS for Desktop  Chris Donohue, GISP ... View more

I usually take the same approach as Darren Wiens.   If Maplex isn't cutting it for full labeling (which oftentimes is the case), use it just to get things are far as you can and then create annotation.  Then bite the bullet and manually edit the annotation.  Use an numbering or letter system, with a table to the side if the client is OK with that.  If the client insists on the full label being on the map, prepare to spend lots of time manually moving labels, as Maplex, while helpful, doesn't always do the whole job effectively. The less technical challenge that often comes with this is explaining to a perplexed client why a map with just a few labels is easy to generate, while a map depicting many things crammed together takes several times longer to complete. By the way, if some aspiring developer out there comes up with a third-party automated labeling system that would solve the complex labeling issues, I know plenty of people who would be willing to spend some cash $$ on buying that solution. Chris Donohue, GISP ... View more

As mentioned by Dan Patterson creating a Service Area is the way to go.  One of the options when running the Solve for the Service Area is to Generate Lines, which once they are created can be exported as a feature class and then used as an aid in finding the specific streets. The biggest challenge will be getting a good line dataset together, as while street datasets can be commonly found for most areas, many are CAD based or other wise loosely digitized so do not have connectivity between the street lines.  All it takes is one very tiny gap to cause the analysis to miss a whole area.  So expect to spend quite a bit of time doing cleanup of the street data before running the analysis.  Running topology on the street dataset is highly recommended. Ffor walkability, you may need to add in pedestrian paths to your street data.   Also, you will need to account for any streets that do not have safe pedestrian walkways along them (so are not traversable). Finally, one will need to take into account overpasses/underpasses in areas where one cannot transfer directly from one to another.  This is typically handled by populating the Elevation field to have different values for the segment(s) above and the segment(s) below so they are treated as non-traversable when the network dataset is built. Chris Donohue, GISP "Network Analysis = 1% analysis + 99% data cleanup" ... View more

I'm not sure if it still holds up in this day and age, but on the database side memory usage is one reason.  For example, if all the values in a field represent dollars, there is no reason to have the $ sign stored with the data, as all those dollar signs take up space.  There is just lots of redundancy there that one could normalize down to a field without them and save space/make the database work more effectively. Depending upon your data, a similar argument can be made for the hastag or other symbols; if all the data for a field would have that character anyways, drop that character.  Of course, one would have to have a process to add that character back in if the data needed to be exported to someone and they are expecting the character to included with each value. The GIS reasoning for not using special characters is that they can cause issues with many GIS processes.  Some processes crash if special characters are included.  Part of this due to the way operating systems work in that they often reserve those characters for operating system commands, so the operating system gets confused when it sees those characters in data. Chris Donohue, GISP ... View more

For those who are interested, background on the development of this AddIn by Tim can be found in Summer 2017 ArcUser article: New Ways to Work | ArcUser  Chris Donohue, GISP ... View more

You could do a Select by Attribute on each layer, then save off the results to feature classes.  Then one could go on to process them by whatever is needed next, be it Merge them, Select from them further, Join them, etc. (to come up with more specifics along this line we would need to know more specifics about the data and what you want to do with it). Also, If this is a common bit of processing that would need to be run on a regular basis, one could automate the process by employing Modelbuilder or Python. Chris Donohue, GISP ... View more

I'd like to offer an alternative approach.  How about employing cartographic Representations?  This way the data can remain accurate, but be portrayed differently.  For example, one can displace a point to the side when there is clutter in the Representation of it, but the underlying data still contains the original location. What are representations?—Help | ArcGIS for Desktop  Caveats: If the project you are working on is a quick "one and done" project, then Representations are probably not worth doing.  But if the data will be displayed over and over in the same map extent (like for a map that is put out every month), I've found it is worth it to take the time to set them up. If you have not used Representations before, it will take some time to get up to speed with using them.  Some of the processes are not intuitive and their are some quirks.  But it can be well-worth it for an ongoing map effort. Chris Donohue, GISP ... View more