nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

Too many towns

CameronBouchard

I'm producing a map that is showing towns with labels. The data that I am using contains a classification for the towns with population ranges (6 classes). The class 6 towns are the ones with the lowest population but there are too many of them to allow for clear labelling. The map is full of town labels and I haven't even added hydrography (with labels) and transportation (with labels).

What I would like to do is only show about half of the class 6 towns, with no preference given to any of them. I have tried assigning random values to them and showing only half (even numbers) but the towns were not spread out spatially - which resulted in clusters that were still too close together.

I would really like to show only half of the points with a spatial dispersion that allows for clearer labelling. I'm not looking for text labelling solutions - I simply have too many points on the map.

Can anyone help with suggestions?

Btw, I'm using ArcInfo 10, if it matters.

I've attached little screengrab to show my headache.

towns1.jpg

Find more posts tagged with

Accepted answers

All comments

JimW1

Hmm this one doesn't ensure spatially distribution but it works like a charm. You can just keep running it until you see a pattern you like.

http://www.spatialecology.com/htools/rndsel.php

(Thanks google)

JimW1

How about adding a field to your towns layer called something like 'DISPLAYME' and then populate the fields with '0' for do not draw and '1' for draw. You can then use a definition query to not display (thus not label) the '0' values. As for chooseing which ones? The Thiessen polygon is an option but I'm with Gregory and you'll have to do this at least partly manually to look good and to capture towns that may have more importance than others (regional nodes, vacation spots, etc.).

There is a random selection script floating out there too that you could use to randomly select class 6 features that are spatially dispersed. I'd find a link but this new forum search drives me batty.

JoshWhite

Thank you for your help, but this is not a labelling issue.
I need to know how I can show a selection of a point file that is spatially dispersed.

I would say this IS a labeling issue but suit yourself. There isn't a whole lot you can do with this issue if the cities are all grouped together. The Thiessen polygons is probably going to be your best bet though. I haven't done work with them before so I didn't say anything about them as I didn't know anything about them. Good luck with your project.

CameronBouchard

This may save you some time (still takes some, though):

1.) Create Thiessen Polygons from your towns (pics 1 & 2)

Sounds like a plausable idea. I think I will try that.
Thanks.

DarrenWiens2

This may save you some time (still takes some, though):

1.) Create Thiessen Polygons from your towns (pics 1 & 2)
2.) Spatial Join the polygon attributes to the town table (transfers INPUT_ID from polygons to points)
3.) Join by INPUT_ID to transfer polygon areas to the points
4.) Select and export the points associated with the top 90% polygon areas (the largest polygons = the farthest apart towns). If you're brave you could rather just delete the smallest 10% points.
5.) Make new Thiessen Polygons from the new set of points
6.) Repeat until you reach an acceptable level of dispersion (I got to pic 3 & 4 in three interations)

4.png

1.png

3.png

2.png

GregoryElwood

Sorry to have to say this, but you may wish to make a copy of the data and hand edit the Class 6 towns to acheive a dispersal pattern that looks good.

Automatic geoprocessing and labelling functions are good, and can save one a great deal of time, but nothing beats the human eye for artistic design.

Gregory

CameronBouchard

Thank you for your help, but this is not a labelling issue.
I need to know how I can show a selection of a point file that is spatially dispersed.

JoshWhite

Can you send me the table or at least a screenshot of it with all of the fields, clearly I'm going to need to see your data to help you out properly. The only other solution I can think of, and its a painful one, is to convert to annotation and remove the ones that are too clustered. I finally resorted to that for my street names as I could never get them to label properly just using the standard labelling engine.

CameronBouchard

I need to keep about half of the class 6 towns.

JoshWhite

That leaves me with unlabelled towns. The result I am hoping for would have fewer towns drawn (then labelled).

Not with a definition query it wouldn't. For example run a definition query that is something like this "CLASS"<>6 or whatever your field name is. In this case not only would they not be labelled they wouldn't be drawn either.

CameronBouchard

That leaves me with unlabelled towns. The result I am hoping for would have fewer towns drawn (then labelled).

JoshWhite

I haven't really used this but you could try going in to your Label placement properties on the conflict detection tab go down to buffer and change the value there. Also adding your other labels for transportation and hydrography (depending on the settings) should wipe out a few of the city labels. The other thing may be to shorten some of the names if possible. You only other viable option really is to use a definition query to exclude certain ones. You also should really consider not labelling the Class 6 cities. I've struggled with that many times, sometimes you just can't fit everything on to a map that you want.

Join the Conversation

Signed in members can post, follow updates, and more. New here? Register a free account.

Quick Links

Community Help
Find useful guides, FAQs, and documents to help you navigate and make the most of Esri Community.