# Proportional Symbol Maps

7 replies to this topic

### #1 kunigami Posted 24 September 2010 - 10:20 AM

kunigami

Newbie

• Validated Member
• 5 posts
• Brazil

Hi,

I'm studying a problem regarding proportional symbol maps, using as symbols, opaque circles. In my problem, an input consists of the set of circles that are showed in the map. Each circle is given by the cartesian coordinates of its center (x, y) and by its radius.

In general, data is given in latitute, longitude and quantity. Coordinates (x, y) are obtained from (lat, lon) through some projection. The area is proportional to the quantity, so radius is proportional to the square root of this quantity. But, choosing the "proportion factor" is done visually, right?

My two questions are:

1) What's the best tool to visualize those disks and choose this "proportion factor", in such a way that I can later obtain the coordinates and radius of the disks?

2) Is there any database with the coordinates of the symbols already projected and its radius already defined by some cartographer?

Thanks,

### #2 CHART Posted 25 September 2010 - 09:55 AM

CHART

Chart

• Validated Member
• 358 posts
• No Country Selected

This link might be of help.

http://jenny.cartography.ch/legend/
Chart

### #3 kunigami Posted 27 September 2010 - 07:00 AM

kunigami

Newbie

• Validated Member
• 5 posts
• Brazil

This link might be of help.

http://jenny.cartography.ch/legend/

Sorry, I couldn't see how this might help with what I've asked.

### #4 l.jegou Posted 27 September 2010 - 09:30 AM

l.jegou

Master Contributor

• Validated Member
• 114 posts
• Gender:Male
• Location:Toulouse
• Interests:Webmapping, thematical mapping.
• France

1) What's the best tool to visualize those disks and choose this "proportion factor", in such a way that I can later obtain the coordinates and radius of the disks?

2) Is there any database with the coordinates of the symbols already projected and its radius already defined by some cartographer?

Hi, i don't know if i understand clearly your questions, it could be useful to provide some context (what type of project : software dev., map production ?)

For your 1) any thematic cartography software, or GIS software with decent thematic cartography functions could do. You load your points, the attributes data to be represented by the surface of the disk, and the software will draw the map.

For your 2), i don't understand. If the symbols are discs, their centers are positioned on the lat/long points. The coordinates of the points composing the perimeter of the disc are function of the radius, and calculated automatically by the cartographic or GIS software. It's possible to do the calculation with the formula of the geographic projection used.

### #5 Dennis McClendon Posted 27 September 2010 - 01:02 PM

Dennis McClendon

Hall of Fame

• Validated Member
• 1,176 posts
• Gender:Male
• Location:Chicago
• Interests:map design, large-scale maps of cities
• United States

Even if using illustration software, it's a simple matter to just set up a little spreadsheet and calculate the diameter of the circles from the quantity.
Dennis McClendon, Chicago CartoGraphics
chicagocarto.com

### #6 kunigami Posted 28 September 2010 - 12:06 PM

kunigami

Newbie

• Validated Member
• 5 posts
• Brazil

1) What's the best tool to visualize those disks and choose this "proportion factor", in such a way that I can later obtain the coordinates and radius of the disks?

2) Is there any database with the coordinates of the symbols already projected and its radius already defined by some cartographer?

Hi, i don't know if i understand clearly your questions, it could be useful to provide some context (what type of project : software dev., map production ?)

For your 1) any thematic cartography software, or GIS software with decent thematic cartography functions could do. You load your points, the attributes data to be represented by the surface of the disk, and the software will draw the map.

For your 2), i don't understand. If the symbols are discs, their centers are positioned on the lat/long points. The coordinates of the points composing the perimeter of the disc are function of the radius, and calculated automatically by the cartographic or GIS software. It's possible to do the calculation with the formula of the geographic projection used.

Actually I'm implementing a software that deals with a set of circles, trying to choose their placement order when they overlap, in such a way an objective funcion is maximized. Computationally, a circle is seen as a coordinate (x,y) and a radius. One can see that, depending on the chosen radius, there may be more or less overlapping.

One obvious application of this work is proportional symbol mapping, thus I'd like to get real instances to test my program. There are several data available in form of (latitute, longitude, quantity), but I'd like a way to transform it to (x, y, radius) automatically (or automagically) since there may be hundreds of such disks. If I use thematic cartography software, it probably won't return me the numerical values of the radius from the disks it is plotting, right?

Anyway, are there any free/open source such softwares you recommend?

Thanks,

### #7 l.jegou Posted 29 September 2010 - 01:13 AM

l.jegou

Master Contributor

• Validated Member
• 114 posts
• Gender:Male
• Location:Toulouse
• Interests:Webmapping, thematical mapping.
• France

Ok, i think that i see what you are looking for. My advice is to try to use an open source thematic cartography software and add some export function to retrieve the radii of the generated discs, but it's somewhat complicated.

As a cartographer, when i've to draw such discs, my methodology is guided by constraints, ordered descending. But the constraints varies very much from map to map, as function of the placement of the points and the values to map. So the largest constraint is often : "where is the biggest disc to draw, and how big can i draw it, without masking too much of the background or overlapping other discs ?" In my own open-source thematic cartography software for pedagogy i start the procedure for such maps with the choice of the size of the largest disc.

You can also look into "Thematic cartography and Visualization" by Terry A. Slocum (Prentice-Hall editors), which has a complete chapter about this type of maps and several sub-techniques.

My personnal opinion is that it will be difficult to computerize a visual / try-and-adapt technique, but current softwares are not very efficient, however

### #8 kunigami Posted 12 October 2010 - 05:04 PM

kunigami

Newbie

• Validated Member
• 5 posts
• Brazil

Ok, i think that i see what you are looking for. My advice is to try to use an open source thematic cartography software and add some export function to retrieve the radii of the generated discs, but it's somewhat complicated.

I managed to implement a simple tool to do this for me, using Google Maps API! It has been serving me quite well, so my new task is to find some suitable data for proportional maps. I'll start a new topic to ask for (free) references of such datasets.

As a cartographer, when i've to draw such discs, my methodology is guided by constraints, ordered descending. But the constraints varies very much from map to map, as function of the placement of the points and the values to map. So the largest constraint is often : "where is the biggest disc to draw, and how big can i draw it, without masking too much of the background or overlapping other discs ?" In my own open-source thematic cartography software for pedagogy i start the procedure for such maps with the choice of the size of the largest disc.

You can also look into "Thematic cartography and Visualization" by Terry A. Slocum (Prentice-Hall editors), which has a complete chapter about this type of maps and several sub-techniques.

I think I already read the chapter of this book about Proportional Symbol Maps, but as far as I remember, the only thing I could learn from that was that the radii of the disks should be chosen so that the map "was neither too empty, neither too full". I'll take a second and deeper look in this and other chapters to see if I'm missing something.

My personnal opinion is that it will be difficult to computerize a visual / try-and-adapt technique, but current softwares are not very efficient, however

#### 0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users

-->