Jump to content


Representing Earthquake Magnitude

- - - - - earthquakes proportional circles symbology

  • Please log in to reply
2 replies to this topic



    Key Contributor

  • Validated Member
  • PipPipPip
  • 90 posts
  • Gender:Male
  • Location:South Carolina
  • Interests:Rivers, Urban Design, Geomorphology, Bicycling, Outdoors
  • United States

Question about representing earthquakes, in light of earthquakes being in the (at least US) news recently.  


Often, you see maps with many earthquakes symbolized by magnitude with some sort of graduated circles. This is mostly done with the circles being sized linearly (magnitude 5 is some linear step larger than magnitude 4).  However, earthquake magnitude is a on a log scale.  My question; does it make more sense, or convey information more accurately, to represent earthquakes such that the circles are sized on a log scale rather than linearly?  I know this presents some other cartographic issues, such as leading people to believe that the circle represents an area in which the earthquakes was felt. 


I included two images... one of earthquakes around South Carolina represented with log scaled circles (quick and dirty, might use R to do this in a nicer way). 

The other is an info graphic I found when searching this idea.  It makes sense on an infographic, but does it on a map? 


To me, scaling the symbols this way gets across that the difference in magnitude between 4 and 5 is actually 10 times. And the difference between 4 and 6 is 100 times!  If the point of the map is to compare relative earthquake "size" in different locations, than this should be considered. 








    Hall of Fame

  • Validated Member
  • PipPipPipPipPipPipPip
  • 1,099 posts
  • Gender:Not Telling
  • Location:Redlands, CA
  • United States

What you're saying makes sense to me, matching the symbology better to what the data represents. Like you said, it does present some other challenges, but it does help make the fewer large earthquakes stand out better from the many smaller ones.

Dave Barnes
Product Engineer
Map Geek




  • New Member
  • Pip
  • 1 posts
  • Gender:Male
  • No Country Selected

There are two extreme approaches: 1) area of circle (not radius) proportional to magnitude M; 2) area proportional 10^M. Both tend to produce unsatisfactory results. I tend to use a compromise solution with area proportional to b^M where the base is 2<b<4.

Also tagged with one or more of these keywords: earthquakes, proportional, circles, symbology

0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users