Tuesday, July 2, 2013

Mine Depth (Make your axes meaningful!)



At the Empire Mine State Historic Park, you can walk 30 feet into the old gold mine, meet a quirky and knowledgeable blacksmith named David, see an awesome scale model of the mine (above--very secret during the years of operation!), and see this chart visualizing the depth of the mine over time:


I usually like axes to be as meaningful as possible, so I redrew the chart (code below) representing the year with the horizontal axis:
The original version probably has its own advantages as well, but some observations and questions comes a little more quickly from my version:

  • From 1886 onward, the mine seems to get deeper as a roughly constant rate (averaged over time).
  • Are they digging it deeper smoothly over time, or does the depth increase in spurts?
  • Why do we have so much more data in the period from 1886 to 1892 than the rest?

Before drawing the chart this way, I figured that depth would increase faster in more recent years (with better technology), which isn't true at all. But note that this is just the maximum depth -- maybe they can dig much more easily with better technology, but that's spread across more tunnels, so the depth doesn't increase any faster? Or maybe each bar on the original chart represents a short, concentrated digging effort (meaning they can increase the depth much faster in recent years, but also go for much longer without increasing it at all)?





The code:

library("ggplot2")

mineDepths.string ='"Year","Depth"
1854,102
1865,201
1874,1250
1886,1600
1888,1700
1890,1900
1892,2100
1901,3000
1914,4600
1924,6200"
1956,11000'


depths = read.csv(text=mineDepths.string)
depths$Depth = -depths$Depth

ggplot(data=depths, aes(x=Year, y=Depth)) + 
  geom_segment(mapping=aes(xend=Year, yend=0), size=3) +
  geom_text(aes(label = Year), y=0, size=5, hjust=0, angle=90) + 
  scale_y_continuous(limits = c(-12000, 500))

No comments:

Post a Comment