Search code examples
rggplot2fill

R - ggplot2 - geom_area - Remove break in color fill under a geom_line


For the data:

   year     N cumulative time.period
1  2000   100        100        blue
2  2001   100        200        blue
3  2002   100        300        blue
4  2003   100        400        blue
5  2004   100        500        blue
6  2005   100        600         red
7  2006   100        700         red
8  2007   100        800         red
9  2008   100        900         red
10 2009   100       1000         red

I want to first plot the line for "cumulative" and then color it using the "time.period" variable. After that I want to fill in the area below the line using the same "time.period" variable. Here's my code for generating that plot:

library(dplyr)
library(ggplot2)

year<-2000:2009
N <- rep(100, 10)

DF<-as.data.frame(cbind(year, N))

DF <- DF %>%
 mutate(cumulative = cumsum(N)) %>%
 mutate(time.period = ifelse(year<2005, "blue", "red"))

 ggplot(DF, aes(x=year, y=cumulative)) + 
  geom_line(color=DF$time.period) +
  ylab("Cumulative count") +
  geom_area(aes(fill=DF$time.period)) +
  scale_fill_manual(values=c("blue", "red"))

This generates the following plot:

Plot

I then get a gap in the fill for year 2004. How can I adjust my code so that there are no breaks in the coloring for geom_area fill?


Solution

  • Since you are working with count data, I think it is reasonable to visualize your data using a bar chart. The figure will need to be cleaned up a bit, but this eliminates the broken part of the visualization.

    DF <- structure(list(year = 2000:2009, N = c(100L, 100L, 100L, 100L, 
    100L, 100L, 100L, 100L, 100L, 100L), cumulative = c(100L, 200L, 
    300L, 400L, 500L, 600L, 700L, 800L, 900L, 1000L), time.period =     
    structure(c(1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L), 
    .Label = c("blue", "red"), class = "factor")), 
    .Names = c("year", "N", "cumulative", 
    "time.period"), class = "data.frame", row.names = c("1", "2", 
    "3", "4", "5", "6", "7", "8", "9", "10"))
    
    ggplot(DF, aes(year, cumulative, fill = time.period)) +     
    geom_bar(stat="identity", position = "identity")
    

    enter image description here