6

I've produced a graph using ggplot and some data that represents the occurrence of specific management actions over the past decade. The graph looks great except for some vertical white spaces that appear within the plotted lines. Is there a way to remove these lines? I'm guessing they are caused by the grouping of my data (group=1). Any help would be appreciated. My code is below.

Also, I've used this stackoverflow question as guidance.

library(scales)

# Build data frame.
a7.data <- data.frame(date = seq(as.Date("2002/01/01"), as.Date("2013/05/01"),     by="day"))
a7.data$year <- as.numeric(format(as.Date(a7.data$date), format="%Y"))
a7.data$month <- as.numeric(format(as.Date(a7.data$date), format="%m"))
a7.data$day <- as.numeric(format(as.Date(a7.data$date), format="%d"))
a7.data$status <- "Yes"
a7.data$filler_value <- 0

# Edit data frame for dates in which the management action was "no".
a7.data$status[a7.data$date >= "2002/01/01" & a7.data$date < "2002/07/01"] <- "No"
a7.data$status[a7.data$date >= "2005/05/20" & a7.data$date < "2005/08/25"] <- "No"
a7.data$status[a7.data$date >= "2005/12/31" & a7.data$date < "2006/04/12"] <- "No"
a7.data$status[a7.data$date >= "2006/11/06" & a7.data$date < "2006/12/31"] <- "No"
a7.data$status[a7.data$date >= "2007/01/31" & a7.data$date < "2007/07/02"] <- "No"
a7.data$status[a7.data$date >= "2008/02/01" & a7.data$date < "2009/08/11"] <- "No"
a7.data$status[a7.data$date >= "2010/02/28" & a7.data$date < "2010/03/15"] <- "No"
a7.data$status[a7.data$date >= "2010/05/09" & a7.data$date < "2010/07/07"] <- "No"

# Create a new column that creates a dummy year with which to plot the data in ggplot using faceting.
a7.data <- transform(a7.data, doy = as.Date(paste(1970, month, day, sep="/")))

# Custom colors.
ccolors <- c("#086CA2", "#FF8B00")

# ggplot code.
bb <- ggplot(a7.data, aes(doy, filler_value)) + 
geom_line(aes(color=status, group=1), size=15, alpha=0.9) + 
scale_x_date(label=date_format("%b"), breaks = "month") + 
xlab("") + ylab("") + facet_grid(year~., scales="free") + 
theme_bw() + theme(axis.text.y=element_blank()) + 
theme(axis.ticks.y=element_blank()) + 
scale_color_manual(values=ccolors, name="Article VII Restrictions?")

# Display plot.
bb
Community
  • 1
  • 1
H2O_Research
  • 165
  • 8
  • hi, are you referring to the white space to the left and right of `Jan` – Ricardo Saporta May 01 '13 at 20:46
  • I should have been more clear. I was referring to the very thin white spaced within the color blocks. SimonO101 solved my problem by recommending I use geom_linerange instead of geom_line to create contiguous color blocks. – H2O_Research May 01 '13 at 23:58

1 Answers1

7

Interesting plot! I think if I understand your problem you either need

  • to add an expand = c(0,0) to scale_x_date to get rid of the space at the boundaries
  • and/or use geom_linerange to make contiguous colour blocks

But you need to specify both a ymin and a ymax value to use geom_linerange. filler_value seems to be a natural choice for ymin so lets make a7.data$filler_value2 <- 1 to be our ymax and use a geom_linerange and include an expand argument:

a7.data$filler_value2 <- 1
bb <- ggplot(a7.data, aes(x = doy)) + 
geom_linerange( aes(ymin = filler_value , ymax = filler_value2 , color=status, group=1), size=15, alpha=0.9) + 
scale_x_date(label=date_format("%b"), breaks = "month" , expand = c(0,0)) + 
xlab("") + ylab("") + facet_grid(year~., scales="free") + 
theme_bw() + theme(axis.text.y=element_blank()) + 
theme(axis.ticks.y=element_blank()) + 
scale_color_manual(values=ccolors, name="Article VII Restrictions?")

# Display plot.
bb

If I make those changes I get a plot that looks like this... enter image description here

Simon O'Hanlon
  • 58,647
  • 14
  • 142
  • 184