How to fill geom_polygon with different colors above and below y = 0 (or any other value)?

by

Get the indices where the y value of two consecutive time steps have different sign. Use linear interpolation between these points to generate new x values where y is zero.

First, a smaller example to make it easier to get a feeling for the linear interpolation and which points are added to the original data:

# original data
d <- data.frame(x = 1:6,
                y = c(-1, 2, 1, 2, -1, 1))

# coerce to data.table
library(data.table)
setDT(d)

# make sure data is ordered by x
setorder(d, x)

# add a grouping variable
# only to keep track of original and interpolated points in this example
d[ , g := "orig"]

# interpolation
d2 = d[ , {
  ix = .I[c(FALSE, abs(diff(sign(d$y))) == 2)]
  if(length(ix)){
    pred_x = sapply(ix, function(i) approx(x = y[c(i-1, i)], y = x[c(i-1, i)], xout = 0)$y)
    rbindlist(.(.SD, data.table(x = pred_x, y = 0, g = "new")))} else .SD
}]

d2   
#           x  y  grp
# 1  1.000000 -1 orig
# 2  2.000000  2 orig
# 3  3.000000  1 orig
# 4  4.000000  2 orig
# 5  5.000000 -1 orig
# 6  6.000000  1 orig
# 13 1.333333  0  new
# 11 4.666667  0  new
# 12 5.500000  0  new

Plot with original and new points differentiated by color:

ggplot(data = d2, aes(x = x, y = y)) +
  geom_area(data = d2[y <= 0], fill = "red", alpha = 0.2) +
  geom_area(data = d2[y >= 0], fill = "blue", alpha = 0.2) +
  geom_point(aes(color = g), size = 4) +
  scale_color_manual(values = c("red", "black")) +
  theme_bw()

enter image description here

Apply on OP’s data:

d = as.data.table(orig)
# setorder(d, year)

d2 = d[ , {
  ix = .I[c(FALSE, abs(diff(sign(d$afw))) == 2)]
  if(length(ix)){
    pred_yr = sapply(ix, function(i) approx(afw[c(i-1, i)], year[c(i-1, i)], xout = 0)$y)
    rbindlist(.(.SD, data.table(year = pred_yr, afw = 0)))} else .SD}]

ggplot(data = d2, aes(x = year, y = afw)) +
  geom_area(data = d2[afw <= 0], fill = "red") +
  geom_area(data = d2[afw >= 0], fill = "blue") +
  theme_bw()

enter image description here

In reply to @Jason Whythe’s comment, the method above can be modified to account for grouped data. The interpolation is made within each group, and the plot is facetted by group:

# data grouped by 'id' 
d = data.table(
  id = rep(c("a", "b", "c"), c(6, 5, 4)),
  x = as.numeric(c(1:6, 1:5, 1:4)),
  y = c(-1, 2, 1, 2, -1, 1,
        0, -2, 0, -1, -2, 
        2, 1, -1, 1.5))

# again, this variable is just added for illustration 
d[ , g := "orig"]

d2 = d[ , {
  ix = .I[c(FALSE, abs(diff(sign(.SD$y))) == 2)]
  if(length(ix)){
    pred_x = sapply(ix, function(i) approx(x = d$y[c(i-1, i)], y = d$x[c(i-1, i)], xout = 0)$y)
    rbindlist(.(.SD, data.table(x = pred_x, y = 0, g = "new")))} else .SD
}, by = id]

ggplot(data = d2, aes(x = x, y = y)) +
  facet_wrap(~ id) +
  geom_area(data = d2[y <= 0], fill = "red", alpha = 0.2) +
  geom_area(data = d2[y >= 0], fill = "blue", alpha = 0.2) +
  geom_point(aes(color = g), size = 4) +
  scale_color_manual(values = c("red", "black")) +
  theme_bw()

enter image description here

For an alternative base solution adapted from @kohske‘s answer here (credits to him), see previous edits.

More Related Contents:

  1. How to fill with different colors between two lines? (originally: fill geom_polygon with different colors above and below y = 0 (or any other value)?)
  2. Manually setting group colors for ggplot2
  3. Plot with conditional colors based on values in R [duplicate]
  4. Change bar plot colour in geom_bar with ggplot2 in r
  5. Create discrete color bar with varying interval widths and no spacing between legend levels
  6. asymmetric color distribution in scale_gradient2?
  7. Fill and border colour in geom_point (scale_colour_manual) in ggplot
  8. How do I change the color value of just one value in ggplot2’s scale_fill_brewer?
  9. Stacked barplot with colour gradients for each bar
  10. ggplot2: How to specify multiple fill colors for points that are connected by lines of different colors
  11. Is it possible to define the “mid” range in scale_fill_gradient2()?
  12. Different colours of geom_line above and below a specific value
  13. Increase resolution of color scale for values close to zero
  14. Creating a density histogram in ggplot2?
  15. Remove lines from color and fill legends
  16. using two scale colour gradients on one ggplot [duplicate]
  17. ggplot: colour points by groups based on user defined colours
  18. Consistent color scale and legend between plots when not all levels of a grouping variable are present in the data
  19. Different colour palettes for two different colour aesthetic mappings in ggplot2
  20. Add color to boxplot – “Continuous value supplied to discrete scale” error
  21. ggplot2: Fix colors to factor levels
  22. R – ggplot line color (using geom_line) doesn’t change
  23. Creating a vertical color gradient for a geom_bar plot
  24. ggplot scale color gradient to range outside of data range
  25. How to use an image as a point in ggplot?
  26. How to put labels over geom_bar in R with ggplot2
  27. Displaying a greater than or equal sign
  28. Modifying fonts in ggplot2
  29. How do I loop through column names and make a ggplot scatteplot for each one
  30. How to use superscript in axis labels with ggplot2

Leave a Comment