Is there a way to change the scale limits of the second y-axis in ggplot2?

January 25, 2024

Suppose that I have two variables with different scales for example;

df = data.table::data.table(
  date = seq(from = lubridate::ymd("2010-01-01"), 
             to = lubridate::ymd("2020-01-01"), length = 100),
  var1 = rpois(n = 100, lambda = 84),
  var2 = rpois(n = 100, lambda = 300)
)

If I decide to use a double y-axis plot, using the code;

library(ggplot)

df |> 
  ggplot(aes(x = date)) +
  geom_line(aes(y = var1), col = "royalblue") +
  geom_line(aes(y = var2), col = "red") +
  scale_y_continuous(
    name = "var1",
    sec.axis = sec_axis(~., name = "var2")
  ) +
  theme_light()

The results are as follows;

How can I adjust the limits of the second y-axis to, for instance, start at 250 and end at 400? This would bring the lines closer together, allowing for a clearer view of the trend.

>Solution :

You can transform the data when you plot and then undo the transformation in the call to sec_axis(). Here, I just subtracted 200 from the data values and then add 200 to the labels to reflect the change:

library(ggplot2)
library(data.table)
library(lubridate)
df = data.table(
  date = seq(from = ymd("2010-01-01"), to = ymd("2020-01-01"), length = 100),
  var1 = rpois(n = 100, lambda = 84),
  var2 = rpois(n = 100, lambda = 300)
)

df |> 
  ggplot(aes(x = date)) +
  geom_line(aes(y = var1), col = "royalblue") +
  geom_line(aes(y = var2-200), col = "red") +
  scale_y_continuous(
    name = "var1",
    sec.axis = sec_axis(~.+200, name = "var2")
  ) +
  theme_light()