creating a common legend for bar plot and line

Question

I'm very new to R so I apologise in advance if this is a very basic question.

I'm trying to plot a graph showing discharge and suspended sediment load (SSL). However, I want to make it clear that the bar plot represents discharge, and the line graph represents SSL. I had two ideas:

color the label for discharge and SSL to correspond to the bar graph and line graph respectively, so the reader intuitively knows which one belongs to which. but ggplot2 doesn't allow me to do this because it'll colour both y axes the same color.
build a legend that indicates clearly that the red line belongs to SSL, blue box plot belongs to discharge. There is a post here which does something similar but I can't seem to achieve it. I would greatly appreciate it if someone could help me out.

This is what my graph looks like now.

This is my script:

library(ggplot2)
library(gridExtra)
library(RColorBrewer)
library(tibble)

P_Discharge <- Pyay$Mean.monthly.discharge
P_MaxTemp <- Pyay$Mean.monthly.max.temperature
P_MinTemp <- Pyay$Mean.monthly.minimum.temperature
P_Rain <- Pyay$Max.monthly.rainfall
P_SSL <- Pyay$Mean.suspended.sediment.load

#reorderingthemonths
Pyay$Month <- factor(Pyay$Month, 
                     levels=c("Jan", "Feb", "Mar", "Apr", "May", "Jun", 
                              "Jul", "Aug", "Sep", "Oct", "Nov", "Dec"))
#PlottingdischargeandSSL
Pgraph1 <- ggplot(Pyay, aes(x=Month, group=2))
Pgraph1 <- Pgraph1 + geom_bar(aes(y=P_Discharge), stat="identity", fill="blue")
Pgraph1 <- Pgraph1 + geom_line(aes(y=P_SSL), colour="red", size=1) +
  labs(y=expression(Q/(m^{3}))) +
  labs(x="")

#addingsecondaxis
Pgraph1 <- Pgraph1 + scale_y_continuous(sec.axis=sec_axis(~., name=expression(
  Suspended~sediment~load~(10^{6}~t))))

#colouringaxistitles
Pgraph1 <- Pgraph1 + theme(axis.title.x=element_blank(), 
                           axis.title.y=element_text(size=14), 
                           axis.text.x=element_text(size=14))

Pgraph1

Data

Pyay <- tibble::tribble(
              ~Month, ~Mean.monthly.discharge, ~Mean.monthly.max.temperature, ~Mean.suspended.sediment.load, ~Max.monthly.rainfall, ~Mean.monthly.minimum.temperature,
               "Jan",                   8.528,                          32.2,                         3.407,                   1.5,                              16.2,
               "Feb",                   6.316,                          35.1,                         2.319,                   0.9,                              17.8,
               "Mar",                       7,                          37.6,                         2.587,                   5.1,                              21.2,
               "Apr",                   8.635,                          38.7,                         3.573,                  27.3,                              24.7,
               "May",                  12.184,                            36,                         5.785,                 145.1,                              25.6,
               "Jun",                  30.414,                          31.9,                        21.811,                 234.8,                              24.8,
               "Jul",                  70.753,                            31,                        70.175,                   198,                              24.8,
               "Aug",                  79.255,                            31,                        81.873,                 227.5,                              24.7,
               "Sep",                  67.079,                          32.3,                        65.798,                 205.7,                              24.6,
               "Oct",                  53.677,                          33.5,                        47.404,                   124,                              24.2,
               "Nov",                  22.937,                          32.7,                        14.468,                    56,                              21.7,
               "Dec",                  12.409,                          31.5,                         5.842,                   1.5,                              18.1
              )

Welcome to Stack Overflow! Could you make your problem reproducible by sharing a sample of your data so others can help (please do not use `str()`, `head()` or screenshot)? You can use the [`reprex`](https://reprex.tidyverse.org/articles/articles/magic-reprex.html) and [`datapasta`](https://cran.r-project.org/web/packages/datapasta/vignettes/how-to-datapasta.html) packages to assist you with that. See also [Help me Help you](https://speakerdeck.com/jennybc/reprex-help-me-help-you?slide=5) & [How to make a great R reproducible example?](https://stackoverflow.com/q/5963269) — Tung, Mar 20 '19 at 05:41
This might help too https://stackoverflow.com/a/51458034/786542 — Tung, Mar 20 '19 at 05:55

score 4 · Accepted Answer · answered Mar 20 '19 at 06:48

4

If you put the color and fill arguments inside the aes() you'll get a legend. With scale_fill_manual we change the bars to blue. setting the color and fill labs() to "" removes them.

##  Plotting discharge and SSL
Pgraph1 <- ggplot(Pyay, aes(x=Month, group = 2))
Pgraph1 <- Pgraph1 + geom_bar(aes(y=P_Discharge, fill = "discharge"), stat="identity")
Pgraph1 <- Pgraph1 + geom_line(aes(y=P_SSL, colour = "SSL"), size=1)+ labs(y=expression(Q/(m^{3}))) + labs(x=" ") 
Pgraph1 <- Pgraph1 + scale_fill_manual(values = c("discharge" = "blue")) + labs(color = "", fill = "")

#adding second axis 
Pgraph1 <- Pgraph1 + scale_y_continuous(sec.axis = sec_axis(~.,name = expression(Suspended~sediment~load~(10^{6}~t))))

#colouring axis titles
Pgraph1 <- Pgraph1 + theme(
  axis.title.x = element_blank(),
  axis.title.y = element_text(size=14),
  axis.text.x = element_text(size=14)
)
Pgraph1

answered Mar 20 '19 at 06:48

f.lechleitner

3,554
1
17
35

1

Thank you so much for the clear explanation and solution. I really appreciate it. I do have one more question - is there a way I can make the SSL and discharge legend entries to be closer to each other? – catlovingtaco Mar 20 '19 at 06:53
Yes, add `+ theme(legend.spacing.y = unit(-0.2, "cm"))` to your plot! Alternative solutions can be found here: https://stackoverflow.com/questions/14103920/smaller-gap-between-two-legends-in-one-plot-e-g-color-and-size-scale – f.lechleitner Mar 20 '19 at 07:04
1

Got it. Thank you Brett :) – catlovingtaco Mar 20 '19 at 07:12

jay.sf · Answer 2 · 2019-03-20T10:34:03.137

You could also consider base R plots, which I find a little more straightforward.

par(mar=c(5, 5, 4, 5) + 0.1)                                   # adjust plot margins
b <- barplot(Pyay$Mean.monthly.discharge, col="blue",          # plots and saves x-coordinates
             ylim=c(0, 90),
             ylab=expression(Q/(m^{3})))
lines(b, Pyay$Mean.suspended.sediment.load, col="red", lwd=2)  # use x-coordinates here
axis(1, b, labels=Pyay$Month)
axis(4, seq(0, 90, 20), labels=, seq(0, 90, 20))
mtext(expression(Suspended~sediment~load~(10^{6}~t)), 4, 3)
legend("topleft", legend=c("discharge", "SSL"), pch=c(15, NA),
       pt.cex=2, lty=c(0, 1), col=c("blue", "red"))
box()

Gives

Data

Pyay <- structure(list(Month = c("Jan", "Feb", "Mar", "Apr", "May", "Jun", 
"Jul", "Aug", "Sep", "Oct", "Nov", "Dec"), Mean.monthly.discharge = c(8.528, 
6.316, 7, 8.635, 12.184, 30.414, 70.753, 79.255, 67.079, 53.677, 
22.937, 12.409), Mean.monthly.max.temperature = c(32.2, 35.1, 
37.6, 38.7, 36, 31.9, 31, 31, 32.3, 33.5, 32.7, 31.5), Mean.suspended.sediment.load = c(3.407, 
2.319, 2.587, 3.573, 5.785, 21.811, 70.175, 81.873, 65.798, 47.404, 
14.468, 5.842), Max.monthly.rainfall = c(1.5, 0.9, 5.1, 27.3, 
145.1, 234.8, 198, 227.5, 205.7, 124, 56, 1.5), Mean.monthly.minimum.temperature = c(16.2, 
17.8, 21.2, 24.7, 25.6, 24.8, 24.8, 24.7, 24.6, 24.2, 21.7, 18.1
)), row.names = c(NA, -12L), class = "data.frame")

creating a common legend for bar plot and line

2 Answers2

Linked