line graph of four variables in r in just one plot

Question

I would like to create a line graph that shows how the trend of five air pollutants were during the years 2009 to 2019.

Year	CO2	NO2	O3	PM2.5
2009	30	18	20	30
2010	32	16	22	20
2011	33	16	24	20
2012	32	15	25	22
2013	34	14	27	24
2014	36	14	28	22
2015	38	13	29	20
2016	39	13	30	18
2017	40	12	32	16
2018	44	13	34	15
2019	45	11	38	14

I gave that code but it is a histogram, i would like to have a line graph were all four are in the same plot.

df %>%
  ggplot(aes(x = Year, y = n, fill = airpollutants)) +
  geom_col() +
  facet_wrap(~Year)  + ggtitle("trend of airpollutants")

I want this output: https://cdn.ablebits.com/_img-blog/line-graph/line-graph-excel.png

score 3 · Answer 1 · answered Jul 26 '22 at 12:13

You could reshape your data from wide to long and colour every airpollutants like this:

df <- read.table(text = "Year   CO2 NO2 O3  PM2.5
2009    30  18  20  30
2010    32  16  22  20
2011    33  16  24  20
2012    32  15  25  22
2013    34  14  27  24
2014    36  14  28  22
2015    38  13  29  20
2016    39  13  30  18
2017    40  12  32  16
2018    44  13  34  15
2019    45  11  38  14
", header = TRUE)

library(ggplot2)
library(dplyr)
library(reshape)
df %>%
  melt(id = "Year") %>%
  mutate(variable = as.factor(variable)) %>%
  ggplot(aes(x = Year, y = value, colour = variable)) +
  geom_line() +
  labs(colour = "airpollutants") +
  ggtitle("trend of airpollutants")

^{Created on 2022-07-26 by the reprex package (v2.0.1)}

score 3 · Answer 2 · 2022-07-26T12:33:37.027

3

Usually you'll want to be in long format when plotting in ggplot2.

One way to draw multiple lines without going long is to map over the columns

ggplot(data = df) + purrr::map2(df[-1], names(df[-1]), \(x,y) geom_line(aes(x = df$Year, y = x, col = y))) +
  labs(x = "Concentration",
       y = "Year",
       col = "Pollutant")

edited Jul 26 '22 at 12:33

answered Jul 26 '22 at 12:26

That is some brutal but effective code. What's \(x,y) doing there? – socialscientist Jul 26 '22 at 12:31
I was just being lazy, see update – Jul 26 '22 at 12:33
Oh, I just meant I don't know what `\(x,y)` does... Is that shorthand for `function(x,y){` in `purrr`? – socialscientist Jul 26 '22 at 12:35
2

Oh yeah, sorry, yeah it's just `function(x,y)` – Jul 26 '22 at 12:36

socialscientist · Accepted Answer · 2022-07-26T12:26:26.617

2

set.seed(123)
library(ggplot2)
library(tidyr)

# Example data
df <- data.frame(year = 2009:2019,
                 CO2 = sample(30:40, 11),
                 NO2 = sample(10:20, 11),
                 O3 = sample(20:30, 11),
                 PM2.5 = sample(15:25, 11))

# Convert to long format
df_long <- pivot_longer(df, 
                        cols = c(CO2, NO2, O3, PM2.5), 
                        values_to = "Concentration",
                        names_to = "Pollutant")


# Plot
ggplot(df_long,
       aes(
         x = year,
         y = Concentration,
         color = Pollutant,
         linetype = Pollutant
       )) +
  geom_line(size = 0.7) +
  ggtitle("Trend of Airpollutants") +
  xlab("Year") +
  ylab("Concentration") +
  scale_x_continuous(breaks = seq(2009, 2019, by = 1), limits = c(2009,2019)) +
  theme_minimal()

edited Jul 26 '22 at 12:26

answered Jul 26 '22 at 12:15

socialscientist

3,759
5
23
58

1

how can I change it that the lines are not dotted, just line? – Jul 26 '22 at 12:30
1

Remove the `linetype = Pollutant` line! That's letting your groups (pollutants) affect the line types as well as the colors of the lines. Note that it is often recommended to try to use different line types because the people who look at your figure may be colorblind or someone might print off your figure in black and white -- changing the line type helps a little with making your work more accessible. – socialscientist Jul 26 '22 at 12:31
2

perfect many thanks. I am very new to R – Jul 26 '22 at 12:32

line graph of four variables in r in just one plot

3 Answers3

Linked