Time series with ggplot2: Using days and hours from different columns - r

Question

I am trying to plot a time series using ggplot, having the day and time stored in different data frame columns. How can I tell ggplot to take into account both the date and time in the plot?
My data looks like this
Date Hour_min Tair Tflower Tbud
Day1 8:35 24,73 29,79 31,41
Day1 8:36 24,29 29,99 31,82
... .. .. ... ...
Day2 00:00 23,62 30,37 32,59
One can load a small sample of the dataset with this:
#Tagua <- read.table(file = "TIMESERIE_OTO32.txt", header = TRUE,dec = ",")
Tagua <- structure(
list(
Date = structure(c(1L, 1L, 2L, 2L), .Label = c("Day1", "Day2"), class = "factor"),
Hour_min = structure(c(1L, 2L, 1L, 2L), .Label = c("8:35", "8:36"), class = "factor"),
Tair = c(24.73, 24.29, 23.62, 24.29),
Tflower = c(29.79, 29.99, 30.37, 29.99),
Tbud = c(31.41, 31.82, 32.59, 31.82)
),
.Names = c("Date", "Hour_min", "Tair", "Tflower", "Tbud"),
class = "data.frame",
row.names = c(NA, -4L))
Days, hours, and 3 temperature from different parts of the flower.
I have 1400 minutes for 2 days.
I wrote this script:
library(ggplot2)
ggplot(aes(x = (Hour_min), group=1), data = Tagua) +
geom_line(aes(y = Tair, colour = "var1")) +
geom_line(aes(y = Tbud, colour = "var2")) +
geom_line(aes(y = Tflower, colour = "var3"))
The problem is that R plots from 00:00 to 23 (of course), without considering the days.
How can I solve this problem?
If possible, I would like to set the x-axis tick just corresponding to the hour (eg. 2:00, 3:00,...).

Answer 1

This may not be the shortest solution, but you can run it step by step and see how it works.
library(lubridate)
library(dplyr)
library(tidyr)
library(ggplot2)
Tagua <- read.table(file = "TIMESERIE_OTO32.txt", header = TRUE, dec = ",")
Tagua_clean <- Tagua %>%
# Separate hours and minutes:
separate(Hour_min, into = c("Hour", "Minute"), sep = ":") %>%
# Convert Day1 -> 0
# Day2 -> 1
mutate(Day = as.numeric(gsub("Day", "", Date)) - 1) %>%
# Create a Period:
mutate(time_period = period(days = Day, hours = Hour, minutes = Minute)) %>%
# Create a Date, using the beginning of the experiment (if you know it):
mutate(Date = as.POSIXct("2017-01-01") + time_period) %>%
# Option 2: Convert the time period to hours:
mutate(Hours = as.numeric(time_period)/3600) %>%
select(Date, Hours, Tair, Tflower, Tbud)
# Option 1: With real dates:
ggplot(aes(x = Date), data = Tagua_clean) +
geom_line(aes(y = Tair, colour = "var1")) +
geom_line(aes(y = Tbud, colour = "var2"))+
geom_line(aes(y = Tflower, colour = "var3"))
# Option 2: With hours:
ggplot(aes(x = Hours), data = Tagua_clean) +
geom_line(aes(y = Tair, colour = "var1")) +
geom_line(aes(y = Tbud, colour = "var2"))+
geom_line(aes(y = Tflower, colour = "var3"))
Update: Restart the hours to 0 every day. Here we use Dates but we customize how they are shown.
scale_x_datetime has the argument date_labels that can be set to "%H" to show the hour of the day or can be set to "Day %d \n Hour: %H" for a combination of day and hour. See ?strptime for more format options. Another argument that can be used is date_breaks to specify "1 hour" if you want a label every hour.
ggplot(aes(x = Date), data = Tagua_clean) +
geom_line(aes(y = Tair, colour = "var1")) +
geom_line(aes(y = Tbud, colour = "var2"))+
geom_line(aes(y = Tflower, colour = "var3")) +
scale_x_datetime(date_labels = "Day %d \n Hour: %H")