How to plot average temperature and variation?

Question

I have hourly temperature data from 1970 to 2021. I would like to summarize the dataset into a graph in ggplot showing the mean temperature of each month and year. And I also want to show the average temperature and temperature volatility increased. And I would like to show average temperature in the '2010s is higher than in the '1990s.

Here is the code of temperature density plot in 2010s and 1990s. But I don't know how to set the x axis.

data_re <- data 
data_re$Year910 <- ifelse(data$Year %in% c(1990:1999), "1990s", 
                             ifelse(data$Year %in% c(2010:2020), "2010s",NA))
data_re1 <- subset(data_re, Year910 %in% c("1990s","2010s"))
data_re1$Year910 <- factor(data_re1$Year910)
ggplot(data_re1, aes(x=, color=Year910)) +
xlab('Temperature')+
geom_density()

Here is the data file. https://docs.google.com/spreadsheets/d/1HwPFJ1wKMMr0845Et60tV36WIWXM66Ig/edit?usp=sharing&ouid=111186166036061320361&rtpof=true&sd=true

Any advice on how to best go about this would be greatly appreciated.

Welcome to stack overflow. It looks like you aren't sure about what kind of plot, exactly, you want, which isn't a programming question. A good guess for a density plot would be x=Temp. Your second question is about comparing means and significance which is a big topic, see t-tests, group_by, and summarize. — PeterK
– PeterK, Commented May 5, 2022 at 13:20

Deepansh Arora · Accepted Answer · 2022-05-05 13:53:28Z

2

I used lubridate library to convert the character to Date.

df = df %>%
  pivot_longer(!c(Date,Month,Year), names_to = "HourNo", values_to = "Temp")

df$Date = as.Date(df$Date, format = "%Y-%m-%d")
df = na.omit(df)
df$Abbr = paste(month.abb[month(df$Date)], "-", year(df$Date), sep="")

df_Aggr = df %>%
  group_by(Abbr) %>%
  summarise(Avg = mean(Temp),
            Begin = min(Date))
ggplot(df_Aggr, aes(x =Begin, y=Avg))+geom_line(color="blue", size=1)+
  scale_x_date(date_breaks = "24 month", date_labels = "%b-%Y")+
  theme(axis.text.x = element_text(vjust = 0.8,
    angle = 35, hjust = 0.9))+ggtitle("Average Temperature")

answered May 5, 2022 at 13:53

Deepansh Arora

7521 gold badge4 silver badges20 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Allan Cameron · Accepted Answer · 2022-05-05 13:48:20Z

I would first of all ensure your data is in the correct format after importing it. I am assuming you have already loaded your data frame from Excel and named it df:

df <- df[1:624,]
df[2:28] <- lapply(df[2:28], as.numeric)
df$Date <- as.POSIXct(df$Date)

Now we can load a couple of useful packages:

library(tidyverse)
library(geomtextpath)

Then we could summarize and plot with something like the following:

df %>%
  rowwise() %>%
  mutate(Temp = sum(across(starts_with('Hour')), na.rm = TRUE) / 24) %>%
  select(-starts_with('Hour')) %>%
  filter(Date > as.POSIXct('1989-12-31')) %>% 
  group_by(Year) %>%
  summarize(Temp = mean(Temp), Date = median(Date)) %>% {
  ggplot(., aes(Date, Temp)) +
  geom_line(color = 'gray') +
  geom_point(color = 'gray75') +
  geom_textsegment(aes(x = as.POSIXct('1990-01-01'), 
                   xend = as.POSIXct('1999-12-31'),
                   y = mean(Temp), yend = mean(Temp), color = '1990s',
                   label = '1990s'), vjust = -0.2, size = 6,
               data = .[.$Date < as.POSIXct('2000-01-01'),], linetype = 2) +
  geom_textsegment(aes(x = as.POSIXct('2000-01-01'), 
                   xend = as.POSIXct('2009-12-31'),
                   y = mean(Temp), yend = mean(Temp), color = '2000s',
                   label = '2000s'), vjust = -0.2, size = 6,
               data = .[.$Date < as.POSIXct('2010-01-01') &
                        .$Date > as.POSIXct('1999-12-31'),], linetype = 2) +
  geom_textsegment(aes(x = as.POSIXct('2010-01-01'), 
                   xend = as.POSIXct('2019-12-31'),
                   y = mean(Temp), yend = mean(Temp), color = '2010s',
                   label = '2010s'), vjust = -0.2, size = 6,
               data = .[.$Date < as.POSIXct('2020-01-01') &
                        .$Date > as.POSIXct('2009-12-31'),], linetype = 2) +
  theme_light(base_size = 16) +
  scale_color_brewer(palette = 'Set1') +
  theme(legend.position = 'none') +
  labs(title = 'Annual mean temperature')

Note that I have used an annual summary of the temperature here. If you use monthly temperatures, the range of temperatures becomes much larger and the message of the plot becomes much weaker:

df %>%
  rowwise() %>%
  mutate(Temp = sum(across(starts_with('Hour')), na.rm = TRUE) / 24) %>%
  select(-starts_with('Hour')) %>%
  filter(Date > as.POSIXct('1989-12-31')) %>% {
  ggplot(., aes(Date, Temp)) +
  geom_line(color = 'gray') +
  geom_point(color = 'gray75') +
  geom_textsegment(aes(x = as.POSIXct('1990-01-01'), 
                   xend = as.POSIXct('1999-12-31'),
                   y = mean(Temp), yend = mean(Temp), color = '1990s',
                   label = '1990s'), vjust = -0.2, size = 6,
               data = .[.$Date < as.POSIXct('2000-01-01'),], linetype = 2) +
  geom_textsegment(aes(x = as.POSIXct('2000-01-01'), 
                   xend = as.POSIXct('2009-12-31'),
                   y = mean(Temp), yend = mean(Temp), color = '2000s',
                   label = '2000s'), vjust = -0.2, size = 6,
               data = .[.$Date < as.POSIXct('2010-01-01') &
                        .$Date > as.POSIXct('1999-12-31'),], linetype = 2) +
  geom_textsegment(aes(x = as.POSIXct('2010-01-01'), 
                   xend = as.POSIXct('2019-12-31'),
                   y = mean(Temp), yend = mean(Temp), color = '2010s',
                   label = '2010s'), vjust = -0.2, size = 6,
               data = .[.$Date < as.POSIXct('2020-01-01') &
                        .$Date > as.POSIXct('2009-12-31'),], linetype = 2) +
  theme_light(base_size = 16) +
  scale_color_brewer(palette = 'Set1') +
  theme(legend.position = 'none') +
  labs(title = 'Annual mean temperature')
  }

I liked your first graph. There is always so much to learn everyday on Stackoverflow.
OMG Thank you for solving it. It didn't work out although I spent a long time trying it. Thank you very much!!
But I have another question: how to show the variation of temperature is also gradually increasing? Thank you so much!!

Collectives™ on Stack Overflow

How to plot average temperature and variation?

2 Answers 2

Comments

3 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

3 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related