Solution_Scatterplot.R

library(tidyverse)
library(knitr)
# read csv data, Note: Semicolon seperated CSVs can be loaded by function 'read_delim()'
airquality <- read_delim("data/AirQualityUpperAut.csv", delim = ";")
# half-hourly temperature measurement of station S108 to data frame
temp_tab <- airquality %>%
  dplyr::filter(component == "TEMP" & meantype == "HMW" & station == "S108")
# half-hourly relative humidity measurement of station S108 to data frame
humidity_tab <- airquality %>%
  dplyr::filter(component == "RF" & meantype == "HMW" & station == "S108")
# join humidity and temperature tables by common field 'time'
temp_tab %>%
  dplyr::inner_join(
    # right table
    humidity_tab,
    # columns to match
    by = c("time" = "time")
  ) %>%
  dplyr::select(time, value.x, value.y) %>% # select relevant columns from joined table

  # create plot
  ggplot2::ggplot(
    .,
    aes(
      x = value.x,
      y = value.y
    )
  ) +
  labs(
    x = "air temperature [°C]",
    y = "relative humidity [%]"
  ) +
  ggplot2::geom_point(color = "blue") + # define geometry scatterplot, with point color blue
  ggplot2::geom_smooth(method = lm, color = "red", fill = "#69b3a2", se = TRUE) + # with linear trend and confidence interval
  ggplot2::theme_minimal()