Analysis with T&H data

Simple stat and plots

Data and Packages

file <- read.csv("tempch4.txt")
library(tidyverse)

── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
✔ dplyr     1.1.4     ✔ readr     2.1.5
✔ forcats   1.0.0     ✔ stringr   1.5.1
✔ ggplot2   3.5.1     ✔ tibble    3.2.1
✔ lubridate 1.9.3     ✔ tidyr     1.3.0
✔ purrr     1.0.2     
── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
✖ dplyr::filter() masks stats::filter()
✖ dplyr::lag()    masks stats::lag()
ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors

Take a glimpse

glimpse(file)

Rows: 24,394
Columns: 3
$ datetime <chr> "2016-02-03 11:00:00", "2016-02-03 12:00:00", "2016-02-03 13:…
$ temp_wew <dbl> 19.5, 20.5, 20.5, 20.0, 19.5, 18.5, 18.0, 17.5, 17.5, 17.0, 1…
$ rh       <dbl> 60.1, 57.6, 55.0, 53.9, 53.8, 55.0, 59.6, 61.5, 66.3, 66.9, 6…

Split the datetime column to filter with different through time variable

data2 <- separate(file, col = datetime, into = c("Date", "Time"), sep = " ")
data <- separate(data2, col = Date, into = c("Year", "Month", "Day"), sep = "-")

Basic descriptive statistic

data |> 
summary()

     Year              Month               Day                Time          
 Length:24394       Length:24394       Length:24394       Length:24394      
 Class :character   Class :character   Class :character   Class :character  
 Mode  :character   Mode  :character   Mode  :character   Mode  :character  
                                                                            
                                                                            
                                                                            
                                                                            
    temp_wew           rh       
 Min.   :15.00   Min.   :30.40  
 1st Qu.:20.00   1st Qu.:70.50  
 Median :21.50   Median :85.00  
 Mean   :22.52   Mean   :81.07  
 3rd Qu.:25.50   3rd Qu.:94.20  
 Max.   :32.50   Max.   :99.90  
                 NA's   :1583   

Generating the variance and Std

data |> 
  summarise(VarT = var(temp_wew), VarH = var(rh, na.rm = TRUE), StdT = sd(temp_wew), StdH = sd(rh, na.rm = TRUE))

      VarT     VarH     StdT     StdH
1 11.76777 253.5742 3.430418 15.92401

file["month"] <- data[["Month"]]

Distribution of the T and H

# Distributio of Temperature
Temperature <- file$temp_wew
hist(Temperature)

Temperature distribution is normally distributed, but there is appearance of almost two peaks.

# Distributio of Humidity
rtv_humidity <- file$rh
hist(rtv_humidity)

Humidity distribution is not normally distributed over the year. higher humidity is prevalent

Temperature Distribuion indicating different month of the year

file |> 
  ggplot(aes(x= datetime, y = temp_wew, colour = month)) +
  geom_point()+
  geom_hline(yintercept = 23)

The Colors represents different months, starting from Feb to Dec. There is trend showing a bell distribution. However, some months (summer) appear to experience higher temperature above 23, while some are below.

For better visualization The monthly average temperature over the year, signify a bell shape.

  data |> 
  group_by(Month) |> 
  summarise(AverageT = mean(temp_wew), AverageH = mean(rh, na.rm = TRUE)) |> 
    ggplot(aes(x = Month,y =AverageT, group = 1)) +
  geom_line(colour = "blue") +
  geom_point() 

BoxPlot emphasis on the bell shape and the means of monthly temperature. There is very few data available for Dec

data |> 
  ggplot(aes(y = temp_wew, x = Month))+
  geom_boxplot()

Humidty Distribuion indicating different month of the year

file |> 
  ggplot(aes(x= datetime, y = rh, colour = month)) +
  geom_point()

Warning: Removed 1583 rows containing missing values or values outside the scale range
(`geom_point()`).

For Humidity, the pattern is different.. higher humidity towards the end of the year and higher variance at the mid months.

  data |> 
  group_by(Month) |> 
  summarise(AverageT = mean(temp_wew), AverageH = mean(rh, na.rm = TRUE)) |> 
    ggplot(aes(x = Month,y =AverageH, group = 1)) +
  geom_line(colour = "green") +
  geom_point() 

Plotting the mean of the distribution, it shows increasing trend of humidity.

data |> 
  ggplot(aes(y = rh, x = Month))+
  geom_boxplot()

Warning: Removed 1583 rows containing non-finite outside the scale range
(`stat_boxplot()`).

The box_plot emphasis the increasing trends of the rh mean over the year.

Distribution during the day (Hourly)

The average hourly temperature of the day, shows rising and falling pattern Temperature decreasing from early hour(00:00) and to lowest temperature at 06:00 and beginning to rise from that hour to the highest temperature at region between 16-19:00 and after which it started to decrease again.

data |> 
  group_by(Time) |> 
  summarise(AverageT = mean(temp_wew), AverageH = mean(rh, na.rm = TRUE)) |> 
  ggplot(aes(x = Time,y =AverageT, group = 1)) +
  geom_line(colour = "blue") +
  geom_point() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

Filtering through different months to see changes in the distribution, the pattern still follow a similar trend.

data |> 
  filter(Month == "07") |> 
  group_by(Time) |> 
  summarise(AverageT = mean(temp_wew), AverageH = mean(rh, na.rm = TRUE)) |> 
  ggplot(aes(x = Time,y =AverageT, group = 1)) +
  geom_line(colour = "blue") +
  geom_point() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

Hour humidity districution of the day reveal a U shape. with the highest rh at around 6:00 and lowest around 16:00 of the day. This is opposite to temperature distributuion

data |> 
  group_by(Time) |> 
  summarise(AverageT = mean(temp_wew), AverageH = mean(rh, na.rm = TRUE)) |> 
  ggplot(aes(x = Time,y =AverageH, group = 1)) +
  geom_line(colour = "red") +
  geom_point() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

Filtering through the months, The distribution is more stable compare to the temperature pattern of the year.

data |> 
  filter(Month == "09") |> 
  group_by(Time) |> 
  summarise(AverageT = mean(temp_wew), AverageH = mean(rh, na.rm = TRUE)) |> 
  ggplot(aes(x = Time,y =AverageH, group = 1)) +
  geom_line(colour = "red") +
  geom_point() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

Daily Max and Min distribution

The above graph represent, daily max and minimum temperature.Both follow the same trends. Red points are the maximum temperature of the day and the blue points are the minimum

  data2 |> 
  group_by(Date) |> 
  summarise(Tmax = max(temp_wew), Tmin = min(temp_wew)) |> 
  ggplot() +
  geom_point(aes(x= Date,y = Tmax), color = "red") +
  geom_point(aes(x= Date,y = Tmin), color = "blue") 

The above graph represent, daily max and minimum humidity. No interesting pattern, months between the year experience varying rh, low and high, unlike the extreme months with higher rh. Red points are the maximum temperature of the day and the blue points are the minimum

data2 |> 
  group_by(Date) |> 
  summarise(Hmax = max(rh), Hmin = min(rh)) |> 
  ggplot() +
  geom_point(aes(x= Date,y = Hmax), color = "red") +
  geom_point(aes(x= Date,y = Hmin, group = 1), color = "blue") 

Warning: Removed 71 rows containing missing values or values outside the scale range
(`geom_point()`).
Removed 71 rows containing missing values or values outside the scale range
(`geom_point()`).

Daily change (Hmax-Hmin) in temperature over the period.

data2 |> 
  group_by(Date) |> 
  summarise(ddT = max(temp_wew)-min(temp_wew), ddH = max(rh)-min(rh)) |> 
  ggplot(aes(x= Date, y = ddT, group = 1)) +
  geom_point()

Daily change (Tmax-Tmin) in humidity over the period. The pattern is like a bell-shape, this explain the differences in the rh variation, with higher var situated at the middle months showing a higher daily rh changes, than the left and right rea/extreme months

data2 |> 
  group_by(Date) |> 
  summarise(ddT = max(temp_wew)-min(temp_wew), ddH = max(rh)-min(rh)) |> 
  ggplot(aes(x= Date, y = ddH, group = 1)) +
  geom_point()

Warning: Removed 71 rows containing missing values or values outside the scale range
(`geom_point()`).