#install.packages(c("readr", "dplyr", "ggplot2", "tidyr"))
# Load packages
library(readr)
## Warning: package 'readr' was built under R version 4.4.3
library(dplyr)
## Warning: package 'dplyr' was built under R version 4.4.3
##
## Attaching package: 'dplyr'
## The following objects are masked from 'package:stats':
##
## filter, lag
## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, union
library(ggplot2)
## Warning: package 'ggplot2' was built under R version 4.4.3
library(tidyr)
## Warning: package 'tidyr' was built under R version 4.4.3
#Import Meteorology Data
met_raw <- read_table(
"../data/uni.barentsburg.20107.dat",
col_names = FALSE,
show_col_types = FALSE
)
head(met_raw)
## # A tibble: 6 × 22
## X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13
## <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 20107 1961 1 -1 -1 1 78.1 14.2 -12.7 1013. 1000. 1000. 6
## 2 20107 1961 2 -1 -1 1 78.1 14.2 -17.2 1005. 1000. 1000. 6.4
## 3 20107 1961 3 -1 -1 1 78.1 14.2 -12.5 998. 1000. 1000. 8.2
## 4 20107 1961 4 -1 -1 1 78.1 14.2 -11.4 1013. 1000. 1000. 5.8
## 5 20107 1961 5 -1 -1 1 78.1 14.2 -3.8 1016. 1000. 1000. 8.4
## 6 20107 1961 6 -1 -1 1 78.1 14.2 2.7 1008. 1000. 1000. 8.4
## # ℹ 9 more variables: X14 <dbl>, X15 <dbl>, X16 <dbl>, X17 <dbl>, X18 <dbl>,
## # X19 <dbl>, X20 <dbl>, X21 <dbl>, X22 <chr>
#I only keep important columns
met_data <- met_raw %>%
select(
station = X1,
year = X2,
month = X3,
latitude = X7,
longitude = X8,
temp = X9,
station_name = X22
)
head(met_data)
## # A tibble: 6 × 7
## station year month latitude longitude temp station_name
## <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 20107 1961 1 78.1 14.2 -12.7 barentsburg
## 2 20107 1961 2 78.1 14.2 -17.2 barentsburg
## 3 20107 1961 3 78.1 14.2 -12.5 barentsburg
## 4 20107 1961 4 78.1 14.2 -11.4 barentsburg
## 5 20107 1961 5 78.1 14.2 -3.8 barentsburg
## 6 20107 1961 6 78.1 14.2 2.7 barentsburg
summary(met_data)
## station year month latitude longitude
## Min. :20107 Min. :1961 Min. : 1.00 Min. :78.07 Min. :14.25
## 1st Qu.:20107 1st Qu.:1971 1st Qu.: 3.75 1st Qu.:78.07 1st Qu.:14.25
## Median :20107 Median :1980 Median : 6.50 Median :78.07 Median :14.25
## Mean :20107 Mean :1980 Mean : 6.50 Mean :78.07 Mean :14.25
## 3rd Qu.:20107 3rd Qu.:1990 3rd Qu.: 9.25 3rd Qu.:78.07 3rd Qu.:14.25
## Max. :20107 Max. :2000 Max. :12.00 Max. :78.07 Max. :14.25
## temp station_name
## Min. : -23.70 Length:480
## 1st Qu.: -11.22 Class :character
## Median : -3.70 Mode :character
## Mean : 161.60
## 3rd Qu.: 4.80
## Max. : 999.99
#Clean temperature data
met_clean <- met_data %>%
mutate(
temp = ifelse(temp >= 999, NA, temp),
period = case_when(
year >= 1961 & year <= 1980 ~ "1961-1980",
year >= 1981 & year <= 2000 ~ "1981-2000"
),
month_name = factor(
month,
levels = 1:12,
labels = c("Jan", "Feb", "Mar", "Apr", "May", "Jun",
"Jul", "Aug", "Sep", "Oct", "Nov", "Dec")
)
)
summary(met_clean$temp)
## Min. 1st Qu. Median Mean 3rd Qu. Max. NA's
## -23.700 -12.725 -6.400 -6.083 0.900 7.700 80
sum(is.na(met_clean$temp))
## [1] 80
#Monthly Statistics Table
monthly_stats <- met_clean %>%
group_by(month_name) %>%
summarise(
overall_mean = mean(temp, na.rm = TRUE),
overall_median = median(temp, na.rm = TRUE),
early_mean = mean(temp[period == "1961-1980"], na.rm = TRUE),
early_median = median(temp[period == "1961-1980"], na.rm = TRUE),
late_mean = mean(temp[period == "1981-2000"], na.rm = TRUE),
late_median = median(temp[period == "1981-2000"], na.rm = TRUE),
sd_temp = sd(temp, na.rm = TRUE)
)
monthly_stats
## # A tibble: 12 × 8
## month_name overall_mean overall_median early_mean early_median late_mean
## <fct> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 Jan -14.1 -14.6 -14.5 -15.0 -13.4
## 2 Feb -14.3 -15 -15.2 -15.2 -13.1
## 3 Mar -14.1 -13.8 -15.0 -15.1 -12.8
## 4 Apr -11.3 -11.3 -11.7 -11.6 -10.6
## 5 May -4.08 -3.8 -4.40 -3.9 -3.58
## 6 Jun 1.61 1.5 1.54 1.45 1.72
## 7 Jul 5.52 5.5 5.47 5.45 5.59
## 8 Aug 4.59 4.5 4.42 4.45 4.86
## 9 Sep 0.479 0.5 0.38 0.4 0.631
## 10 Oct -5.07 -4.9 -4.80 -4.1 -5.48
## 11 Nov -9.11 -8.5 -9.53 -9.6 -8.46
## 12 Dec -12.3 -12 -12.2 -11.8 -12.5
## # ℹ 2 more variables: late_median <dbl>, sd_temp <dbl>
#The first plot is monthly mean temperatures for 1961-2000
ggplot(monthly_stats,
aes(x = month_name,
y = overall_mean)) +
geom_col(
fill = "skyblue",
alpha = 0.9
) +
geom_errorbar(
aes(
ymin = overall_mean - sd_temp,
ymax = overall_mean + sd_temp
),
width = 0.3,
linewidth = 0.8
) +
labs(
title = "Mean Monthly Air Temperature at Barentsburg Station (1961–2000)",
subtitle = "Bars show monthly mean temperatures and the error bars show standard deviation",
x = "Month",
y = "Temperature (C)"
) +
theme_gray() +
theme(
plot.title = element_text(size = 14, face = "bold"),
axis.title = element_text(size = 12),
axis.text = element_text(size = 14),
legend.title = element_text(size = 12),
legend.text = element_text(size = 11)
)
#The second plot is Mean vs Median Temperatures
mean_median <- data.frame(
month = rep(monthly_stats$month_name, 2),
temperature = c(
monthly_stats$overall_mean,
monthly_stats$overall_median
),
statistic = c(
rep("Mean", 12),
rep("Median", 12)
)
)
ggplot(mean_median,
aes(
x = month,
y = temperature,
fill = statistic
)) +
geom_col(
position = "dodge",
alpha = 0.9,
width = 0.9
) +
labs(
title = "Monthly Mean and Median Temperatures",
subtitle = "Barentsburg Station, Arctic Russia (1961-2000)",
x = "Month",
y = "Temperature (C)"
) +
scale_fill_manual(
values = c(
"Mean" = "skyblue",
"Median" = "#FF7900"
)
) +
theme_gray() +
theme(
plot.title = element_text(size = 18, face = "bold"),
plot.subtitle = element_text(size = 12),
axis.title = element_text(size = 14),
axis.text = element_text(size = 12),
legend.title = element_text(size = 12),
legend.text = element_text(size = 11)
)
#The third plot will compare Early vs Late period
early_late <- data.frame(
month = rep(monthly_stats$month_name, 2),
temperature = c(
monthly_stats$early_mean,
monthly_stats$late_mean
),
period = c(
rep("1961-1980", 12),
rep("1981-2000", 12)
)
)
ggplot(early_late,
aes(
x = month,
y = temperature,
fill = period
)) +
geom_col(position = "dodge") +
labs(
title = "Comparison of Monthly Mean Temperatures",
subtitle = ("Barentsburg Station, Arctic Russia: 1961-1980 vs 1981-2000"),
x = "Month",
y = "Temperature (C)",
fill = "Period"
) +
scale_fill_manual(
values = c(
"1961-1980" = "navy",
"1981-2000" = "red"
)
) +
theme(
plot.title = element_text(size = 16, face = "bold"),
axis.title = element_text(size = 13),
axis.text = element_text(size = 11),
legend.title = element_text(size = 12),
legend.text = element_text(size = 11)
)
#I made a Monthly Statistics Table to show the overall changes in mean and median.
#the earlier period and late period of mean and median.
final_monthly_table <- monthly_stats %>%
select(
Month = month_name,
`Mean Temp 1961-2000` = overall_mean,
`Median Temp 1961-2000` = overall_median,
`Mean Temp 1961-1980` = early_mean,
`Median Temp 1961-1980` = early_median,
`Mean Temp 1981-2000` = late_mean,
`Median Temp 1981-2000` = late_median
) %>%
mutate(
across(
where(is.numeric),
~ round(.x, 2)
)
)
final_monthly_table
## # A tibble: 12 × 7
## Month `Mean Temp 1961-2000` `Median Temp 1961-2000` `Mean Temp 1961-1980`
## <fct> <dbl> <dbl> <dbl>
## 1 Jan -14.0 -14.6 -14.5
## 2 Feb -14.3 -15 -15.2
## 3 Mar -14.1 -13.8 -15.0
## 4 Apr -11.3 -11.3 -11.7
## 5 May -4.08 -3.8 -4.4
## 6 Jun 1.61 1.5 1.54
## 7 Jul 5.52 5.5 5.47
## 8 Aug 4.59 4.5 4.42
## 9 Sep 0.48 0.5 0.38
## 10 Oct -5.07 -4.9 -4.8
## 11 Nov -9.11 -8.5 -9.53
## 12 Dec -12.3 -12 -12.2
## # ℹ 3 more variables: `Median Temp 1961-1980` <dbl>,
## # `Mean Temp 1981-2000` <dbl>, `Median Temp 1981-2000` <dbl>
write.csv(
final_monthly_table,
"../output/final_monthly_temperature_table.csv",
row.names = FALSE
)
#Lastly, I performed a statistical test
#it will create vectors for the earlier and later climate periods
early_period <- monthly_stats$early_mean
late_period <- monthly_stats$late_mean
#This will conduct paired t-test
t_test_result <- t.test(
early_period,
late_period,
paired = TRUE
)
#Display the results
t_test_result
##
## Paired t-test
##
## data: early_period and late_period
## t = -2.7521, df = 11, p-value = 0.01882
## alternative hypothesis: true mean difference is not equal to 0
## 95 percent confidence interval:
## -1.2463185 -0.1386632
## sample estimates:
## mean difference
## -0.6924908