# Step 1: Combine Date and Time columns into a DateTime column
PortTalbotAQ$DateTime <- as.POSIXct(paste(PortTalbotAQ$Date, PortTalbotAQ$Time), format="%Y-%m-%d %H:%M:%S")

# Step 2: Filter the data for the three periods (November 1st to December 31st)
data_2016 <- subset(PortTalbotAQ, DateTime >= "2016-11-01" & DateTime <= "2016-12-31" & format(DateTime, "%H") == "17")
data_2018 <- subset(PortTalbotAQ, DateTime >= "2018-11-01" & DateTime <= "2018-12-31" & format(DateTime, "%H") == "17")
data_2020 <- subset(PortTalbotAQ, DateTime >= "2020-11-01" & DateTime <= "2020-12-31" & format(DateTime, "%H") == "17")

# Step 3: Combine the data for the three years into one dataframe
data_combined <- rbind(data_2016, data_2018, data_2020)
data_combined$Year <- factor(c(rep("2016", nrow(data_2016)), rep("2018", nrow(data_2018)), rep("2020", nrow(data_2020))))

# Step 4: Perform the ANOVA test to compare NO2 levels across the three years
anova_result <- aov(NO2 ~ Year, data = data_combined)
summary(anova_result)

# Step 5: If significant differences are found, perform post-hoc test (Tukey's HSD)
if (summary(anova_result)[[1]]$`Pr(>F)`[1] < 0.05) {
  post_hoc_result <- TukeyHSD(anova_result)
  print(post_hoc_result)
} else {
  print("No significant differences found.")
}