KYSTREV_Chla_summer_limits.Rmd

---
title: "KYSTREV_Chla_limits"
author: "Sandra Gran & Sonja Kistenich"
date: "2025-02-25"
output: html_document
description: "Script to calculate chlorophyll a summer means based on monitoring data from all ecoregions"
---

```{r, include=FALSE}
# List of required packages
packages <- c("xfun", "tidyverse", "readxl", "lubridate", "hms", "terra", "sf", "dplyr", "tidyr")

# Install missing packages
install_if_missing <- function(pkg) {
  if (!requireNamespace(pkg, quietly = TRUE)) {
    install.packages(pkg)
  }
}

# Check and install packages
lapply(packages, install_if_missing)

# Load libraries
lapply(packages, library, character.only = TRUE)

Sys.setlocale(locale='no_NB.utf8')


```
#Load dataset
```{r}
load("KYSTREV_2024_cleandata_final_v3.RData")
```

#Select relevant parameters
```{r}
nutchla_data = df_final_newsnox %>% 
  filter(Parameter %in% c("KlfA","NH4","SiO2", "SNOx","TOTN", "PO4", "TOTP", "Temp","Salt")) %>% 
  dplyr::select(Master_Ecoregion, Master_RegType,Master_station_code, Sample_date, Depth, Parameter,Master_value) %>% 
  pivot_wider(names_from = Parameter, values_from = Master_value) %>% 
  mutate(Master_station_code = gsub("Ærøydypet", "Aeroydypet", Master_station_code))
```

#Set new limits
```{r}
chlfa <- nutchla_data %>%
  filter(!(Master_Ecoregion == "Skagerrak" & year(Sample_date) < 2000)) %>%
  select(Master_Ecoregion, Master_RegType, Master_station_code, Sample_date, Depth, KlfA) %>% 
  rename(Chla = KlfA) %>% 
  mutate(Year = year(Sample_date),
         Month = month(Sample_date, label = TRUE, abbr = FALSE)) %>%
  na.omit() %>% 
  group_by(Master_Ecoregion, Master_RegType, Master_station_code, Sample_date, Year, Month) %>%
  summarise(Chla = mean(Chla, na.rm = TRUE),
            count_per_month = n()) %>% 
  ungroup()


# Filter the data for each ecoregion and combine
filtered_data <- chlfa %>% 
  filter(Month %in% c("juni", "juli", "august", "september")) %>%
  group_by(Master_Ecoregion, Master_RegType, Year) %>%
  filter(n_distinct(Month) >=2) %>%
  ungroup()

# Calculate the summer average of Chla per ecoregion and regtype
summer_avg <- filtered_data %>%
  group_by(Master_Ecoregion, Master_RegType, Year) %>%
  summarise(summer_avg_Chla = mean(Chla, na.rm = TRUE),
            count = n()) %>% 
  ungroup() %>%
  arrange(Master_RegType)


# Calculate the 5th percentile and the required limits
limits_1 <- summer_avg %>%
  group_by(Master_Ecoregion, Master_RegType) %>%
  summarise(
    fifth_percentile = quantile(summer_avg_Chla, probs = 0.05, na.rm = TRUE),
    count_n = sum(count)
  )

limits <- summer_avg %>%
  group_by(Master_Ecoregion) %>%
  summarise(
    fifth_percentile = quantile(summer_avg_Chla, probs = 0.05, na.rm = TRUE),
    count_n = sum(count)
  ) %>%
  bind_rows(limits_1) %>%
  mutate(
    lim_SG_G = fifth_percentile * 1.5,
    lim_G_M = lim_SG_G * 1.5,
    lim_M_D = lim_G_M * 1.5,
    lim_D_SD = lim_M_D * 1.5
  ) 

# Select and rename columns as required
result_table <- limits %>%
  select(Master_Ecoregion, Master_RegType, count_n, fifth_percentile, lim_SG_G, lim_G_M, lim_M_D, lim_D_SD) %>%
  rename(
    Ecoregion = Master_Ecoregion,
    RegType = Master_RegType,
    `5th_Percentile` = fifth_percentile
  ) %>%
  arrange(Ecoregion, RegType) %>%
  mutate(across(4:8, round, 2))

# Print the result table
print(result_table)
writexl::write_xlsx(result_table, "KYSTREV_5thperc_New_limits_Chla.xlsx")
```