WS1207

setup

if (!nzchar(system.file(package = "librarian"))) {
  install.packages("librarian")
}

librarian::shelf(
  quiet = TRUE,
  readr, here, fs, ggplot2, glue, "jiho/castr", dplyr, oce, patchwork, purrr, tidyr
)

Code

ctd_dir <- here("data/01_raw/combined_fl_keys_data")
fpath <- here(glue(
  "{ctd_dir}/{params$cruise_id}_fl_keys_ctd.csv"
))

cruise_df <- read_csv(fpath, show_col_types = FALSE, col_types = cols(station = col_character()))

plot depth & pressure vs time elapsed

tryCatch({
  plots <- list()
  
  for (station_name in unique(cruise_df$station)) {
    subset_df <- filter(cruise_df, station == station_name)
    plots[[station_name]] <- ggplot(subset_df, aes(x = time_elapsed)) +
      geom_point(aes(y = depth), color = "blue") +  # Plot depth in blue
      geom_line(aes(y = sea_water_pressure), color = "red") +  # Plot sea water pressure in red
      ggtitle(glue("{station_name}")) +
      theme(
        axis.title.x = element_blank(),
        axis.title.y = element_blank(),
        axis.text.x = element_blank(),
        axis.text.y = element_blank(),
        axis.ticks.x = element_blank(),
        axis.ticks.y = element_blank()
      )  # Remove individual axis titles and text
  }
  
  # drop nulls
  plots <- purrr::compact(plots)
  
  # Combine all plots into a grid
  combined_plot <- wrap_plots(plots) + 
    plot_layout(ncol = 4) +  # Adjust ncol to set number of columns in the grid
    plot_annotation(
      title = "Depth (blue) and Pressure (red)",
      subtitle = "Each panel represents a different station",
      caption = "Time Elapsed (x-axis) vs Depth & Pressure (y-axis)"
    ) +
    theme(
      plot.tag = element_text(size = 12, face = "bold"),
      plot.tag.position = "topleft",
      axis.title.x = element_text(margin = margin(t = 10)),
      axis.title.y = element_text(margin = margin(r = 10))
    )
  
  print(combined_plot)
}, error = function(er){
  print(er)
});

plot across all stations

p <- ggplot(cruise_df, aes(x = time, y = depth, fill = station)) +
  geom_col() +  # This creates the bars
  # geom_text(aes(label = station), vjust = -0.3) +  # This adds labels to each bar, adjust vjust for position
  labs(x = "Time", y = "Depth", title = "Depth over Time by Station") +  # Set labels and title
  theme_minimal()  # Use a minimal theme
print(p)

create oce.ctd objects from dataframes

ctd_load <- function(data, other_params = NULL) {
  
  # create csv into ctd object
  test_ctd <-
    as.ctd(
      salinity    = data$sea_water_salinity,
      temperature = data$sea_water_temperature,
      pressure    = data$sea_water_pressure,
      station     = data$station
    )
  
  # add additional columns to ctd object
  if (!is.null(other_params)) {
    for (param_name in other_params) {
      test_ctd <-
        oceSetData(
          object = test_ctd,
          name   = param_name,
          value  = data[[param_name]]
        )
    }
  }
  print(glue("{data$station[1]}:\t{length(test_ctd@data$scan)} scans"))


  return(test_ctd)
}

# Define other parameters to add
other_params <- c(
  "cruise_id", "station", "time", "time_elapsed", 
  "latitude", "longitude", "sea_water_electrical_conductivity", 
  "CDOM", "dissolved_oxygen","oxygen_saturation", "chlorophyll_concentration", 
  "chlorophyll_fluorescence", "photosynthetically_available_radiation", 
  "beam_attenuation","beam_transmission", "depth", "sea_water_sigma_t",
  "descent_rate", "sound_velocity"
)

# Split data by station and create data list
ctd_FK <- cruise_df %>%
  split(.$station) %>%
  map(~ ctd_load(.x, other_params = other_params)) # ~ is a lambda(x)

1:  6229 scans
10: 942 scans
11: 5932 scans
16: 2668 scans
17: 2897 scans
18: 3082 scans
19: 600 scans
20: 981 scans
21.5:   6642 scans
21/LK:  2 scans
22.5:   6773 scans
23: 1437 scans
24: 2437 scans
25.5:   10862 scans
3:  4950 scans
5:  1140 scans
6:  4531 scans
6.5:    4442 scans
7:  3066 scans
9:  2669 scans
9.5:    7519 scans

plotting scans for each cast in the first list

for (i in seq(ctd_FK)){
  cast <- ctd_FK[[i]]  # 1 is selecting only the first sublist
  # print(i)
  print(glue("=== station: {cast@metadata$station[1]}"))
  print(glue("# scans: {length(cast@data$scan)}"))
  plotScan(cast)
}

=== station: 1
# scans: 6229

=== station: 10
# scans: 942

=== station: 11
# scans: 5932

=== station: 16
# scans: 2668

=== station: 17
# scans: 2897

=== station: 18
# scans: 3082

=== station: 19
# scans: 600

=== station: 20
# scans: 981

=== station: 21.5
# scans: 6642

=== station: 21/LK
# scans: 2

=== station: 22.5
# scans: 6773

=== station: 23
# scans: 1437

=== station: 24
# scans: 2437

=== station: 25.5
# scans: 10862

=== station: 3
# scans: 4950

=== station: 5
# scans: 1140

=== station: 6
# scans: 4531

=== station: 6.5
# scans: 4442

=== station: 7
# scans: 3066

=== station: 9
# scans: 2669

=== station: 9.5
# scans: 7519

plotting each cast in the first list

for (i in seq(ctd_FK)){
  cast <- ctd_FK[[i]]  # 1 is selecting only the first sublist
  tryCatch({
    plot(ctdDecimate(ctdTrim(cast)))
  }, error = function(e){
    print(e)
  })
}

plotting other physical parameters for each cast

# Loop through each CTD cast
for (i in seq(ctd_FK)){
  cast <- ctd_FK[[i]]  # Assuming each sublist contains only one relevant CTD object
  tryCatch({
    # Extract metadata for station name and cruise ID
    station_name <- cast@metadata$station[1]
    cruise_id <- cast@data$cruise_id[1]

    # Generate a title with station name and cruise ID
    overall_title <- glue::glue("Station: {station_name}, Cruise ID: {cruise_id}")

    # Set margins: increase the outer margin for the title
    par(oma = c(0, 0, 3, 0))  # Top outer margin increased for title

    # Plotting function with specific parameters
    oce::plot(
      x = ctdDecimate(ctdTrim(cast)),
      which = c(
        "sea_water_electrical_conductivity",
        "descent_rate", "sound_velocity",
        "sea_water_sigma_t"
      ),
      main = ""  # No main title for individual subplots
    )

    # Place a single overall title at the top of the plot frame
    mtext(overall_title, side = 3, line = 1, outer = TRUE, cex = 1.5)

    # Reset outer margins to default
    par(oma = c(0, 0, 0, 0))

  }, error = function(e) {
    print(e$message)  # Print any errors that occur during plotting
  })
}

plotting other nutrient parameters for each cast

# Loop through each CTD cast
for (i in seq(ctd_FK)){
  cast <- ctd_FK[[i]]  # Assuming each sublist contains only one relevant CTD object
  tryCatch({
    # Extract metadata for station name and cruise ID
    station_name <- cast@metadata$station[1]
    cruise_id <- cast@data$cruise_id[1]

    # Generate a title with station name and cruise ID
    overall_title <- glue::glue("Station: {station_name}, Cruise ID: {cruise_id}")

    # Set margins: increase the outer margin for the title
    par(oma = c(0, 0, 3, 0))  # Top outer margin increased for title

    # Plotting function with specific parameters
    oce::plot(
      x = ctdDecimate(ctdTrim(cast)),
      which = c(
        "CDOM", "dissolved_oxygen",
        "oxygen_saturation",
        "chlorophyll_concentration", "chlorophyll_fluorescence"
      ),
      main = ""  # No main title for individual subplots
    )

    # Place a single overall title at the top of the plot frame
    mtext(overall_title, side = 3, line = 1, outer = TRUE, cex = 1.5)

    # Reset outer margins to default
    par(oma = c(0, 0, 0, 0))

  }, error = function(e) {
    print(e$message)  # Print any errors that occur during plotting
  })
}

[1] "In plot,ctd-method() : which=\"CDOMdissolved_oxygenoxygen_saturationchlorophyll_concentrationchlorophyll_fluorescence\" cannot be handled"
[1] "In plot,ctd-method() : which=\"CDOMdissolved_oxygenoxygen_saturationchlorophyll_concentrationchlorophyll_fluorescence\" cannot be handled"
[1] "In plot,ctd-method() : which=\"CDOMdissolved_oxygenoxygen_saturationchlorophyll_concentrationchlorophyll_fluorescence\" cannot be handled"
[1] "In plot,ctd-method() : which=\"CDOMdissolved_oxygenoxygen_saturationchlorophyll_concentrationchlorophyll_fluorescence\" cannot be handled"
[1] "In plot,ctd-method() : which=\"CDOMdissolved_oxygenoxygen_saturationchlorophyll_concentrationchlorophyll_fluorescence\" cannot be handled"
[1] "In plot,ctd-method() : which=\"CDOMdissolved_oxygenoxygen_saturationchlorophyll_concentrationchlorophyll_fluorescence\" cannot be handled"
[1] "In plot,ctd-method() : which=\"CDOMdissolved_oxygenoxygen_saturationchlorophyll_concentrationchlorophyll_fluorescence\" cannot be handled"
[1] "In plot,ctd-method() : which=\"CDOMdissolved_oxygenoxygen_saturationchlorophyll_concentrationchlorophyll_fluorescence\" cannot be handled"
[1] "In plot,ctd-method() : which=\"CDOMdissolved_oxygenoxygen_saturationchlorophyll_concentrationchlorophyll_fluorescence\" cannot be handled"
[1] "In plot,ctd-method() : which=\"CDOMdissolved_oxygenoxygen_saturationchlorophyll_concentrationchlorophyll_fluorescence\" cannot be handled"
[1] "In plot,ctd-method() : which=\"CDOMdissolved_oxygenoxygen_saturationchlorophyll_concentrationchlorophyll_fluorescence\" cannot be handled"
[1] "In plot,ctd-method() : which=\"CDOMdissolved_oxygenoxygen_saturationchlorophyll_concentrationchlorophyll_fluorescence\" cannot be handled"
[1] "In plot,ctd-method() : which=\"CDOMdissolved_oxygenoxygen_saturationchlorophyll_concentrationchlorophyll_fluorescence\" cannot be handled"
[1] "In plot,ctd-method() : which=\"CDOMdissolved_oxygenoxygen_saturationchlorophyll_concentrationchlorophyll_fluorescence\" cannot be handled"
[1] "In plot,ctd-method() : which=\"CDOMdissolved_oxygenoxygen_saturationchlorophyll_concentrationchlorophyll_fluorescence\" cannot be handled"
[1] "In plot,ctd-method() : which=\"CDOMdissolved_oxygenoxygen_saturationchlorophyll_concentrationchlorophyll_fluorescence\" cannot be handled"
[1] "In plot,ctd-method() : which=\"CDOMdissolved_oxygenoxygen_saturationchlorophyll_concentrationchlorophyll_fluorescence\" cannot be handled"
[1] "In plot,ctd-method() : which=\"CDOMdissolved_oxygenoxygen_saturationchlorophyll_concentrationchlorophyll_fluorescence\" cannot be handled"
[1] "In plot,ctd-method() : which=\"CDOMdissolved_oxygenoxygen_saturationchlorophyll_concentrationchlorophyll_fluorescence\" cannot be handled"
[1] "In plot,ctd-method() : which=\"CDOMdissolved_oxygenoxygen_saturationchlorophyll_concentrationchlorophyll_fluorescence\" cannot be handled"
[1] "In plot,ctd-method() : which=\"CDOMdissolved_oxygenoxygen_saturationchlorophyll_concentrationchlorophyll_fluorescence\" cannot be handled"

plotting other optical parameters for each cast

# Loop through each CTD cast
for (i in seq(ctd_FK)){
  cast <- ctd_FK[[i]]  # Assuming each sublist contains only one relevant CTD object
  tryCatch({
    # Extract metadata for station name and cruise ID
    station_name <- cast@metadata$station[1]
    cruise_id <- cast@data$cruise_id[1]

    # Generate a title with station name and cruise ID
    overall_title <- glue::glue("Station: {station_name}, Cruise ID: {cruise_id}")

    # Set margins: increase the outer margin for the title
    par(oma = c(0, 0, 3, 0))  # Top outer margin increased for title

    # Plotting function with specific parameters
    oce::plot(
      x = ctdDecimate(ctdTrim(cast)),
      which = c(
        "photosynthetically_available_radiation",
        "beam_attenuation","beam_transmission"
      ),
      main = ""  # No main title for individual subplots
    )

    # Place a single overall title at the top of the plot frame
    mtext(overall_title, side = 3, line = 1, outer = TRUE, cex = 1.5)

    # Reset outer margins to default
    par(oma = c(0, 0, 0, 0))

  }, error = function(e) {
    print(e$message)  # Print any errors that occur during plotting
  })
}

loop through every cast, clean, & save

combined_df <- data.frame()
for (i in seq(ctd_FK)){
  cast <- ctd_FK[[i]]  # 1 is selecting only the first sublist
  # print(class(cast))
  # clean cast 
  trimmed_cast <- ctdTrim(cast)
  decimated_cast <- ctdDecimate(trimmed_cast, p = 0.5)  # binned to 0.5 m
  
  # convert to df
  cast_df <- as.data.frame(decimated_cast@data)
  
  # Add metadata
  # assumes station ID and cruise ID the same for all & just uses 1st one
  cast_df <- mutate(
    cast_df,
    station = cast@data$station[1],
    cruise_id = cast@data$cruise_id[1]
  )

  # drop NA rows left by cleaning
  cast_df <- subset(cast_df, !is.na(scan))
  
  # Append the data to the combined dataframe
  combined_df <- rbind(combined_df, cast_df)
}
# Save to CSV
file_path <- here(glue("data/cleaned/{cruise_id}.csv"))
write.csv(combined_df, file_path, row.names = FALSE)