# nolint start: commented_code_linter, line_length_linter,object_usage_linter. library(here) library(sf) library(terra) library(tidyverse) library(lubridate) library(exactextractr) library(readxl) # Vang alle command line argumenten op args <- commandArgs(trailingOnly = TRUE) # Controleer of er ten minste één argument is doorgegeven if (length(args) == 0) { stop("Geen argumenten doorgegeven aan het script") } # Converteer het eerste argument naar een numerieke waarde end_date <- as.Date(args[1]) offset <- as.numeric(args[2]) # Controleer of weeks_ago een geldig getal is if (is.na(offset)) { # stop("Het argument is geen geldig getal") offset <- 7 } # Converteer het tweede argument naar een string waarde project_dir <- as.character(args[3]) # Controleer of data_dir een geldige waarde is if (!is.character(project_dir)) { project_dir <- "chemba" } laravel_storage_dir <- here("laravel_app/storage/app", project_dir) #preparing directories planet_tif_folder <- here(laravel_storage_dir, "merged_tif") merged_final <- here(laravel_storage_dir, "merged_final_tif") new_project_question = FALSE planet_tif_folder <- here(laravel_storage_dir, "merged_tif") merged_final <- here(laravel_storage_dir, "merged_final_tif") data_dir <- here(laravel_storage_dir, "Data") extracted_CI_dir <- here(data_dir, "extracted_ci") daily_CI_vals_dir <- here(extracted_CI_dir, "daily_vals") cumulative_CI_vals_dir <- here(extracted_CI_dir, "cumulative_vals") weekly_CI_mosaic <- here(laravel_storage_dir, "weekly_mosaic") daily_vrt <- here(data_dir, "vrt") harvest_dir <- here(data_dir, "HarvestData") source("parameters_project.R") source("ci_extraction_utils.R") dir.create(here(laravel_storage_dir)) dir.create(here(data_dir)) dir.create(here(extracted_CI_dir)) dir.create(here(daily_CI_vals_dir)) dir.create(here(cumulative_CI_vals_dir)) dir.create(here(weekly_CI_mosaic)) dir.create(here(daily_vrt)) dir.create(merged_final) dir.create(harvest_dir) # end_date <- lubridate::dmy("20-6-2024") week <- week(end_date) #weeks_ago = 0 # Creating weekly mosaic #dates <- date_list(weeks_ago) dates <- date_list(end_date, offset) print(dates) #load pivot geojson # pivot_sf_q <- st_read(here(data_dir, "pivot.geojson")) %>% dplyr::select(pivot, pivot_quadrant) %>% vect() # raster_files <- list.files(planet_tif_folder,full.names = T, pattern = ".tif") # filtered_files <- map(dates$days_filter, ~ raster_files[grepl(pattern = .x, x = raster_files)]) %>% # compact() %>% # flatten_chr() # head(filtered_files) # Extracting CI # pivot_sf_q <- st_read(here("..", "Data", "pivot_20210625.geojson")) %>% dplyr::select(pivot, pivot_quadrant) %>% vect() # pivot_sf <- st_read(here(data_dir, "pivot.geojson")) %>% dplyr::select(pivot, pivot_quadrant) %>% group_by(pivot) %>% summarise() %>% vect() # message("pivot loaded") raster_files_NEW <- list.files(merged_final,full.names = T, pattern = ".tif") # pivots_dates0 <- readRDS(here(harvest_dir, "harvest_data_new")) %>% filter( # pivot %in% c("1.1", "1.2", "1.3", "1.4", "1.6", "1.7", "1.8", "1.9", "1.10", "1.11", "1.12", "1.13", # "1.14" , "1.16" , "1.17" , "1.18" ,"2.1", "2.2", "2.3" , "2.4", "2.5", "3.1", "3.2", "3.3", # "4.1", "4.2", "4.3", "4.4", "4.5", "4.6", "5.1" ,"5.2", "5.3", "5.4", "6.1", "6.2", "DL1.1", "DL1.3") # ) # harvesting_data <- pivots_dates0 %>% # select(c("pivot_quadrant", "season_start_2021", "season_end_2021", "season_start_2022", "season_end_2022", "season_start_2023", "season_end_2023", "season_start_2024", "season_end_2024")) %>% # pivot_longer(cols = starts_with("season"), names_to = "Year", values_to = "value") %>% # separate(Year, into = c("name", "Year"), sep = "(?<=season_start|season_end)\\_", remove = FALSE) %>% # mutate(name = str_to_title(name)) %>% # pivot_wider(names_from = name, values_from = value) %>% # rename(field = pivot_quadrant) #If run for the firsttime, it will extract all data since the start and put into a table.rds. otherwise it will only add on to the existing table. # Define the path to the file file_path <- here(cumulative_CI_vals_dir,"combined_CI_data.rds") # Check if the file exists if (!file.exists(file_path)) { # Create the file with columns "column1" and "column2" data <- data.frame(sub_field=NA, field=NA) saveRDS(data, file_path) } print("combined_CI_data.rds exists, adding the latest image data to the table.") filtered_files <- map(dates$days_filter, ~ raster_files_NEW[grepl(pattern = .x, x = raster_files_NEW)]) %>% compact() %>% flatten_chr() walk(filtered_files, extract_rasters_daily, field_geojson= field_boundaries, quadrants = TRUE, daily_CI_vals_dir) extracted_values <- list.files(daily_CI_vals_dir, full.names = TRUE) extracted_values <- map(dates$days_filter, ~ extracted_values[grepl(pattern = .x, x = extracted_values)]) %>% compact() %>% flatten_chr() pivot_stats <- extracted_values %>% map(readRDS) %>% list_rbind() %>% group_by(sub_field) %>% summarise(across(everything(), ~ first(na.omit(.)))) combined_CI_data <- readRDS(here(cumulative_CI_vals_dir,"combined_CI_data.rds")) #%>% drop_na(pivot_quadrant) pivot_stats2 <- bind_rows(pivot_stats, combined_CI_data) # pivot_stats2 <- combined_CI_data print("All CI values extracted from latest 7 images.") saveRDS(combined_CI_data, here(cumulative_CI_vals_dir,"combined_CI_data.rds")) #used to save the rest of the data into one file # gather data into long format for easier calculation and visualisation pivot_stats_long <- pivot_stats2 %>% tidyr::gather("Date", value, -field, -sub_field ) %>% mutate(Date = right(Date, 8), Date = lubridate::ymd(Date) ) %>% drop_na(c("value","Date")) %>% mutate(value = as.numeric(value))%>% filter_all(all_vars(!is.infinite(.))) %>% # rename(field = pivot_quadrant, # sub_field = field) %>% unique() # #2021 # pivot_select_model_Data_2021 <- harvesting_data %>% filter(Year == 2021) %>% pull(field) # # pivot_select_model_Data_2022 <- harvesting_data %>% filter(Year == 2022) %>% pull(field) # pivot_select_model_Data_2023 <- harvesting_data %>% filter(year == 2023) %>% filter(!is.na(season_start)) %>% pull(sub_field) pivot_select_model_Data_2024 <- harvesting_data %>% filter(year == 2024)%>% filter(!is.na(season_start)) %>% pull(sub_field) # pivots_dates_Data_2022 <- pivots_dates0 %>% filter(!is.na(season_end_2022)) # pivot_select_model_Data_2022 <- unique(pivots_dates_Data_2022$pivot_quadrant ) # # pivots_dates_Data_2023 <- pivots_dates0 %>% filter(!is.na(season_start_2023)) # pivot_select_model_Data_2023 <- unique(pivots_dates_Data_2023$pivot_quadrant) # # pivots_dates_Data_2024 <- pivots_dates0 %>% filter(!is.na(season_start_2024)) # pivot_select_model_Data_2024 <- unique(pivots_dates_Data_2024$pivot_quadrant) extract_CI_data <- function(field_names, harvesting_data, field_CI_data, season) { # nolint: line_length_linter. # field_names = "AG1D06P" # field_names = "1.13A" # harvesting_data = harvesting_data # field_CI_data = pivot_stats_long # season= 2023 filtered_harvesting_data <- harvesting_data %>% na.omit() %>% filter(year == season, sub_field %in% field_names) filtered_field_CI_data <- field_CI_data %>% filter(sub_field %in% field_names) # CI <- map_df(field_names, ~ { ApproxFun <- approxfun(x = filtered_field_CI_data$Date, y = filtered_field_CI_data$value) Dates <- seq.Date(ymd(min(filtered_field_CI_data$Date)), ymd(max(filtered_field_CI_data$Date)), by = 1) LinearFit <- ApproxFun(Dates) CI <- data.frame(Date = Dates, FitData = LinearFit) %>% left_join(., filtered_field_CI_data, by = "Date") %>% filter(Date > filtered_harvesting_data$season_start & Date < filtered_harvesting_data$season_end) %>% mutate(DOY = seq(1, n(), 1), model = paste0("Data", season, " : ", field_names), season = season, sub_field = field_names) # }) #%>% #{if (length(field_names) > 0) message("Done!")} return(CI) } ## Extracting the correct CI values # Data_2021 <- map(pivot_select_model_Data_2021, ~ extract_CI_data(.x, harvesting_data = harvesting_data, field_CI_data = pivot_stats_long, season = 2021)) %>% list_rbind() # nolint: line_length_linter. # message('2021') # Data_2022 <- map(pivot_select_model_Data_2022, ~ extract_CI_data(.x, harvesting_data = harvesting_data, field_CI_data = pivot_stats_long, season = 2022)) %>% list_rbind() # message('2022') # Data_2023 <- map(pivot_select_model_Data_2023, ~ extract_CI_data(.x, harvesting_data = harvesting_data, field_CI_data = pivot_stats_long, season = 2023)) %>% list_rbind() # message('2023') Data_2024 <- map(pivot_select_model_Data_2024, ~ extract_CI_data(.x, harvesting_data = harvesting_data, field_CI_data = pivot_stats_long, season = 2024)) %>% list_rbind() message('2024') #CI_all <- rbind(Data_2023, Data_2024) CI_all <- Data_2024 message('CI_all created') #CI_all <- Data_2023 CI_all <- CI_all %>% group_by(model) %>% mutate(CI_per_day = FitData - lag(FitData), cumulative_CI = cumsum(FitData)) message('CI_all cumulative') head(CI_all) message('show head') saveRDS(CI_all, here(cumulative_CI_vals_dir,"All_pivots_Cumulative_CI_quadrant_year_v2.rds")) message('rds saved') # nolint end