## ----setup, include = FALSE---------------------------------------------------
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

## ----season-data, message = FALSE---------------------------------------------
library(tidycreel)

set.seed(2024)

# Full season: May 1 – September 30 (184 days)
all_dates <- seq(as.Date("2024-05-01"), as.Date("2024-09-30"), by = "1 day")
day_types <- ifelse(
  as.integer(format(all_dates, "%u")) %in% c(6L, 7L), "weekend", "weekday"
)

season_calendar <- data.frame(
  date     = all_dates,
  day_type = day_types
)

# Stratified random sample: 30% of weekdays, 50% of weekends
set.seed(2024)
sampled <- ave(
  seq_along(all_dates), day_types,
  FUN = function(i) {
    rate <- if (day_types[i[1]] == "weekend") 0.50 else 0.30
    as.integer(runif(length(i)) < rate)
  }
)
sampled_calendar <- season_calendar[as.logical(sampled), ]

# Generate plausible effort counts (angler-hours per count)
# Weekend counts higher than weekday
sampled_counts <- data.frame(
  date = sampled_calendar$date,
  day_type = sampled_calendar$day_type,
  n_anglers = ifelse(
    sampled_calendar$day_type == "weekend",
    round(rnorm(sum(sampled), mean = 55, sd = 18)),
    round(rnorm(sum(sampled), mean = 28, sd = 12))
  )
)
sampled_counts$n_anglers <- pmax(sampled_counts$n_anglers, 0L)

nrow(sampled_calendar) # Number of sampled days

## ----season-effort, message = FALSE-------------------------------------------
season_design <- creel_design(
  sampled_calendar,
  date = date, strata = day_type
)
season_design <- add_counts(season_design, sampled_counts)

season_effort <- estimate_effort(season_design)
season_effort$estimates

## ----monthly-effort, message = FALSE------------------------------------------
months <- 5:9
month_labels <- c("May", "June", "July", "August", "September")

monthly_effort <- lapply(seq_along(months), function(i) {
  m <- months[i]
  cal_m <- sampled_calendar[format(sampled_calendar$date, "%m") == sprintf("%02d", m), ]
  cnt_m <- sampled_counts[format(sampled_counts$date, "%m") == sprintf("%02d", m), ]

  if (nrow(cal_m) == 0) {
    return(NULL)
  }

  des_m <- creel_design(cal_m, date = date, strata = day_type)
  des_m <- add_counts(des_m, cnt_m)
  est <- estimate_effort(des_m)$estimates

  cbind(month = month_labels[i], est)
})

do.call(rbind, monthly_effort)

## ----season-from-months, message = FALSE--------------------------------------
monthly_df <- do.call(rbind, monthly_effort)

season_from_months <- data.frame(
  stratum  = "Season total",
  estimate = sum(monthly_df$estimate),
  se       = sqrt(sum(monthly_df$se^2))
)

season_from_months

## ----interview-data, message = FALSE------------------------------------------
set.seed(2024)

# Three interviews per sampled day (ensures ≥10 complete trips per month)
n_per_day <- 3L
n_int <- nrow(sampled_calendar) * n_per_day
catch_total <- rpois(n_int, lambda = 1.8)
interviews <- data.frame(
  date         = rep(sampled_calendar$date, each = n_per_day),
  day_type     = rep(sampled_calendar$day_type, each = n_per_day),
  trip_status  = "complete",
  hours_fished = round(rnorm(n_int, mean = 3.5, sd = 1.2), 1),
  catch_total  = catch_total,
  catch_kept   = pmin(rpois(n_int, lambda = 0.6), catch_total)
)
interviews$hours_fished <- pmax(interviews$hours_fished, 0.5)

## ----monthly-catch, message = FALSE-------------------------------------------
monthly_catch <- lapply(seq_along(months), function(i) {
  m <- months[i]
  cal_m <- sampled_calendar[format(sampled_calendar$date, "%m") == sprintf("%02d", m), ]
  cnt_m <- sampled_counts[format(sampled_counts$date, "%m") == sprintf("%02d", m), ]
  int_m <- interviews[format(interviews$date, "%m") == sprintf("%02d", m), ]

  if (nrow(cal_m) == 0) {
    return(NULL)
  }

  des_m <- creel_design(cal_m, date = date, strata = day_type)
  des_m <- add_counts(des_m, cnt_m)
  des_m <- add_interviews(des_m, int_m,
    trip_status = trip_status,
    catch = catch_total,
    effort = hours_fished,
    n_anglers = 1, # every interview is a single angler
    harvest = catch_kept
  )

  effort_est <- estimate_effort(des_m)
  catch_est <- estimate_total_catch(des_m)

  cbind(month = month_labels[i], catch_est$estimates)
})

do.call(rbind, monthly_catch)

## ----season-catch-------------------------------------------------------------
catch_df <- do.call(rbind, monthly_catch)
season_catch <- data.frame(
  stratum  = "Season total",
  estimate = sum(catch_df$estimate),
  se       = sqrt(sum(catch_df$se^2))
)
season_catch

## ----season-summary, message = FALSE------------------------------------------
# Collect effort and catch estimates for each month into named lists
effort_list <- list()
catch_list <- list()

for (i in seq_along(months)) {
  m <- months[i]
  label <- month_labels[i]

  cal_m <- sampled_calendar[format(sampled_calendar$date, "%m") == sprintf("%02d", m), ]
  cnt_m <- sampled_counts[format(sampled_counts$date, "%m") == sprintf("%02d", m), ]
  int_m <- interviews[format(interviews$date, "%m") == sprintf("%02d", m), ]

  if (nrow(cal_m) == 0) next

  des_m <- creel_design(cal_m, date = date, strata = day_type)
  des_m <- add_counts(des_m, cnt_m)
  des_m <- add_interviews(des_m, int_m,
    trip_status = trip_status,
    catch = catch_total,
    effort = hours_fished,
    n_anglers = 1, # every interview is a single angler
    harvest = catch_kept
  )

  effort_list[[label]] <- estimate_effort(des_m)
  catch_list[[label]] <- estimate_total_catch(des_m)
}

# Assemble effort report
effort_summary <- season_summary(effort_list)
effort_summary$table

## ----catch-summary, message = FALSE-------------------------------------------
catch_summary <- season_summary(catch_list)
catch_summary$table

## ----multi-year, eval = FALSE-------------------------------------------------
# # Conceptual pattern — replace with actual data per year
# year_effort <- list()
# 
# for (yr in c(2022, 2023, 2024)) {
#   cal_yr <- subset(full_calendar, format(date, "%Y") == yr)
#   cnt_yr <- subset(full_counts, format(date, "%Y") == yr)
# 
#   des_yr <- creel_design(cal_yr, date = date, strata = day_type)
#   des_yr <- add_counts(des_yr, cnt_yr)
# 
#   year_effort[[as.character(yr)]] <- estimate_effort(des_yr)
# }
# 
# # Multi-year report table
# season_summary(year_effort)$table

## ----export, eval = FALSE-----------------------------------------------------
# write_schedule(effort_summary$table, "effort_by_month_2024.csv")
# write_schedule(effort_summary$table, "effort_by_month_2024.xlsx")