Merge pull request #45 from noaa-afsc/21-update_inputs_for_2026

21 update inputs for 2026

Author: GeoffMayhew-NOAA

.gitignore

@@ -3,4 +3,7 @@
 *.Rproj.user
 *.Ruserdata
 *.RData
+/analyses/monitoring_costs/ais_billable_sea_days/*.xlsx
+/analyses/monitoring_costs/ais_billable_sea_days/travel/*.xlsx
+*SCRATCH.*
 .Rproj.user

analyses/effort_prediction/effort_prediction.R

@@ -14,9 +14,7 @@
 ## User inputs ---- ADPyear, whether outputs are to be saved this run, and the auth token for getting data.
 #==============================#
 
-ADPyear <- as.numeric(rstudioapi::showPrompt(title = "Enter year",
-                                             message = "What year is the ADP year? (four digits, example: 2025)"))
-
+ADPyear <- 2026
 saveoutputs <- "YES" # Must be "YES" to save figures and rdata file. Use any other value to skip saves.
 
 if(!require("FMAtools")) devtools::install_github("Alaska-Fisheries-Monitoring-Analytics/FMAtools")

analyses/effort_prediction/figures/figure_c3.png

@@ -0,0 +1,9 @@
+This folder contains the monthly travel cost spreadsheets given to Lisa Thompson by AIS. Occasionally the .xlsx files are not alwas immediately readable by R so copies that work are made (e.g., `April 2025 Travel_FINAL.xlsx` has a copy that works called `April 2025 Travel_FIXED.xlsx`).
+
+The travel spreadsheet folder is here: 
+https://drive.google.com/drive/folders/15lu3k2glo5Msa9F_bEo9lUSstRlSE6J6
+
+...and resides within the folder that contains the AIS billable days spreadsheets: https://drive.google.com/drive/folders/1T1u8JuXQ1XLEAJR9So4HtHEA8upz1I8C
+
+Lisa Thompson is the owner.
+

analyses/monitoring_costs/ais_billable_sea_days/travel/README.txt

@@ -652,37 +652,11 @@ fg_em <- unique(fg_em)
 # Counts of vessels listed in EM
 fg_em[, .N, by = FLAG]  
 
-# * Effort prediction ----
-effort_strata <- work.data |>
-  # select necessary columns
-  _[CVG_NEW == "PARTIAL", .(ADP, STRATA = STRATA_NEW, TRIP_TARGET_DATE, TRIP_ID)
-    # ensure one trip target date per trip, making it the minimum trip target date
-  ][, TRIP_TARGET_DATE := min(TRIP_TARGET_DATE), by = TRIP_ID] |>
-  unique() |>
-  setorder(ADP, STRATA, TRIP_TARGET_DATE) |>
-  _[ , TRIPS := 1/.N, by = TRIP_ID
-     # find julian dates
-  ][, JULIAN_DATE := yday(TRIP_TARGET_DATE)
-    # set julian date to 1 for trips that left in year adp - 1
-  ][, JULIAN_DATE := ifelse(year(TRIP_TARGET_DATE) < ADP, 1, JULIAN_DATE)
-    # set julian date to 366 for trips that left in year adp + 1
-  ][, JULIAN_DATE := ifelse(year(TRIP_TARGET_DATE) > ADP, 366, JULIAN_DATE)][] 
-
-#' isolate the latest date for which we have data in the most recent year of VALHALLA
-effort_strata.max_date <- max(effort_strata[ADP == ADPyear - 1, JULIAN_DATE])
-# count trips through max_date and total trips by year and stratum
-effort_strata <- effort_strata[, .(
-  MAX_DATE_TRIPS = sum(TRIPS[JULIAN_DATE <= effort_strata.max_date]),
-  TOTAL_TRIPS = sum(TRIPS)
-), by = .(ADP, STRATA)
-# make total trips NA for ADPyear - 1, since the year is not over
-][ADP == ADPyear - 1, TOTAL_TRIPS := NA][]
-
 #' *===================================================================================================================*
 
-# MANUAL CORRECTIONS : Split long TRIP_IDs ----
+# FINAL MANUAL CORRECTIONS : Split long TRIP_IDs ----
 
-# These corrections need to be made before they affect trip_duration.R
+#' *These corrections need to be made before they affect effort_prediction.R and trip_duration.R*
 
 #' [2026FinalADP: Splitting two OB_TRW_GOA tender trips that merged trips over a large time span]
 
@@ -717,6 +691,32 @@ work.data |>
 
 #' *===================================================================================================================*
 
+# * Effort prediction ----
+effort_strata <- work.data |>
+  # select necessary columns
+  _[CVG_NEW == "PARTIAL", .(ADP, STRATA = STRATA_NEW, TRIP_TARGET_DATE, TRIP_ID)
+    # ensure one trip target date per trip, making it the minimum trip target date
+  ][, TRIP_TARGET_DATE := min(TRIP_TARGET_DATE), by = TRIP_ID] |>
+  unique() |>
+  setorder(ADP, STRATA, TRIP_TARGET_DATE) |>
+  _[ , TRIPS := 1/.N, by = TRIP_ID
+     # find julian dates
+  ][, JULIAN_DATE := yday(TRIP_TARGET_DATE)
+    # set julian date to 1 for trips that left in year adp - 1
+  ][, JULIAN_DATE := ifelse(year(TRIP_TARGET_DATE) < ADP, 1, JULIAN_DATE)
+    # set julian date to 366 for trips that left in year adp + 1
+  ][, JULIAN_DATE := ifelse(year(TRIP_TARGET_DATE) > ADP, 366, JULIAN_DATE)][] 
+
+#' isolate the latest date for which we have data in the most recent year of VALHALLA
+effort_strata.max_date <- max(effort_strata[ADP == ADPyear - 1, JULIAN_DATE])
+# count trips through max_date and total trips by year and stratum
+effort_strata <- effort_strata[, .(
+  MAX_DATE_TRIPS = sum(TRIPS[JULIAN_DATE <= effort_strata.max_date]),
+  TOTAL_TRIPS = sum(TRIPS)
+), by = .(ADP, STRATA)
+# make total trips NA for ADPyear - 1, since the year is not over
+][ADP == ADPyear - 1, TOTAL_TRIPS := NA][]
+
 # * Trip duration ----
 
 #' First, trim off the most recent 4 years of trips. Modeling trip duration over the entire dataset is not needed and