_targets.R

library(targets)
library(tarchetypes)
source("src/functions.R")
source("src/glider_funcs.R")
source("src/util_funcs.R")
options(tidyverse.quiet = TRUE)

tar_option_set(packages = c("data.table", "sf", "glatos", "geosphere", "viridisLite",  "ggplot2", "raster", "flextable", "terra", "geosphere", "leaflet", "readxl", "leafgl", "spatstat.core", "tidyverse", "DT", "kableExtra"))

list(
  # load path to range curve data collected by Binder in 2016 in GB with V13 tags
  tar_target(
    raw_log_reg_H_GBE,
    "data/Results/Results/GBE/LogisticRegressions/HighPower/logisticResultsH.csv",
    format = "file"
  ),
  
  # load path to Saginaw Bay Reef locations
  tar_target(
    raw_SB_reefs,
    "data/Saginaw Bay Reef Locations.xlsx",
    format = "file"
    ),
  
  # load path to LWF receivers
  tar_target(
    raw_grid,
    "data/us_receivers_20201215.csv",
    format = "file"
  ),

  # load path to Sag Bay polygon from EPA-AOC
  tar_target(
    raw_sbay,
    "data/aoc_mi_saginaw_river_bay/AOC_MI_SAGINAW_RIVER_BAY.shp",
    format = "file"
  ),

  # Convert SB polygon of SB to wgs84
  tar_target(
    sbay,
    {shp <- st_read(raw_sbay, quiet = TRUE);
      shp <- shp[shp$AOC == "AREA OF CONERN",];
      st_transform(shp, crs = 4326)
      },
    format = "rds"
  ),

  # Convert LWF grid to sf object
  tar_target(
    grid,
    {grd <- fread(raw_grid);
      grd <- grd[!(station %in% c("YTZ-035", "YTZ-031", "YTZ-030", "YTZ-034", "YTZ-039", "YTZ-042", "YTZ-038")),];
      st_as_sf(grd, coords = c("longitude", "latitude"), remove = FALSE, crs = 4326, agr = "constant")
    },
    format = "rds"
  ),

  # write out planning map
  tar_target(
    plan_map,
    .plan_map(grid, reefs, dirty_lines, grid10km = grid_10km_dup_reefs, bay_outline = sbay, pth = "output/comb_plan_map.html", write = TRUE),
    format = "file"
  ),
  

# raw file for lake huron bathymetry data
  tar_target(
    raw_bathy,
    "data/LH_bathy/huron_lld.tif", 
  ),
  
  tar_target(
    bathy,
    prep_bathy(raw_bathy, xmin_out = -83.96, xmax_out = -82.535, ymin_out = 43.5, ymax_out = 44.44, out_pth = "~/Documents/WA_sources_of_reproduction_2021/grid_planing/output/bathy_SB.tif"),
    format = "file"
  ),

tar_target(
  grid_depth,
  depth_extract(raw_recs = grid, bathy = bathy, lidar = NULL),
  format = "rds"
  ),

## tar_target(
##   map,
##   grid_map(bathy,
##            grid = grid_depth,
##            sbay,
##            reefs = reefs,
##            rec_grid = inner_bay_rec_grid,
##            spawn_rivers = spawn_rivs,
##            bay_mth = bay_mth_grd,
##            sim_fish = sim_tracks,
##            rand_grid = inner_bay_rec_grid_rand,
##            pth = "docs/index.html"),
##   format = "file"
## ),

tar_target(
  rec_grid,
  .grid(poly = sbay, cellsize = c(10000, 10000), in_crs = 3175, out_crs = 4326),
  format = "rds"
),

# read in Sag Bay reefs
# subset only 4 main reefs
tar_target(
  dirty_reefs,
  {out <- as.data.table(read_xlsx(path = raw_SB_reefs, range = "Reefs!D6:I15"));
    out <- out[Reef %in% c("Tawas Reef", "Little Charity Island Reef", "Duck Reef", "Coreyon (restored)", "North Island Reef"),]
   #fwrite(out, "data/SB_reefs.csv")
  },
  format = "fst_dt"
),

# proposed sturgeon river recs
tar_target(
  sturgeon,
  "data/Potential Receiver Deployments Saginaw River System 3.3.22.xlsx",
  format = "file"
),

# hard coded release locations!!!! this is going to be a problem!!!!
tar_target(
  dirty_sturgeon,
  {out <- as.data.table(read_xlsx(path = sturgeon, range = "Sheet1!A1:D16"));
    out <- out[c(1,3,5,9), id := "Rel"][is.na(id), id := "rec_pt"]
    out <- st_as_sf(out, coords = c("Longitude", "Latitide"), crs = 4326, agr = "constant")
  },
  format = "rds"
  ),


# contains possible tributary deployments
tar_target(
  rec_lines,
  "data/LH_tribs_lines.csv",
  format = "file"
),


tar_target(
  dirty_lines,
  {out <- fread(rec_lines);
    out <- out[GLATOS_ARRAY %in% c("HBC", "OSC"),];
    out <- out[STATION_NO %% 2 == 1,];
    out[,STATION_NO := 1:.N, by = .(GLATOS_ARRAY)]
#    out[, `:=`(center_lat = mean(DEPLOY_LAT), center_lon = mean(DEPLOY_LONG)), by = .(GLATOS_ARRAY, STATION_NO)]
    out[, `:=`(site = paste(GLATOS_ARRAY, STATION_NO, sep = "_"))]
    out<- st_as_sf(out, coords = c("DEPLOY_LONG", "DEPLOY_LAT"), agr = "constant", crs = 4326)
    },  
  format = "rds"
  ),

tar_target(
  reefs,
  clean_reefs(x = dirty_reefs),
  format = "rds"
),

tar_target(
  raw_spawn_riv,
  "data/spawning_rivers.csv",
  format = "file"
),

tar_target(
  spawn_rivs,
  .spawn_rivs(raw = raw_spawn_riv),
  format = "rds"
),

# this is raw layer for Saginaw Bay mouth sub-array.  Adjust the size of this and the inner- bay sub-array should automatically adjust.
tar_target(
  sag_bay_mth_raw,
  "data/SB_mouth.gpkg",
  format = "file"
),

tar_target(
  sag_bay_mth,
  sf::st_read(sag_bay_mth_raw, quiet = TRUE, agr = "constant"),
  format = "rds"
),

tar_target(
  bay_mth,
  sf::st_intersection(sag_bay_mth, sbay),
  format = "rds"
),


# create bay mouth array
tar_target(
  bay_mth_grd,
  .grid(poly = bay_mth, cellsize = c(5000, 5000), in_crs = 3175, out_crs = 4326),
  format = "rds"
),

tar_target(
  GL_poly,
  fix_CRS(),
  format = "rds"
),

tar_target(
  LH,
  crop(GL_poly, xmin = -84.387, xmax = -82.0915, ymin = 43, ymax = 45),
  format = "rds"
),

# contains summary of map and receiver deployments locations for most recent design (2022-07-06)
tar_render(
  dtc_summary,
  "src/coords.rmd",
  output_dir = "docs",
  output_file = "index.html"
),


# create simulated tracks.  Adjust inputs in "crw_in_polygon_updated" to change nature of tracks
## tar_target(
##   sim_tracks,
##   .sim_tracks(
##     n_trks = 2,
##     poly = LH,
##     theta = c(0,25),
##     stepLen = 1000,
##     initHeading = 40,
##     nsteps = 150,
##     initPos = c(-83.6248, 43.8824),
##     sp_out = FALSE,
##     show_progress = FALSE,
##     recs = bay_mth_grd),
##   format = "fst_dt"
##   ),

# calculate min distance between fish and all receivers in grid
## tar_target(
##   min_dist,
##   sim_fish_rec_dist(fsh_trks = sim_tracks, rec_grid = bay_mth_grd),
##   format = "fst_dt"
## ),

#transmissions for all simulated fish tracks and simulated tags
## tar_target(
##   sim_tag_trans,
##   .sim_tag_trans(
##     path = sim_tracks,
##     vel = 0.5,
##     delayRng = c(280,400),
##     burstDur = 5,
##     EPSG = 3175,
##     sp_out = FALSE),
##   format = "fst_dt"
## ),

## # determine which transmission were detected or not on bay mouth receivers only
## tar_target(
##   sim_dtc_low_perf,
##   .sim_dtc(dtc_trans = sim_tag_trans,
##            recLoc = bay_mth_grd,
##            EPSG = 3175,
##            sp_out = FALSE,
##            show_progress = FALSE,
##            min_dist = sim_tracks,
##            ba = c(2.913904, -0.0051440101)),
##   format = "fst_dt"
## ),

## # determine which transmissions were detected or not on bay mouth receivers only
## tar_target(
##   sim_dtc_high_perf,
##   .sim_dtc(dtc_trans = sim_tag_trans,
##            recLoc = bay_mth_grd,
##            EPSG = 3175,
##            sp_out = FALSE,
##            show_progress = FALSE,
##            min_dist = sim_tracks,
##            ba = c(3.104209, -0.0021253279)),
##   format = "fst_dt"
## ),

# create "inner bay" area by removing bay mouth region from sag bay polygon
## tar_target(
##   in_bay,
##   .difference(x = sbay, y = sag_bay_mth),
##   format = "rds"
## ),

# create "inner bay" receiver grid
# also consider spatstat::nncross for nearest neighbour type analysis
# see nngeo::st_nn function.  Prehaps easier to randomly place points in Sag Bay and then remove all points that are within a spedified distance?
# generate random points in polygon with minimum distance from one another and from polygon boundary
# https://gis.stackexchange.com/questions/163287/randomly-sampling-points-in-r-with-minimum-distance-constraint

## tar_target(
##   inner_bay_rec_grid,
##   .inner_bay_rec_grid(bbox = c(xmin = -83.948193, xmax = -82.946270, ymin = 43.595602, ymax = 44.277664),
##                       cellsize = c(10000, 10000),
##                       inner_bay_poly = in_bay,
##                       reefs = reefs),
##   format = "rds"
## ),


# grid entire sag bay
# 5 km
tar_target(
  grid_5km,
  .inner_bay_rec_grid(bbox = c(xmin = -83.948193, xmax = -82.946270, ymin = 43.595602, ymax = 44.277664),
                      cellsize = c(5000, 5000),
                      inner_bay_poly = sbay,
                      reefs = reefs),
  format = "rds"
),

# 10 km
tar_target(
  grid_10km,
  .inner_bay_rec_grid(bbox = c(xmin = -83.948193, xmax = -82.946270, ymin = 43.595602, ymax = 44.277664),
                      cellsize = c(10000, 10000),
                      inner_bay_poly = sbay,
                      reefs = reefs),
  format = "rds"
),

# 10 km base grid(includes both recevier and grid points, some overlap, 2022-07-11)
tar_target(
  grid_10km_dup_reefs,
  .inner_bay_rec_grid(bbox = c(xmin = -83.948193, xmax = -82.946270, ymin = 43.595602, ymax = 44.277664),
                      cellsize = c(10000, 10000),
                      inner_bay_poly = sbay,
                      reefs = reefs,
                      all_dups = TRUE),
  format = "rds"
),

# 15 km
tar_target(
  grid_15km,
  .inner_bay_rec_grid(bbox = c(xmin = -83.948193, xmax = -82.946270, ymin = 43.595602, ymax = 44.277664),
                      cellsize = c(15000, 15000),
                      inner_bay_poly = sbay,
                      reefs = reefs),
  format = "rds"
),

# adjusted grid to avoid shipping lanes, shallow water, etc
tar_target(
  raw_adj_grid,
  "data/grid_adj_20220621.csv",
  format = "file"
),

# visualize and clean up pts for rec deployments (latest version of grid 2022-06-23)
tar_target(
  leaflet_pts,
  clean_leaflet(reefs = reefs, LH_grid = grid, dirty_lines = dirty_lines, sag_grid = grid_10km_dup_reefs, adjustments = raw_adj_grid, sturgeon = dirty_sturgeon_pts, rivs = spawn_rivs),
  format = "rds"
),

# create gpx of only points in bay that HBBS is responsible for tending (2022-07-06)
tar_target(
  gpx_file,
  {
    foo <- leaflet_pts %>% filter(glatos_array != "YTZ" & !is.na(glatos_array) & !(glatos_array %in% c("KAW", "RIF", "AUG", "TAW", "PIN", "SGN", "PNC", "QUA", "PIG"))) %>% select(site) %>% rename(name = site)
    st_write(foo, "output/Sbay_walleye_recs.gpx", driver = "GPX", append = FALSE)
    return("output/Sbay_walleye_recs.gpx")},
    format = "file"
),

# create gpx of all points in bay (no LWF receivers, but include all sag tribs)(2022-07-06)
tar_target(
  gpx_file_all,
  {foo <- leaflet_pts %>% filter(glatos_array != "YTZ" | is.na(glatos_array)) %>% select(site) %>% rename(name = site)
    st_write(foo, "output/Sbay_walleye_recs_all.gpx", driver = "GPX", append = FALSE)
    return("output/Sbay_walleye_recs_all.gpx")},
  format = "file"
),

# create kml of all points in bay (no LWF receivers, but all sag tribs) (2022-07-06)
tar_target(
  kml_all,
  {foo <- leaflet_pts %>% filter(glatos_array != "YTZ" | is.na(glatos_array)) %>% select(site) %>% rename(name = site)
    st_write(foo, "output/Sbay_walleye_recs_all.kml", driver = "kml", append = FALSE)
    return("output/Sbay_walleye_recs_all.kml")},
  format = "file"
  ),

tar_target(
  raw_sturgeon,
  "data/Proposed Sturgeon Receivers in Tribs v1 (1).kmz",
  format = "file"
),

tar_target(
  dirty_sturgeon_pts,
  {foo <- sf::st_read(raw_sturgeon)
    foo <- st_zm(foo)
    setDT(foo)
    # add site names
    foo[, site := c("TRA_1", "TRR_1", "TRL_1", "CRA_1", "CRR_1", "CRL_1", "FRR_1", "FRL_1", "FRL_2", "FRL_3", "SHA_1", "SHR_1", "SHL_1", "SHL_2", "SHL_3", "SAG_1", "SAG_2", "SAG_3", "SAG_4", "SAG_5")]

    foo <- st_as_sf(foo)
  },
  format = "rds"
),

# add in receiver maintenance data for all Sag Bay
tar_target(
  raw_maintenance,
  "data/WA_rec_maintenance_20220701.csv",
  format = "file"
),

# latest packing list for receiver deployments (2022-07-06)
tar_render(
  HBBS_rec_prep,
  "src/HBBS_rec_list.rmd",
  output_dir = "output",
  output_file = "HBBS_rec_deps_WA.html"
),

# add in MDNR and Purdue gear deployments in Saginaw Bay (2022-07-07)
tar_target(
  purdue_raw,
  "data/Saginaw Bay instrument package locations.xlsx",
  format = "file"
),

tar_target(
  purdue,
  .purdue(x = purdue_raw),
  format = "rds"
)


)


    
 
#)