FINAL prototype1 changes: moved making a locations df to a separate s…

…cript (because the stations on the main page were not showing up properly). This script creates separate location dfs for the datasets, and one for all for the main page. Also added HAB locations to the map (different colors and maps and all!)

02_Cleaning_scripts/Create_location_data.R

@@ -0,0 +1,115 @@
+########################################################################
+########## NERRS Science Transfer project - GTMNERR        #############
+########################################################################
+
+# Geraldine Klarenberg, PhD
+# [email protected]
+# 6 Sep 2024
+
+# Load packages
+library(tidyverse)
+
+# Script to create file with locations, for mapping
+
+#### Water Quality ####
+# import all WQ data
+WQ_df <- readRDS("./03_Data_for_app/WQ_all.Rds")
+
+WQ_years <- WQ_df %>% # site friendly and station code are the names in common
+  filter(variable %in% c("StationCode", "site_friendly",
+                         "geometry", 
+                         "SampleDate", # changed, from StartDate - but also not necessary for locations
+                         "Latitude",
+                         "Longitude",
+                         "data_source")
+  ) %>%
+  #select(c(RowID, variable, value)) %>% # not necessary
+  #distinct(RowID, variable, value) %>% # not necessary
+  pivot_wider(
+    names_from = variable,
+    values_from = value,
+    values_fill = list(value = NA)
+  ) %>%
+  mutate(
+    SampleDate = parse_date_time(SampleDate, c("ymd_HMS", "ymd")), # get a warning about 311 failed to parse, but they did, only as ymd
+    year = year(SampleDate),
+    Latitude = as.numeric(Latitude),
+    Longitude = as.numeric(Longitude)
+  ) %>% 
+  group_by(Latitude, Longitude, site_friendly) %>% 
+  summarize(minYear = min(year(SampleDate)),
+            maxYear = max(year(SampleDate)))
+
+# make dataframe for map display and hover data
+WQ_data_locations = WQ_df %>% # site friendly and station code are the names in common
+  filter(variable %in% c("StationCode", "site_friendly",
+                         "geometry", 
+                         "SampleDate", # changed, from StartDate - but also not necessary for locations
+                         "Latitude",
+                         "Longitude",
+                         "data_source")
+  ) %>%
+  #select(c(RowID, variable, value)) %>% # not necessary
+  #distinct(RowID, variable, value) %>% # not necessary
+  pivot_wider(
+    names_from = variable,
+    values_from = value,
+    values_fill = list(value = NA)
+  ) %>%
+  mutate(
+    #SampleDate = ymd_hms(SampleDate), # get a warning about 311 failed to parse, but they did, only as ymd
+    Latitude = as.numeric(Latitude),
+    Longitude = as.numeric(Longitude)
+  ) %>% 
+  select(-SampleDate) %>% 
+  distinct(Latitude, Longitude, geometry, data_source, StationCode, site_friendly)
+
+WQ_data_locations <- WQ_data_locations %>% 
+  left_join(WQ_years)
+
+# Save data
+saveRDS(WQ_data_locations, "03_Data_for_app/WQ_data_locations.Rds")
+
+WQ_data_locations <- WQ_data_locations %>% 
+  select(-geometry)
+
+#### Algae ####
+HAB_df <- readRDS("03_Data_for_app/HAB.Rds")
+
+# Get min/max years of measurements
+HAB_years <- HAB_df %>% 
+  mutate(date = dmy(`Sample Date`),
+         year = year(date)) %>% 
+  group_by(Latitude, Longitude, Site) %>% 
+  summarize(minYear = min(year),
+            maxYear = max(year)) %>% 
+  rename(site_friendly = Site)
+
+HAB_data_locations <- HAB_df %>%
+  distinct(Latitude, Longitude, Site) %>% 
+  rename(site_friendly = Site) %>% 
+  mutate(data_source = "FWC") # or GTMNERR?
+
+HAB_data_locations <- HAB_data_locations %>% 
+  left_join(HAB_years)
+
+saveRDS(HAB_data_locations, "03_Data_for_app/HAB_data_locations.Rds")
+
+
+#### Shellfish ####
+
+
+
+#### All - main page ####
+# Add dataset type to dfs
+WQ_data_locations["type"] <- "Water Quality"
+HAB_data_locations["type"] <- "Algae"
+
+
+# Merge
+all_data_locations <- WQ_data_locations %>% 
+  full_join(HAB_data_locations)
+
+# Save
+saveRDS(all_data_locations, "03_Data_for_app/all_data_locations.Rds")
+

modules/main_page.R

@@ -14,48 +14,46 @@
 # I am putting this here right now, but I feel we should move this to a cleaning
 # script so it doesn't need to be run every time someone uses the app (as with
 # WQ locations)
-WIN_df <- readRDS("./03_Data_for_app/WIN.Rds")
+# WIN_df <- readRDS("./03_Data_for_app/WIN.Rds")
+# 
+# WIN_data_locations = WIN_df %>%
+#   filter(variable %in% c("geometry", 
+#                          "StationCode", 
+#                          "SampleDate",
+#                          "Latitude",
+#                          "Longitude")
+#   ) %>%
+#   select(c(RowID, variable, value)) %>%
+#   distinct(RowID, variable, value) %>%
+#   pivot_wider(
+#     names_from = variable,
+#     values_from = value,
+#     values_fill = list(value = NA)
+#   ) %>%
+#   distinct(geometry, StationCode, SampleDate, Latitude, Longitude) %>%
+#   mutate(
+#     SampleDate = ymd_hms(SampleDate),
+#     Latitude = as.numeric(Latitude),
+#     Longitude = as.numeric(Longitude),
+#     type = "Water quality",
+#     dataset = "Watershed Information Network (DEP)", # Update this so we use data_source
+#     minYear = min(year(SampleDate)),
+#     maxYear = max(year(SampleDate))
+#   ) %>% 
+#   select(-geometry, -SampleDate)
 
-WIN_data_locations = WIN_df %>%
-  filter(variable %in% c("geometry", 
-                         "StationCode", 
-                         "SampleDate",
-                         "Latitude",
-                         "Longitude")
-  ) %>%
-  select(c(RowID, variable, value)) %>%
-  distinct(RowID, variable, value) %>%
-  pivot_wider(
-    names_from = variable,
-    values_from = value,
-    values_fill = list(value = NA)
-  ) %>%
-  distinct(geometry, StationCode, SampleDate, Latitude, Longitude) %>%
-  mutate(
-    SampleDate = ymd_hms(SampleDate),
-    Latitude = as.numeric(Latitude),
-    Longitude = as.numeric(Longitude),
-    type = "Water quality",
-    dataset = "Watershed Information Network (DEP)", # Update this so we use data_source
-    minYear = min(year(SampleDate)),
-    maxYear = max(year(SampleDate))
-  ) %>% 
-  select(-geometry, -SampleDate)
+#### Location data ------------------------------------------------
+all_data_locations <- readRDS("./03_Data_for_app/all_data_locations.Rds")
 
-#### WQ locations data ------------------------------------------------
-WQ_data_locations <- readRDS("./03_Data_for_app/WQ_locations.Rds")
+# add info for icons and colors
+all_data_locations <- all_data_locations %>%
+  mutate(group_icon = case_when(
+    type == "Water Quality" ~ "flask",
+    type == "Algae" ~ "microscope"),
+    group_color = case_when(
+      type == "Water Quality" ~ "orange",
+      type == "Algae" ~ "purple"))
 
-# For now, for testing, make WQ_locations the dataframe to be used for filtering
-datasets_location <- full_join(WQ_data_locations, WIN_data_locations)
-datasets_location <- st_as_sf(datasets_location, coords = c("Longitude", "Latitude"), crs = 4326)
-datasets_location <- datasets_location[!duplicated(datasets_location),]
-
-color_palette <- colorFactor(palette = c("red", "goldenrod1"), #, "blue", "green"
-                             domain = datasets_location$dataset)
-
-#### HAB locations data ------------------------------------------------
-# Add this at some point so there is another dataset that shows up in the dropdown
-# menu
 
 ### Define the UI -------------------------------------------------------------
 mainPageUI <- function(id) {
@@ -92,8 +90,8 @@ mainPageServer <- function(input, output, session) {
     selectInput(
       inputId = ns("datatype_selector"),
       label = "Select a type of data to see locations with data availability",
-      choices = unique(datasets_location$type),
-      selected = unique(datasets_location$type)[1]
+      choices = unique(all_data_locations$type),
+      selected = unique(all_data_locations$type)[1]
     )
   })
 
@@ -147,15 +145,16 @@ mainPageServer <- function(input, output, session) {
   # Select dataset to add markers to the plot
   observeEvent(input$datatype_selector, {
     # Filter data based on selected group
-    filtered_data <- datasets_location[datasets_location$type == input$datatype_selector,]
+    filtered_data <- all_data_locations[all_data_locations$type == input$datatype_selector,]
     print(filtered_data)
     # Add markers to the map - commented out the popup bc this only works for WQ data
     print("Adding markers")
     leafletProxy(ns("map")) %>%
       clearMarkers() %>%
-      addAwesomeMarkers(icon = makeAwesomeIcon(icon = "flask", markerColor = "orange", library = "fa",
-                                               iconColor = "black"),
+      addAwesomeMarkers(
                         data = filtered_data,
+                        icon = makeAwesomeIcon(icon = ~group_icon, markerColor = ~group_color, library = "fa",
+                                               iconColor = "black"),
                         options = markerOptions(riseOnHover = TRUE), # Brings marker forward when hovering
                         popup = ~paste("<b>Station:</b> ", site_friendly, "<br>", # popups appear when clicking
                                        "<b>Sampling start year:</b> ", minYear, "<br>",

modules/waterquality.R

@@ -13,31 +13,11 @@ library(tidyverse)
 # import all WQ data for this page (WIN and Guana spreadsheet)
 WQ_df <- readRDS("./03_Data_for_app/WQ_all.Rds")
 
-#### Process data further ####
-# MOVE THIS TO THE CLEANING SCRIPT!!
-# make datframe for map display and hover data
-WQ_data_locations = WQ_df %>% # site friendly and station code are the names in common
-  filter(variable %in% c("StationCode", "site_friendly",
-                         "geometry", 
-                      #"HUC12Name", # Changed for now (bc GTM WQ data does not have that variable) BUT we should also include a station name of some sort
-                       "SampleDate", # changed, from StartDate - but also not necessary for locations
-                       "Latitude",
-                       "Longitude",
-                       "data_source")
-         ) %>%
-  #select(c(RowID, variable, value)) %>% # not necessary
-  #distinct(RowID, variable, value) %>% # not necessary
-  pivot_wider(
-    names_from = variable,
-    values_from = value,
-    values_fill = list(value = NA)
-  ) %>%
-  distinct(Latitude, Longitude, data_source, geometry, StationCode, site_friendly) %>%
-  mutate(
-    Latitude = as.numeric(Latitude),
-    Longitude = as.numeric(Longitude)
-  )
+# Location data
+WQ_data_locations <- readRDS("./03_Data_for_app/WQ_data_locations.Rds")
 
+#### Process data further ####
+# Also move this to cleaning script
 WQ_data_units = WQ_df %>%
   filter(variable %in% c("ComponentLong", # some will be duplicated because they differ between the 2 sets
                          "Unit", "data_source")