CreateASpeciesList.R

# Create a species list for eBird checklist hotspots for a specific location.
## e.g. eBird data for Andalucia, Spain from Jan 2016 - Sept 2022. 

# Download  eBird data.
## Log into ebird.org to download the data
## The dataset will be emailed to you. When it is, download the zip file.
## Create a folder on your computer and move the file into this. This folder will be the working directory in R.


# In R Studio Desktop, install and run packages
install.packages("tidyverse")
library(tidyverse)


# Set your working directory
getwd() # See which directory you are currently in
setwd("C:/Users/filepath") # To set the working directory, copy the pathway into the setwd() function. Be sure to use "" and change \ -> /



# Import data 
# eBird files can come in .csv or .txt.

# To read in .csv
data <-read.csv("filepath.csv") 

# To read in .txt
data<-read.delim("filename.txt", sep = "\t", 
                header = TRUE, quote = "", 
                stringsAsFactors = FALSe, na.strings = c(""," ", NA))




# Preliminary exploration of the data

names(data)   # View column names
head(data)    # View first six rows
str(data)     # View vector types for columns e.g. character, numeric, integer
unique(data$COUNTY) #View all Counties listed. unique(data$LOCALITY)




# Process the data

# Filter by COUNTY  
data <- data |>
  filter(COUNTY %in% c("Cádiz", "Málaga"))

# Remove X in COUNT column
# View Counts with X
data[data$OBSERVATION.COUNT == "X",]

# Remove rows where Count = X
data<- data[data$OBSERVATION.COUNT != "X",]



# Create a smaller dataset by defining the data you want to analyse
data <- data |>
  select(GLOBAL.UNIQUE.IDENTIFIER, COMMON.NAME, SCIENTIFIC.NAME, 
         OBSERVATION.COUNT, COUNTY, LOCALITY, OBSERVATION.DATE,
         TIME.OBSERVATIONS.STARTED, SAMPLING.EVENT.IDENTIFIER, PROTOCOL.TYPE,
         DURATION.MINUTES, EFFORT.DISTANCE.KM, NUMBER.OBSERVERS, 
         ALL.SPECIES.REPORTED, GROUP.IDENTIFIER, TRIP.COMMENTS, SPECIES.COMMENTS)


# Filter by date. First separate Year, Month, and Day into their own columns

library(lubridate)

data <- data |>
  separate(OBSERVATION.DATE, c("Year", "Month", "Day"), "-")  

# Filter for all observations in a specific month 
data <-data |>
  filter(Month == "09")

# Filter by complete checklists
data <-data |>
  filter(ALL.SPECIES.REPORTED == 1)


# Convert the column Count to a numeric vector
data$OBSERVATION.COUNT<-as.numeric(data$OBSERVATION.COUNT) 
# If you get the error "NAs introduced by coercion", you can ignore.


str(data) # To check it has been converted


# Rename columns
data<-data |>
  rename(Duration = DURATION.MINUTES, Distance = EFFORT.DISTANCE.KM, 
         Count = OBSERVATION.COUNT)

names(data) # Check they are renamed


# Remove duplicated observations 
# data |>
 # distinct(GROUP.IDENTIFIER, .keep_all = TRUE)


# Create and save the smaller data file from the new data frame. Can delete the larger one.
write.csv(data, "filename.csv", row.names = FALSE)



# Create lists

# Create a species list
Species <- data |>
  distinct(COMMON.NAME)


# Identify checklist localities
data |>
  distinct(LOCALITY)

# Create a species list for a specific checklist locality
Guadalhorce <- data |>
  filter(LOCALITY == "Desembocadura del Guadalhorce--Área General")

Guadalhorce <- Guadalhorce |> 
  distinct(COMMON.NAME, SCIENTIFIC.NAME)

# Create a list of species for an area with multiple checklist localities
Ronda<-data |>
  filter(LOCALITY %in% c("Sierra de Grazalema PNat--Llanos de Libar, Área General","Ronda--Pueblo" ))

# Create a species list that shows the species for each locality in the area
Ronda<-Ronda |>
  distinct(COMMON.NAME, LOCALITY)


# Create a species list that shows which checklist localities each species can be seen
Trip<-data |>
  filter(LOCALITY %in% c("Desembocadura del Guadalhorce--Área General",
                         "Casares--Castillo",
                         "Sierra de Grazalema PNat--Llanos de Libar, Área General",
                         "Ronda--Pueblo",
                         "La Janda--Área General",
                         "Mirador del Estrecho",
                         "El Estrecho PNat--Punta Camorro",
                         "El Estrecho PNat--Isla de Las Palomas",
                         "El Estrecho PNat--Los Lances",
                         "Algeciras--Parque María Cristina",
                         "Doñana PNat--Salinas de Bonanza",
                         "Chipiona--Puerto",
                         "Lagunas de Camino Colorado",
                         "La Breña y Marismas de Barbate PNat--Área General",
                         "El Palmar--Campiña",
                         "Los Alcornocales PNat--Vega del Puente")) %>% 
  arrange(LOCALITY)

# Change the order of columns to show which species can be seen at each locality
Locations <- Trip |>
  relocate(COMMON.NAME, .after = LOCALITY)


# Save the checklists
write.csv(data, "filename.csv", row.names = FALSE)



# Analyse the data

# Find which year(s) have seen the most species
Guadalhorce |>
  group_by(Year) |>
  summarise(species = n_distinct(COMMON.NAME))
  

# If the data cover multiple months, find which months have the most species
Guadalhorce |>
  group_by(Month) |>
  summarise(species = n_distinct(COMMON.NAME))




# Look at total number of species seen in each week
# need to create weeks

Week <-Ronda |>
  filter(Day %in% 1:7) # I will only look at the first week

Tot <-Week |>
  group_by(Year, Day) |>
  summarise(count = n_distinct(COMMON.NAME)) |>
  ungroup()
  
ggplot(Tot, aes(x= Day, y = count))+
  geom_bar(stat = "identity")+
  facet_wrap(~Year)+
  labs(title = "Unique species observations across September",
       x= "Day of month",
       y= "Number of species")+
  theme_bw()