R for the curious but terrified

What is R?

- a Statistical Programming Language

• Syntax
• Code
• Statistical analysis

What is R?

- a Data Science Toolkit

• Analysis
• Visualisation
• Data wrangling
• Data access
• Reproducibility
• Publication and sharing

A few examples

Map the distribution of species x over time using NBN atlas data

Get the data from the NBN atlas API (aka webservice)

library(needs)
needs(tidyverse, jsonlite, sf, mapview, tictoc, ggspatial)


search <- "https://records-ws.nbnatlas.org/occurrences/search?q=*:*&fq=genus:Falco&lat=51.5&lon=0.127&radius=300&pageSize=30000" # Set the base URL for the NBN Atlas web service.

 
tic()
df <- fromJSON(search, simplifyDataFrame = TRUE)
toc() 

40.498 sec elapsed

Trend plot

## plot trends

df$occurrences |>
  count(year, vernacularName) |>
  filter(year > 1999, 
         n > 4) |>
  ggplot() +
  geom_col(aes(year, n)) +
  facet_wrap(~vernacularName, scales = "free_y")

Interactive map

df$occurrences |>
  filter(year > 1999) |>
  sample_frac(.1) |>
  st_as_sf(coords = c("decimalLongitude", "decimalLatitude"), crs = 4326) |>
  mapview(zcol = "vernacularName")

Static maps

df$occurrences |>
  filter(year > 1999) |>
  sample_frac(.25) |>
  mutate(month = as.numeric(month), 
         season = case_when(between(month, 4, 9) ~ "summer", 
                            TRUE ~ "winter")) |>
  select(season, vernacularName, year, contains("decimal")) |>
  st_as_sf(coords = c("decimalLongitude", "decimalLatitude"), crs = 4326) |>
  ggplot() +
  ggspatial::annotation_map_tile() +
  geom_sf(aes(colour = season), size = 0.4) +
  facet_wrap(~vernacularName, nrow = 2) +
  theme_void()