Skip to content

Latest commit

 

History

History
231 lines (172 loc) · 7.76 KB

setup.md

File metadata and controls

231 lines (172 loc) · 7.76 KB
title subtitle author date output
GBIF data access with R - Session 1 - R settings
R workshop by GBIF.no and ForBio
February 18-19, 2018
html_document
keep_md toc toc_depth
true
true
3

You are here: R workshop >> Session 1 Intro >> R settings

GBIF data access with R - Session 1 - R settings

Scientific reuse of openly published biodiversity information: Programmatic access to and analysis of primary biodiversity information using R. Nordic Oikos 2018, pre-conference R workshop, 18th and 19th February 2018. Further information here.

Session 1 gives an introduction to GBIF, use of GBIF data and the GBIF API. This script presents some useful settings, in particular how to install some of the R-packages we will use during the workshop. For more information see session 2.

  • Notice that most Rmd chuncks have settings eval=FALSE, which will exclude execution of this chunk in knitr, but enable manual execution in RStudio.
  • Notice that some of the R-commands below are commented out (with a hash #). To execute selected lines on your computer, you need to uncomment (remove the hash #) first. You may add comments inside chucks (or in R scripts) using a hash at the start of lines, or in the middle to make the text after the hash a comment.

R Markdown is a convinient format (R Notebooks) for writing R-scripts, introduced in 2014. R Markdown integrates very well in RStudio (and can also be used form the command line). Session 2 will provide an introduction to R-scripting including R Markdown.

## Setting the working directory, here: to the same directory as the RMD-script. 
## Primarily useful "outside" of RMarkup chunks.
## RMarkup (Rmd) seems to always change current working directory to the same as the Rmd-script.
require(rstudioapi)
setwd(dirname(rstudioapi::getActiveDocumentContext()$path))
getwd() ## will display the working directory
[1] "/Users/dag/workspace/gitHub/nordic_oikos_2018_r/s1_gbif_intro"

Create a folder for demo data

## Note however, that RMarkdown with knitr works best with all files in the same directory.
#dir.create(file.path("./demo_data")) ## create a folder for demo files (inside working directory)
#dir.create(file.path("../demo_data")) ## create a folder for demo files (next to working directory)
#dir.create(file.path("..", "demo_data")) # will work better accross OS? on Windows?

Some useful R-packages to install, that we will use during the workshop

## It is possible to install R-packages from the script
install.packages("rmarkdown") # R Markdown
## rOpenSci
install.packages("rgbif") # species occurrence data from GBIF
install.packages("mapr", dependencies = TRUE) # rOpenSci mapping tools
install.packages("spocc", dependencies = TRUE) # species occurrence, more than GBIF
## Mapping tools
install.packages("maps") # mapping tools
install.packages("mapproj") # map projection tools
install.packages("rgdal") # r gdal tools
install.packages('mapdata') # map data
## Dismo and Raster by R Hijmans
install.packages("dismo") # species occurrence data from GBIF with dismo
install.packages("raster") # raster data tools
## Plotting tools
install.packages("plotly")
install.packages("RColorBrewer")
## Data management
install.packages("tidyverse") ## tidyverse
install.packages("readxl") ## Read Excel, from tidyverse (without external Java dependencies)
## If you need Java-based packages (such as eg. xlsx)
## Sys.setenv(JAVA_HOME="PATH-TO-JAVA_HOME")
#Sys.setenv(JAVA_HOME='/Library/Java/Home') ## or use '/usr/libexec/java_home -d 64'
#install.packages("rJava")
#install.packages("xlsx")
##

Load R-packages

## Use functions library() or require() to load the R-packages you need
##
## rOpenSci
require('rgbif') # r-package for accessing GBIF
library('mapr') # rOpenSci r-package for mapping (occurrence data)
library('spocc') # rOpenSci r-package with more biodiversity data sources than GBIF
## Mapping tools
require('raster') # spatial raster data management
library('dismo') # spatial data and distribution modelling tools
#library(maps)
#library(mapproj)
library(mapdata)
#library(maptools) # mapping tools for spatial objects
#library(rgdal) # provides the Geospatial Data Abstraction Library
## Data mangement
#library(gapminder)
library(tidyverse) ## includes dplyr etc for handling data, select(), filter()
library('plyr') ##
## Plotting tools
#library('RColorBrewer')

Clear workspace (be careful)

## Large arrays and dataframes can consume huge parts of your working memory,
## and you probably want to remove some of them from memory when they are not needed anymore.
#rm(list = ls()) # Be careful, this line cleans the entire R workspace
#rm(sp) # Probably better to remove large arrays and dataframes individually

NOTES

## BOUNDING BOX
##bb <- c(lowerLon, lowerLat, higherLon, higherLat) # bounding box
bb <- c(10.2,63.3,10.6,63.5) # Trondheim
bb <- c(5.25, 60.3, 5.4, 60.4) # Bergen
bb <- c(18.7, 69.6, 19.2, 69.8) # Tromsoe
bb <- c(10.6, 59.9, 10.9, 60.0) # Oslo
bb <- c(4.5, 54.9, 31.0, 71.0) # Scandinavia = lon(4.5, 31.0), lon(55, 71)
##
#bbox(norway_mask) ## use bbox(spatialObj) to find the bounding box of a spatial object

Norway: min max x 4.6280575 31.078054 y 58.0702763 71.113054

## Print text and variables
message(sprintf("Current working dir: %s\n", getwd()))
paste("Today is", date())
sprintf(5.55555, fmt = '%#.3f') ## 5.556
sprintf(5.55555, fmt = '%#.3g') ## 5.56
library('plyr') ## r-pkg plyr for data handling
spp_df <- ldply(spp) ## ldply - split list, apply function, return dataframe (here list to df)

List to dataframe

#spp <- occ_search(taxonKey=3073, limit=100, return='data', country='NO', hasCoordinate=TRUE)
library('plyr') ## r-pkg plyr for 
spp_df <- ldply(spp) ## ldply - split list, apply function, return dataframe (here list to df)
#spp_m <- spp_df[c("name", "decimalLongitude","decimalLatitude", "basisOfRecord", "year", "municipality")]
#map_leaflet(spp_m, "decimalLongitude", "decimalLatitude", size=3, color=cols)
## coordinate reference system
crs(r) <- "+proj=longlat +datum=WGS84 +no_defs +ellps=WGS84 +towgs84=0,0,0" ## set CRS
crs(r) <- CRS('+init=EPSG:4326') ## same result, but using EPSG is shorter to write, and easier

Read.xlsx depend on JAVA which might be an obstacle

##Sys.setenv(JAVA_HOME="PATH-TO-JAVA_HOME")
#Sys.setenv(JAVA_HOME='/Library/Java/Home') ## or use '/usr/libexec/java_home -d 64'
#Sys.setenv(JAVA_HOME='/Library/Java/JavaVirtualMachines/jdk1.7.0_45.jdk/Contents/Home')
library(xlsx) ## first row contains variable names
spp_dq <- read.xlsx("./demo_data/spp_dq.xlsx", 1) ## manipulated occurrence data, introduced errors
#spp_da <- read.xlsx("./demo_data/spp_dq.xlsx", sheetName = "spp_dq") ## alt: choose named worksheet
head(spp_da, n=5)

Diverse color palettes

library(RColorBrewer)
#display.brewer.all()
display.brewer.pal(n=9, name='Set1')

colorBrewer Set1

Read more about colors at the [https://www.r-bloggers.com/palettes-in-r/](R-bloggers story about color palettes in R)

Navigate back to GitHub project home or GitHub.io html pages.