The document provides an overview of R, a free software environment for statistical computing, detailing its capabilities, including statistical analysis, graphic generation, and package extensions. It outlines essential R functionalities, data types, and data structures, alongside practical guidance on importing, manipulating, and exporting both regular and spatial data. It emphasizes R's suitability for statistical modeling while noting its steep learning curve and the need for specific packages to work with spatial data.

1. Essentials of a
Bas Kempen
© ISRIC-World Soil Information, 2017. Reproduction or dissemination of the work as a whole or parts is not permitted without
consent of the author. Sale or placement on a website where payment must be made to access the document is strictly prohibited. https://creativecommons.org/licenses/by-nc-nd/4.0/

2. What is R?
• R is a free software environment for statistical computing and
graphics
• R provides a wide variety of statistical (linear and nonlinear
modelling, classical statistical tests, time-series analysis,
classification, clustering, …) and graphical techniques, and is highly
extensible.
– Base functionality (comes with R installation)
– Extension via ‘Packages’ (~6,700)
• www.r-project.org/
• www.r-tutor.com

3. Why R?
• It is free
• It runs on a variety of platforms
• Platform for (advanced) statistical data analyses
• State-of-the-art graphic capabilities
• Connects with other software (SAGA GIS, GE, Python)
• Very large user community on the web; lots of resources
• R has a steep learning curve
• Thousands of packages, not always easy to find what you are
looking for
• Sometimes cryptic error messages
• Not a GIS

4. Working with R

5. RStudio

6. First steps.....
• Tell R where to find and save your files: setting the
working directory using the ‘setwd’ command:
– setwd(“D:/Bas/SpringSchool/Rintro/workingdir”)
– setwd(“D:BasSpringSchoolRintroworkingdir”)
• Load packages that are required for your analyses:
– library(gstat)
– require(gstat)
(packages should be installed on your computer)
• Note: R is case-sensitive!

7. Setting up your session

8. R scripts and data objects
• R scripts are saved as “<name>.R” files
• R objects in the environment can be saved for future
use.
• Save the entire environment or only a couple of
objects.
• Objects are saved as “<name>.rda” or
“<name>.RDATA” files

9. Basic Data Types
• Numeric
• Integer
• Logical
• Character
• Factor
Vector
Function

10. Vectors
• Sequence of data elements of the same basic
type.
• Vectors can be combined.

11. Vector arithmetic
• Vectorized operations: most operations work on vectors with
the same syntax as they work on scalars (no need for looping)
• Vector arithmetic:
• Recycling of vector elements:

12. Other data structures
• Matrices
• Lists
• Data frames
• Data frame is the fundamental data structure
for statistical modelling in R.
• Data frame is a table with columns and rows
(fields and records).

13. Data frame
• Columns can have different data types
(numeric, integer, logical, character, factor)
• All columns must have the same length

14. Selecting subsets
• Selection is done with ‘[...]’
• Vector:
• Data frame:

15. Functions
• Data analyses and modelling is done through functions.
• These can be very simple:
• More complex functions have multiple arguments (inputs)
• Arguments have specific requirements
• Access help: ?fit.variogram

16. Plotting
• Large number of packages and functions for
generating plots with basic functionality to
‘high-level’: e.g. lattice and ggplot.
• The basic function for plotting is ‘plot’

17. ggplot (http://ggplot2.org/)
• library(ggplot2)
• Build your plot layer by layer
• Building blocks:
– geom: the geometric object that describes the
type of plot that is produced.
– aes: ‘aesthetics’, defines the visual properties of
the variables that are going to be plotted.
– scales: control the legend, plot layout

18. Plotting with ggplot

19. Plotting with ggplot

20. Plotting with ggplot

21. Plotting with ggplot

22. Importing data
• Importing from tables:
– csv: read.csv()
– txt: read.table()
– xlxs: read.xlsx() [requires package ‘xlsx’]
• Data is imported as a data.frame

23. Exporting data
• Variety of exporting formats for tabular data:
• Saving plots

24. Working with spatial data in R

25. Spatial Data in R
• R offers a wide variety of packages and tools that
can handle spatial data.
• Note: R is not a GIS.
• R is not so memory efficient.
• Relevant packages:
– sp: handling spatial data
– raster: reading/manipulating/writing spatial raster
data
– rgdal: reading/writing spatial data
– maptools: reading/manipulating/writing spatial
polygon data (not maintained anymore)

26. Spatial data classes and formats
• Vector: points, lines and polygons (areal).
• For storing data that has discrete boundaries, such as
country borders, land parcels, and streets.
• Format: shapefile

27. Spatial data classes and formats
• Raster: surface divided into a regular grid of cells.
• For storing data that varies continuously, as in a
satellite image, a surface of chemical concentrations,
or an elevation surface.
• Format: GeoTiff (allows embedding spatial reference
information, metadata and color legends. It also
supports internal compression)
• (Ascii, ESRI Grid)

28. Structures for spatial data
• Spatial data is nothing more than a data frame
that has columns with X and Y coordinates.
• Example:
• Let’s now take a look at R classes for spatial data
(sp package)

29. Spatial data classes I
• Convert a data frame to a SpatialPointsDataFrame
object with the coordinates function.

30. Spatial data classes I
• Data frame with data points at regular
intervals.

31. Spatial data classes II
• Create a grid object with the gridded function:
SpatialPixelsDataFrame

32. Spatial data classes III
• Create a full grid object with the fullgrid function:
SpatialGridDataFrame

33. Spatial data classes IV
• The sp class for polygon data is the
SpatialPolygonsDataFrame.

34. Spatial data classes VI
• The raster package comes with its own class
for raster data: RasterLayer.

35. Importing spatial data
• rgdal: readOGR (vector), readGDAL (raster)
• maptools: readShapePoly, readShapePoints,
readShapeLines (functions not maintained anymore)
• raster: raster

36. Projections
• Once you have loaded your spatial data in R,
you have to tell R its geographic projection.
• Check the current projection: proj4string
function.
• Setting a projection: CRS function.
• Reprojecting to another coordinate system:
spTransform function.

37. Projections

38. Projections

39. Plotting
• sp package: spplot

40. Plotting

41. ggplot I

42. ggplot II

43. ggplot III

44. Interactive maps
• Interactive maps can be generated with the
leaflet package.

45. Exporting spatial data
• rgdal: writeOGR (vector), writeGDAL (raster)
• maptools: writePolyShape, writePointsShape
• raster: writeRaster

46. Now lets practice. Have fun!!