Preface

Last updated: 2020-10-26 15:55:30

0.1 What is R?

R is a programming language and environment, originally developed for statistical computing and graphics. As of October 2020, there are ~16,000 R packages in the official repository CRAN1.

Notable advantages of R are that it is a full-featured programming language, yet customized for working with data, relatively simple and has a huge collection of over 100,000 functions from various areas of interest.

R’s popularity has been steadily increasing in recent years (Figures 0.10.3).

Stack Overflow Trend for the 'r' question tag^[https://insights.stackoverflow.com/trends?tags=r]

Figure 0.1: Stack Overflow Trend for the ‘r’ question tag2

IEEE Language Rankings 2019^[https://spectrum.ieee.org/computing/software/the-top-programming-languages-2019]

Figure 0.2: IEEE Language Rankings 20193

Proportion of research papers citing R^[https://www.nature.com/news/programming-tools-adventures-with-r-1.16609]

Figure 0.3: Proportion of research papers citing R4

A brief overview of the capabilities and packages for several domains of R use, are available in the “CRAN Task Views” (Figure 0.4).

CRAN Task Views^[http://www.maths.lancs.ac.uk/~rowlings/R/TaskViews/]

Figure 0.4: CRAN Task Views5

0.2 R and analysis of spatial data

0.2.1 Introduction

Over time, there was an increasing number of contributed packages for handling and analyzing spatial data in R. Today, spatial analysis is a major functionality in R. As of October 2020, there are at least 185 packages6 specifically addressing spatial analysis in R.

Books on Spatial Data Analysis with R

Figure 0.5: Books on Spatial Data Analysis with R

Some important events in the history of spatial analysis support in R are summarized in Table 0.1.

Table 0.1: Significant events in the history of R-spatial
Year Event
pre-2003 Variable and incomplete approaches (MASS, spatstat, maptools, geoR, splancs, gstat, …)
2003 Consensus that a package defining standard data structures should be useful; rgdal released on CRAN
2005 sp released on CRAN; sp support in rgdal
2008 Applied Spatial Data Analysis with R, 1st ed.
2010 raster released on CRAN
2011 rgeos released on CRAN
2013 Applied Spatial Data Analysis with R, 2nd ed.
2016 sf released on CRAN
2018 stars released on CRAN
2019 Geocomputation with R (https://geocompr.robinlovelace.net/)
2021(?) Spatial Data Science (https://www.r-spatial.org/book/)

The question that arises here is: can R be used as a Geographic Information System (GIS), or as a comprehensive toolbox for doing spatial analysis? The answer is definitely yes. Moreover, R has some important advantages over traditional approaches to GIS, i.e., software with graphical user interfaces such as ArcGIS or QGIS.

General advantages of Command Line Interface (CLI) software include:

  • Automation—Doing otherwise unfeasible repetitive tasks
  • Reproducibility—Precise control of instructions to the computer

Moreover, specific strengths of R as a GIS are:

  • R capabilities in data processing and visualization, combined with dedicated packages for spatial data
  • A single environment encompassing all analysis aspects—acquiring data, computation, statistics, visualization, Web, etc.

Nevertheless, there are situations when other tools are needed:

  • Interactive editing or georeferencing (but see mapedit package)
  • Unique GIS algorithms (3D analysis, label placement, splitting lines at intersections)
  • Data that cannot fit in RAM (but R can connect to spatial databases7 and other softwere for working with big data)

The following sections (0.2.20.2.11) highlight some of the capabilities of spatial data analysis packages in R, through short examples. We are going to elaborate on most of these packages later on in the book, and many of those examples will become clear.

0.2.2 Input and output of spatial data

Reading spatial layers from a file into an R data structure, or writing the R data structure into a file, are handled by external libraries:

  • GDAL/OGR is used for reading/writing vector and raster files, with sf and stars
  • PROJ is used for handling CRS, in both sf and stars
  • Working with specialized formats, e.g., HDF with gdalUtils or NetCDF with ncdf4

Package sf combined with RPostgreSQL can be used to read from, and write to, a PostGIS spatial database:

0.2.3 sf: Processing Vector Layers

GEOS is used for geometric operations on vector layers with sf:

  • Numeric operators—Area, Length, Distance…
  • Logical operators—Contains, Within, Within distance, Crosses, Overlaps, Equals, Intersects, Disjoint, Touches…
  • Geometry generating operators—Centroid, Buffer, Intersection, Union, Difference, Convex-Hull, Simplification…
Buffer function

Figure 0.6: Buffer function

0.2.4 stars: Processing Rasters

Geometric operations on rasters can be done with package stars:

  • Accessing cell values—As matrix / array, Extracting to points / lines / polygons
  • Raster algebra—Arithmetic (+, -, …), Math (sqrt, log10, …), logical (!, ==, >, …), summary (mean, max, …), Masking
  • Changing resolution and extent—Cropping, Mosaic, Resampling, Reprojection
  • Transformations—Raster <-> Points / Contour lines / Polygons

0.2.5 geosphere: Geometric calculations on longitude/latitude

Package geosphere implements spherical geometry functions for distance- and direction-related calculations on geographic coordinates (lon-lat).

Points on Great Circle

Figure 0.7: Points on Great Circle

Visualizing Facebook Friends with `geosphere`^[http://paulbutler.org/archives/visualizing-facebook-friends/]

Figure 0.8: Visualizing Facebook Friends with geosphere8

0.2.6 gstat: Geostatistical Modelling

As mentioned above, R was initially developed for statistical computing (Section 0.1). Accordingly, there is an extensive set of R packages for spatial statistics. For example, package gstat provides a comprehensive set of functions for univariate and multivariate geostatistics, mainly for the purpose of spatial interpolation:

  • Variogram modelling
  • Ordinary and universal point or block (co)kriging
  • Cross-validation
Predicted Zinc concentration, using Ordinary Kriging

Figure 0.9: Predicted Zinc concentration, using Ordinary Kriging

We are going to learn about the gstat package in Chapter 12. An introduction to the package can also be found in Chapter 8 of Applied Spatial Data Analysis with R (Bivand, Pebesma, and Gomez-Rubio 2013).

0.2.7 spdep: Spatial dependence modelling

Modelling with spatial weights:

  • Building neighbor lists and spatial weights
  • Tests for spatial autocorrelation for areal data (e.g., Moran’s I)
  • Spatial regression models (e.g., SAR, CAR)
Neighbours list based on regions with contiguous boundaries

Figure 0.10: Neighbours list based on regions with contiguous boundaries

The spdep package is beyond the scope of this book. An introduction to the package can be found in Chapter 9 of Applied Spatial Data Analysis with R (Bivand, Pebesma, and Gomez-Rubio 2013).

0.2.8 spatstat: Spatial point pattern analysis

Package spatstat provides a comprehensive collection of techniques for statistical analysis of spatial point patterns, such as:

  • Kernel density estimation
  • Detection of clustering using Ripley’s K-function
  • Spatial logistic regression
Distance map for the Biological Cells point pattern dataset

Figure 0.11: Distance map for the Biological Cells point pattern dataset

The book Spatial point patterns: methodology and applications with R (Baddeley, Rubak, and Turner 2015) provides a thorough introduction to the subject of point pattern analysis using the spatstat package. A more brief introduction can also be found in Chapter 7 of Applied Spatial Data Analysis with R (Bivand, Pebesma, and Gomez-Rubio 2013).

0.2.9 osmdata: Access to OpenStreetMap data

Package osmdata gives access to OpenStreetMap (OSM) data—the most extensive open-source map database in the worls—using the Overpass API9.

Beer-Sheva road types map, using data downloaded from OpenStreetMap (OSM)

Figure 0.12: Beer-Sheva road types map, using data downloaded from OpenStreetMap (OSM)

0.2.10 ggplot2: Visualization

The ggplot2 package is one of the most popular packages in R. It provides advanced visualization methods through a well-designed and consistent syntax. The package supports visualization of both vector layers10 and rasters11.

The ggplot2 package is highly customizable and capable of producing publication-quality figures and maps as well as original and innovative designs (Figure 0.13). One of its strengths is in easy preparation of “small-multiple”—or facet, in the terminology of ggplot2—figures (Figure 0.14).

London cycle hire journeys with `ggplot2`^[http://spatial.ly/2012/02/great-maps-ggplot2/]

Figure 0.13: London cycle hire journeys with ggplot212

Crime density by day with `ggplot2`

Figure 0.14: Crime density by day with ggplot2

The ggplot2 package is beyond the scope of this book. A good place to start is the book ggplot2: Elegant Graphics for Data Analysis, by package author (Wickham 2016). The book is available online13.

0.2.11 leaflet, mapview: Web mapping

Packages leaflet and mapview provide methods to produce interactive maps using the Leaflet JavaScript library.

Package leaflet gives more low-level control. Package mapview is a wrapper around leaflet, automating addition of useful features:

  • Commonly used basemaps
  • Color scales and legends
  • Labels
  • Popups

Function mapview produces an interactive map given a spatial object. The zcol parameter is used to specify the attribute used for symbology:

Intractive map made with `mapview`

Figure 0.15: Intractive map made with mapview

0.3 Other materials

0.3.1 Books

  • Model-based Geostatistics (2007)
  • A Practical Guide to Geostatistical Mapping (2009)
  • Spatial Data Analysis in Ecology and Agriculture using R (2012)
  • Learning R for Geospatial Analysis (2014)
  • Applied Spatial Data Analysis with R (1st ed. 2008, 2nd ed. 2013)
  • Hierarchical Modeling and Analysis for Spatial Data (1st ed. 2003, 2nd ed. 2014)
  • An Introduction to R for Spatial Analysis and Mapping 1st ed. 2015, 2nd ed. 2018)
  • Spatial Point Patterns: Methodology and Applications with R (2015)
  • Displaying Time Series, Spatial, and Space-Time Data with R (1st ed. 2014, 2nd ed. 2018)
  • Predictive Soil Mapping with R (2019)
  • Geocomputation with R (2019)
  • Spatial Data Science (2021?)

0.3.2 Papers

  • Pebesma, E., Bivand, R. S. (2005). Classes and Methods for Spatial Data: the sp Package. R news, 5(2), 9-13. [PDF]
  • Pebesma, E. (2018). Simple Features for R: Standardized Support for Spatial Vector Data. The R Journal, 10(1):439-446. [PDF]

References

Baddeley, Adrian, Ege Rubak, and Rolf Turner. 2015. Spatial Point Patterns: Methodology and Applications with R. CRC press.

Bivand, Roger S., Edzer Pebesma, and Virgilio Gomez-Rubio. 2013. Applied Spatial Data Analysis with R, Second Edition. Springer, NY. https://asdar-book.org/.

Wickham, Hadley. 2016. Ggplot2: Elegant Graphics for Data Analysis. springer. https://ggplot2-book.org/.