Introduction to Spatial Data Programming with R

Michael Dorman

Preface

Last updated: 2020-08-12 00:35:24

0.1 What is R?

R is a programming language and environment, originally developed for statistical computing and graphics. As of October 2019, there are ~15,000 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 0.1–0.3).

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

Figure 0.2: IEEE Language Rankings 2019³

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).

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 2019, there are 194 packages⁶ specifically addressing spatial analysis in 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/)
2020(?)	Spatial Data Science (https://keen-swartz-3146c4.netlify.com/)

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, mainly software with graphical user interfaces, like 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, network routing, splitting lines at intersections, etc.)
• Data that cannot fit in RAM (but R can connect to spatial databases⁷)

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:

library(sf)
library(RPostgreSQL)
con = dbConnect(
  PostgreSQL(),
  dbname = "gisdb",
  host = "159.89.13.241",
  port = 5432,
  user = "geobgu",
  password = "*******"
)
dat = st_read(con, query = "SELECT name_lat, geometry FROM plants LIMIT 5;")

dat
##         name_lat           geometry
## 1    Iris haynei 35.67976, 32.77013
## 2    Iris haynei 35.65400, 32.74137
## 3 Iris atrofusca 35.19337, 31.44711
## 4 Iris atrofusca 35.18914, 31.51475
## 5  Iris vartanii 35.13970, 31.47415

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…

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)

Figure 0.7: Points on Great Circle

Figure 0.8: Visualizing Facebook Friends with geosphere⁸

0.2.6 gstat: Geostatistical Modelling

Univariate and multivariate geostatistics:

• Variogram modelling
• Ordinary and universal point or block (co)kriging
• Cross-validation

Figure 0.9: Predicted Zinc concentration, using Ordinary Kriging

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)

Figure 0.10: Neighbours list based on regions with contiguous boundaries

0.2.8 spatstat: Spatial point pattern analysis

Techniques for statistical analysis of spatial point patterns, such as:

• Kernel density estimation
• Detection of clustering using Ripley’s K-function
• Spatial logistic regression

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

0.2.9 osmdata: Access to OpenStreetMap data

Accessing OpenStreetMap (OSM) data using the Overpass API⁹

library(sf)
library(osmdata)
q = opq(bbox = "Beer-Sheva, Israel")
q = add_osm_feature(q, key = "highway")
dat = osmdata_sf(q)
lines = dat$osm_lines
pol = dat$osm_polygons
pol = st_cast(pol, "MULTILINESTRING")
pol = st_cast(pol, "LINESTRING")
lines = rbind(lines, pol)
lines = lines[, c("osm_id", "highway")]
lines = st_transform(lines, 32636)
plot(lines)

Figure 0.12: Beer-Sheva road network data from OpenStreetMap (OSM)

0.2.10 ggplot2, ggmap: Visualization

Figure 0.13: London cycle hire journeys with ggplot2¹⁰

Figure 0.14: Crime density by day with ggplot2

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:

• zcol="..." specifies the attribute used for symbology
• legend=TRUE adds a legend

library(sf)
library(mapview)
states = st_read("_book/data/USA_2_GADM_fips.shp")
mapview(states, zcol = "NAME_1", legend = TRUE)

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 (1^st ed. 2008, 2^nd ed. 2013)
• Hierarchical Modeling and Analysis for Spatial Data (1^st ed. 2003, 2^nd ed. 2014)
• An Introduction to R for Spatial Analysis and Mapping 1^st ed. 2015, 2^nd ed. 2018)
• Spatial Point Patterns: Methodology and Applications with R (2015)
• Displaying Time Series, Spatial, and Space-Time Data with R (1^st ed. 2014, 2^nd ed. 2018)
• Predictive Soil Mapping with R (2019)
• Geocomputation with R (2019)
• Spatial Data Science (2020?)

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]

0.3.3 Courses and tutorials

0.3.3.1 Courses

• GEOG 4/595: Geographic Data Analysis
• CP6521 Advanced GIS
• ES214 Introduction to GIS and Spatial Analysis
• GEOG 4/590: R for Earth-System Science
• GEOG 4/595: Geographic Data Analysis
• Spatial Data Science with R (Robert J. Hijmans)
• Introduction to Spatial Data Programming with R (this course)
• GISC 422 Spatial Analysis and Modelling
• CASA0005 Geographic Information Systems and Science
• Another list here

0.3.3.2 Tutorials

• Geospatial Data Science with R
• Data Carpentry Workshops
• GIS in R (Nick Eubank)
• NEON Data Tutorials
• Learn Spatial Analysis (University of Chicago)
• WUR Geoscripting
• Mapping in R
• Spatial Analysis notes
• Classifying Satellite Imagery in R
• Fundamentals of Spatial Analysis in R

0.3.3.3 Presentations

• UseR! 2019 Spatial workshop part I
• UseR! 2019 Spatial workshop part II
• UseR! 2017 Tutorial “Spatial Data in R: New Directions”

0.3.3.4 Official materials

• sf Vignettes
• stars Vignettes

---

1. Comprehensive R Archive Network↩

2. https://insights.stackoverflow.com/trends?tags=r↩

3. https://spectrum.ieee.org/computing/software/the-top-programming-languages-2019↩

4. https://www.nature.com/news/programming-tools-adventures-with-r-1.16609↩

5. http://www.maths.lancs.ac.uk/~rowlings/R/TaskViews/↩

6. https://cran.r-project.org/web/views/Spatial.html↩

7. https://cran.r-project.org/web/packages/sf/vignettes/sf2.html#reading_and_writing_directly_to_and_from_spatial_databases↩

8. http://paulbutler.org/archives/visualizing-facebook-friends/↩

9. http://wiki.openstreetmap.org/wiki/Overpass_API↩

10. http://spatial.ly/2012/02/great-maps-ggplot2/↩