GIS and Mapping with R

The course will cover mapping and spatial analyses using R. Most of the features of Arcgis or other GIS software are available in R: some are missing, but R also opens more possibilities by making full data accessible. The main topics will be the use of shape files and raster maps, including the creation of maps as well as spatial analyses.

The class will start with a two-week review of basic R. Participants who have reasonable experience with R might decide to skip these. These will cover data types in R, use of dataframes, and basic graphing functions in base R and how they are used for creating maps.

During each session, I will explain methods and present examples of their use, then students will work on assignments using these methods. Datasets will be provided, but students are encouraged to bring their own geographic data as well. A course web site will provide sample code, data, and a list of key R functions.

Students should

have their own computer to work on during the sessions,
have R and several packages already installed and working,
know their computer's file system (ie, how to find files by typing the path plus file name),

Schedule

Time: Each Wednesday, 11:00-1:00PM (except 27 Sept, when it will run noon-2:PM)
Dates: 27 September - 15 November
Where: Conference room

Software required

R base package from http://cran.us.r-project.org/
A programming editor such as Geany (NOT Wordpad NOT Notepad)
R contributed packages, available at http://cran.r-project.org/
- gdata (for 27 Sept)
- sp (for GIS sections)
- maptools
- spatstat
- splancs
- raster

Web site at http://richardcondit.org/workshops/Rgis/MortonArboretum including:

An outline http://richardcondit.org/workshops/Rgis/MortonArboretum
Sample R and csv datasets http://richardcondit.org/workshops/Rgis/data
Assignments http://richardcondit.org/workshops/Rgis/MortonArboretum/assignments.php
R scripts
History of R commands I enter during the course

Each will be updated regularly throughout the course.

Course syllabus

Basic R: Data handling [week 1]

Data types
1. Numeric
2. Character (surrounded by quotes)
3. Boolean
Objects
1. Atomic (or scalar): a single value
2. Vector: many values
3. Dataframe: table
4. * List [*not necessary for basics]
5. * Matrix, Array [*not necessary for basics]
Assigning objects and variables
Vectorizing calculations
Filtering
Installing packages
Reading external data

Programming R [week 2]

Writing functions

Define a function: source
Function arguments, defaults
Declaring empty vectors
Loops
Debugging
Return values
Scope of variables

Saving R objects

Maps with R's base package [week 3]

Basic R graphing
Plotting a dataframe with a line
Basic cosmetics

Color: col
Point character: pch
Point size: cex
Axis: xlab, ylab, axes

Structuring multiple polygons in a list
Polygon vs. line
UTM coordinates

GIS tools in R [week 4]

Packages sp, maptools, rgdal
Classes (type S4)
1. SpatialPolygonsDataFrame
2. SpatialPolygons
3. Polygons
4. Polygon
5. Etc. for Lines, Points
Subsetting the data
Bounding box
Maps
Area

GIS rasters in R [week 5]

World elevation maps
Package raster
- merge
- plot
- crop
- zoom
- contour
Extracting the data

Spatial analyses [week 6]

Packages splancs, sgeostat, spatstat
Coordinate transformation
Distance to point or line

Spatial analyses [week 7]

Coordinate transformation (review)
Point in polygon (review)
Distance to point or line
Nearest neighbor

GPSVisualizer [week 8]

Create maps with world coordinates and GPS Visualizer
Show the map on Google
Create R spatial classes

Principal R functions

Basic R
- table or vector size: length, dim, nrow, ncol
- view a few rows: head, tail
- create vectors: c, seq, rep, sample, sample.int
- declare vectors: numeric, character
- load data: read.delim, read.table, load, attach
- save work: write.table, save, savehistory
Data exploration
- math: log, exp, sin, cos, etc
- stats: sum, mean, median, sd, var, range, min, max
- unique values: unique
- filter: subset
- calculations on table: apply, rowSums, colSums, etc
- grouping: table, tapply
- random numbers: runif, rnorm, rbinom, etc
Writing functions
- execute from file: source
- definition: function
- looping: for
- finishing: return
- debugging: debug, browser
Basic mapping
- start a graph: plot
- add points to graph: points
- add lines to graph: line
- fill a polygon: polygon
- open a new graphics window: X11
- switch between graphics windows: dev.set
- add a straight line: abline
- add all axes without labels: box
- add one axis with specified ticks and labels: axis(at=__,lab=__)
- label points: identify
- set graphics parameters: par
- add legend: legend with many options (see help)
- add text: text, mtext
- arguments to plot
  1. point type: pch
  2. point size: cex
  3. color: col
  4. line thickness: lwd
  5. line type: lty 'dashed' or 'solid' or 'dotted'
  6. without axes: axes=FALSE
  7. title at top: main='yourtitle'
- arguments to par (look at par help for many more)
  1. multiple plots in a grid: mfcol
  2. plot size in inches: pin
  3. label and axis sizes: cex.lab, cex.axis
  4. tighten label and axis: mgp
  5. extend labels or lines off graph: xpd
R GIS
- read shape files: readShapePoly, readShapeLines, readShapePoints
- mapping: sp::plot
- subsetting subset
- create class Polygons: Polygons
- create class Polygon Polygon
- mapping: sp::plot
- set projection CRS (requires sp)
- change projection spTransform (requires sp)
R Raster
- read geotiff files: raster
- graphing plot, contour
- define a subrectangle: extent
- extract the extent: zoom, crop
- get the data: xyFromCell
- change projection: projectRaster
Spatial analyses
- Distances: nndistF, dsquare
- Points within polygon point.in.polygon, inout, crop
Creating a polygon and spatial classes
- One polygon: Polygon etc for Lines, Points
- Several polygons Polygons etc
- The full spatial class SpatialPolygons,SpatialPolygonsDataFrame etc

GIS and Mapping with R

Richard Condit -- Morton Arboretum -- Fall 2017