R: creating a map of selected Canadian provinces and U.S. states

Question

I am attempting to create a map of selected Canadian provinces/territories and selected U.S. states. So far the nicest maps appear to be those generated with GADM data: http://www.gadm.org/

However, I have not been able to plot the U.S. and Canada on the same map or plot only selected provinces/territories and states. For example, I am interested in Alaska, Yukon, NWT, British Columbia, Alberta, and Montana among others.

Also, the U.S. map appears to be split along the international dateline.

Can someone please help me to:

1. plot the aforementioned provinces/territories and states on a single map
2. avoid having the U.S. split along the International dateline
3. overlay a latitude-longitude grid
4. select a specific projection, maybe the polyconic.

Maybe spplot does not allow users to specify projections. I did not see an option to select a projection on the spplot help page. I know how to select projections with the map function in the maps package but those maps did not appear to look as nice and I could not plot the desired subset of provinces/territories and states with that function either.

I do not know how to begin adding a latitude-longitude grid. However, Section 3.2 of the file 'sp.pdf' seems to address the topic.

Below is the code I have come up with so far. I have loaded every map-related package I have stumbled upon and commented out GADM data except for provincial/territorial or state boundaries.

Unfortunately, so far I have only managed to plot maps of Canada or the U.S.

library(maps)
library(mapproj)
library(mapdata)
library(rgeos)
library(maptools)
library(sp)
library(raster)
library(rgdal)

# can0<-getData('GADM', country="CAN", level=0) # Canada
  can1<-getData('GADM', country="CAN", level=1) # provinces
# can2<-getData('GADM', country="CAN", level=2) # counties

plot(can1)    
spplot(can1, "NAME_1") # colors the provinces and provides
                       # a color-coded legend for them
can1$NAME_1            # returns names of provinces/territories
# us0 <- getData('GADM', country="USA", level=0)
  us1 <- getData('GADM', country="USA", level=1)
# us2 <- getData('GADM', country="USA", level=2)
plot(us1)              # state boundaries split at 
                       # the dateline
us1$NAME_1             # returns names of the states + DC
spplot(us1, "ID_1")
spplot(us1, "NAME_1")  # color codes states and
                       # provides their names
#
# Here attempting unsuccessfully to combine U.S. and Canada on one map.
# Attempts at selecting given states or provinces have been unsuccessful.
#
plot(us1,can1)
us.can1 <- rbind(us1,can1)

Thanks for any help. So far I have made no progress with Steps 2 - 4 above. Perhaps I am asking for too much. Perhaps I should simply switch to ArcGIS and try that software.

I have read this StackOverflow post:

Can R be used for GIS?

EDIT

I have now borrowed an electronic copy of 'Applied Spatial Data Analysis with R' Bevand et al. (2008) and downloaded (or located) associated R code and data from the book's website:

http://www.asdar-book.org/

I also found some nice-looking GIS-related R code here:

https://sites.google.com/site/rodriguezsanchezf/news/usingrasagis

If and when I learn how to accomplish the desired objectives I will post solutions here. Although I may eventually move to ArcGIS if I cannot accomplish the objectives in R.

Answer 1

To plot multiple SpatialPolygons objects on the same device, one approach is to specify the geographic extent you wish to plot first, and then using plot(..., add=TRUE). This will add to the map only those points that are of interest.

Plotting using a projection, (e.g. a polyconic projection) requires first using the spTransform() function in the rgdal package to make sure all the layers are in the same projection.

## Specify a geographic extent for the map
## by defining the top-left and bottom-right geographic coordinates
mapExtent <- rbind(c(-156, 80), c(-68, 19))

## Specify the required projection using a proj4 string
## Use http://www.spatialreference.org/ to find the required string
## Polyconic for North America
newProj <- CRS("+proj=poly +lat_0=0 +lon_0=-100 +x_0=0 
            +y_0=0 +ellps=WGS84 +datum=WGS84 +units=m +no_defs")

## Project the map extent (first need to specify that it is longlat) 
mapExtentPr <- spTransform(SpatialPoints(mapExtent, 
                  proj4string=CRS("+proj=longlat")),
                  newProj)

## Project other layers
can1Pr <- spTransform(can1, newProj)
us1Pr <- spTransform(us1, newProj) 

## Plot each projected layer, beginning with the projected extent
plot(mapExtentPr, pch=NA)
plot(can1Pr, border="white", col="lightgrey", add=TRUE)
plot(us1Pr, border="white", col="lightgrey", add=TRUE)

Adding other features to the map, such as highlighting jurisdictions of interest, can easily be done using the same approach:

## Highlight provinces and states of interest
theseJurisdictions <- c("British Columbia",
                        "Yukon",
                        "Northwest Territories",
                        "Alberta",
                        "Montana",
                        "Alaska")

plot(can1Pr[can1Pr$NAME_1 %in% theseJurisdictions, ], border="white", 
    col="pink", add=TRUE)

plot(us1Pr[us1Pr$NAME_1 %in% theseJurisdictions, ], border="white", 
    col="pink", add=TRUE)

Here is the result:

Add grid-lines when a projection is used is sufficiently complex that it requires another post, I think. Looks as if @Mark Miller as added it below!

Answer 2

Below I have modified PaulG's outstanding answer to display a latitude-longitude grid. The grid is coarser than I would like, but might be adequate. I use the United Kingdom with the code below. I do not know how to include the result in this post.

library(rgdal)
library(raster)

# define extent of map area
mapExtent <- rbind(c(0, 62), c(5, 45))

# BNG is British National Grid    
newProj <- CRS("+proj=tmerc +lat_0=49 +lon_0=-2 +k=0.999601271625 
           +x_0=400000 +y_0=-100000 +ellps=airy +units=m +no_defs") 

mapExtentPr <- spTransform(SpatialPoints(mapExtent, 
                  proj4string=CRS("+proj=longlat")),
                  newProj)

# provide a valid 3 letter ISO country code
# obtain a list with: getData("ISO3")
uk0 <- getData('GADM', country="GBR", level=0) # UK
uk1 <- getData('GADM', country="GBR", level=1) # UK countries
uk2 <- getData('GADM', country="GBR", level=2) # UK counties

# United Kingdom projection
uk1Pr      <- spTransform(uk1, newProj)

# latitude-longitude grid projection
grd.LL     <- gridlines(uk1, ndiscr=100)
lat.longPR <- spTransform(grd.LL, newProj)

# latitude-longitude text projection
grdtxt_LL  <- gridat(uk1)
grdtxtPR   <- spTransform(grdtxt_LL, newProj)

# plot the map, lat-long grid and grid labels
plot(mapExtentPr, pch=NA)
plot(uk1Pr, border="white", col="lightgrey", add=TRUE)
plot(lat.longPR, col="black", add=TRUE)
text(coordinates(grdtxtPR),
   labels=parse(text=as.character(grdtxtPR$labels)))

Result looks like: