[R-sig-Geo] overlay 1D graph on a map or map image: van Langren data

[hadley wickham]

On Wed, Dec 3, 2008 at 8:33 AM, Michael Friendly <friendly at yorku.ca> wrote:
> Hi
> I'm working with some historical data on determinations of longitude used by
> Michael F van Langren
> in 1644 to draw what is believed to be the first graph of statistical data.
> The values are estimates
> of the distance in longitude between Toledo and Rome.
>
> G. Iansonius      17.736  1501  Jan Jansson              Flanders
>  [1588-1664] sb:~ 1615
> G. Mercator       19.872  1567  Gerardus Mercator        Flanders
> I. Schonerus      20.638  1536  Johann Schöner           Germany
> P. Lantsbergius   21.106  1530  Phillip van Lansberge    Belgium
> T. Brahe          21.447  1578  Tycho Brahe              Denmark
> I. Regiomontanus  25.617  1471  Johann Muller            Germany
>  [1436-1476] sb: 1463
> Orontius          26.000  1542  Oronce Finé              France
> [1494-1555] C. Clavius        26.340  1567  Christoph Clavius        Germany
>  C. Ptolomeus      27.787   150  Claudius Ptolemaeus      Alexandria
> A. Argelius       28.170  1610  Andrea Argoli            Italy        sb:
> 1610
> A. Maginus        29.787  1582  Giovanni Antonio Magini  Italy
> D. Origanus       30.128  1601  David Origanus           Germany
>
> I want to show these overlaid on a map, something like the following:
>
> http://euclid.psych.yorku.ca/SCS/Gallery/images/Private/Langren/langren-google-overlay.jpg
>
> using *either* a jpeg image,
> http://euclid.psych.yorku.ca/SCS/Gallery/images/Private/Langren/google-toledo-rome3.jpg
> or a comparable portion of the world map, something like
>
> # approximate coordinates of the BBox of this map
> bbox <- c( 38.186, -9.184,
>          43.692, 28.674 )
> bbox <- matrix(bbox, 2, 2, byrow=TRUE)
> map("world", xlim=bbox[,2], ylim=bbox[,1], fill=TRUE, col=colors())
> map.axes()
>
> I created the overlay by manually rescaling and calibrating his graph with a
> portion of a Google map.
> But I'd like to be able to do this more exactly in R, possibly using other
> data.
>
> I can read in  the data as follows:
>
> langren <-
> read.csv("http://euclid.psych.yorku.ca/SCS/Gallery/Private/langrens.csv",
> header=TRUE)
>
> # Lat/Long of Toledo & Rome, from Google map
>
> toledo <- c(39.86, -4.03)
> rome   <- c(41.89, 12.5)
>
> # locations of the estimates, converted to lat, long
> locations <- cbind( rep(rome[1], 12), toledo[2] + langren$long)
> rownames(locations) <- langren$name
> locations <- rbind( toledo, locations)
> colnames(locations) <- c("lat", "long")
>
> The above steps give:
>
> locations <-
> structure(c(39.86, 41.89, 41.89, 41.89, 41.89, 41.89, 41.89,
> 41.89, 41.89, 41.89, 41.89, 41.89, 41.89, -4.03, 13.706, 15.842,
> 16.608, 17.076, 17.417, 21.587, 21.97, 22.31, 23.757, 24.14,
> 25.757, 26.098), .Dim = c(13L, 2L), .Dimnames = list(c("toledo",
> "G. Iansonius", "G. Mercator", "I. Schonerus", "P. Lantsbergius",
> "T. Brahe", "I. Regiomontanus", "Orontius", "C. Clavius", "C. Ptolomeus",
> "A. Argelius", "A. Maginus", "D. Origanus"), c("lat", "long")))
>
> I can also read the .jpg Google map image, at least from a local file:
>
> # read the google image
> library(rimage)
> gimage <-
> read.jpeg("C:/Documents/milestone/images/vanLangren/google-toledo-rome3.jpg")
>
> #gimageloc <-
> "http://euclid.psych.yorku.ca/SCS/Gallery/images/Private/Langren/google-toledo-rome3.jpg"
> #dest <- paste(tempfile(),'.jpg', sep='')
> #download.file(gimageloc, dest)
> #  why doesn't this work?
> #gimage <- read.jpeg(dest)
>
> plot(gimage)
> # approx  pixel coordinates of Toledo and Rome in the image, measured from
> the *top left* corner as (0,0)
> toledo.map <- c(130, 119)
> rome.map <-(505, 59)
>
> Here's where I'm stumped.  How can I recreate a version of van Langren's
> graph shown on top of
> either the gimage or on top of the portion of the R world map?

Here's a quick start using ggplot2 - one trick to make the map look
nicer is to use a bigger range when calling map, and then adjust the
limits of the plot to zoom in on the region of interest.

library(maps)
library(ggplot2)

langren <- read.csv("http://euclid.psych.yorku.ca/SCS/Gallery/Private/langrens.csv",
header=TRUE)

bbox <- c( 38.186, -9.184,
         43.692, 28.674 )
bbox <- matrix(bbox, 2, 2, byrow=TRUE)

borders <- as.data.frame(map("world", plot = F,
  xlim = expand_range(bbox[,2], 0.2),
  ylim = expand_range(bbox[,1], 0.2))[c("x", "y")])

data(world.cities)
cities <- subset(world.cities,
  name %in% c("Rome", "Toledo") & country.etc %in% c("Spain", "Italy"))

ggplot(langren, aes(long, lat)) +
  geom_path(aes(x, y), borders, colour = "grey60") +
  geom_point(y = 40) +
  geom_text(aes(label = name), y = 40.1, angle = 90, hjust = 0, size = 3) +
  geom_point(data = cities, colour = "red", size = 2) +
  coord_cartesian(xlim=bbox[,2], ylim=bbox[,1])

Regards,

Hadley

-- 
http://had.co.nz/