Clarification on raster::adjacent() output values:

Question

I am trying to create a rooks case adjacency matrix using raster::adjacent(), but I am having trouble understanding the output.

Reproducible Example:

library(raster)

r <- raster(nrow = 3, ncol = 3)

rAdjacent <- raster::adjacent(r, cells = 1:ncell(r), pairs = TRUE, directions = 4) 

rAdjacent
      from to
 [1,]    1  3
 [2,]    2  1
 [3,]    3  2
 [4,]    4  6
...

From my interpretation of the output, the output says that 1 and 3 have a rooks case relationship (and if I understand this type of relationship correctly, they do not).

Question 1. Is this correct? Am I interpreting this output correctly?

Question 2. How do I create an output using adjacent() or something else that gives me pairs of adjacent cell numbers?

Thanks :)

Answer 1

You need to project your raster.

The function connects the outer meridians if the raster is not projeced (in a geographic (lat/lon) "projection") and there is data at longitudes -180 and 180 degrees. (from here).

r <- raster(nrow = 3, ncol = 3)
crs(r) <- CRS("+proj=robin +datum=WGS84")

rAdjacent <- raster::adjacent(r, cells = 1:ncell(r), 
                              pairs = TRUE, directions = 4) 
rAdjacent[order(rAdjacent[,1]),]

      from to
 [1,]    1  2
 [2,]    1  4
 [3,]    2  1
 [4,]    2  3
 [5,]    2  5
 [6,]    3  2
 [7,]    3  6
 [8,]    4  5
 [9,]    4  1
[10,]    4  7
[11,]    5  4
[12,]    5  6
[13,]    5  2
[14,]    5  8
[15,]    6  5
[16,]    6  3
[17,]    6  9
[18,]    7  8
[19,]    7  4
[20,]    8  7
[21,]    8  9
[22,]    8  5
[23,]    9  8
[24,]    9  6