coldata.to.imagemat: Map response variables using 'contour()', 'image()', and...
In forestgeo/ctfs: Tools for the Analysis of Forest Dynamics

Description Usage Arguments Details Value Examples

Map response variables using contour(), image(), and filled.contour().

1 2	coldata.to.imagemat(z, x = NULL, y = NULL, gridsize = 20, plotmin = c(0, 0), plotmax = c(1000, 500))

`z`	A numeric vector of responses, which means an attribute which has map coordinates.
`x`	The map coordinates of each column in the matrix, and the values with which the x-axis of the map will be numbered. The number of columns of `z`, `dim(z)\[2]`, must be exactly `length(x)`. If `x` is `NULL`, then it is calculated using `plotmax`.
`y`	The map coordinates of each row in the matrix, and the values with which the y-axis of the map will be numbered; dim(z)[1] must be exactly length(y). If y is NULL, then it is calculated using plotmax.
`gridsize`	Size of each individual map.
`plotmin`	The minimum plot x and y coordinates; in CTFS plots, these are always 0,0. Ignored if x and y are submitted.
`plotmax`	Maximum plot x and y coordinates. The default, 1000,500, is typical CTFS 50-ha plot. Ignored if x and y are submitted.

contour(), image(), and filled.contour() are powerful and do the difficult work, but they are a bit tricky to learn and have many options. If you are expert with those 3 R functions, the functions below will not be especially useful. They simply rearrange various types of data into the correct format.

The first, coldata.to.imagemat(), converts a vector of response variables and converts it into a matrix of the correct form for use by contour(), image(), and filled.contour(). There must be one value of the response z for every position in a grid overlain on the map. In the case of CTFS Plots, this means a value for z at every quadrat, though quadrat can be any size. The function abundanceperquad() produces exactly the correct kind of vector for use by coldata.to.imagemat(). The function gxgy.to.index() produces quadrat numbers that are the correct vector for use by coldata.to.imagemat().

For example, consider a 50 ha plot of 1000 x 500 meters. If gridsize = 20, the standard quadrat, then there are 50 columns x 25 rows in the plot.

In that case, the length of vector z must be exactly 1250, length(x) must be exactly 50, and length(y) exactly 25. Alternatively, x and y can be set

NULL and it will be calculated from plotmax = c(1000, 500) and plotmin = c(0, 0). The simplest application of this function for CTFS plots is to set x and y NULL, and plotmax to the correct plot dimensions. Then make sure z is the right size for the number of quadrats.

The common error with image and contour maps is getting the proper dimensions for z and the proper sizes for plotmax, x, and y. If you get an error data length is not a sub-multiple or multiple of the number of rows'then one of those inputs is wrong.

gridsize must be of square quadrats, and cannot be NULL.

A list designed to match exactly the format of the RGDAL raster object. It has components:

x: same as the argument x; if x is submitted as NULL, this is the vector calculated by using plotmax; * y: likewise for y;
z: exactly as submitted;
columns: number of columns in the grid, same as length(x);
rows: Number of rows in the grid, same as length(y);
mat: The matrix needed for graphing, exactly as needed for instance by imageGraph().

## Not run: 
CTFSplot('bci',6,'full')

BA = abundanceperquad(
  bciex::bci12t6mini,
  gridsize = 20,
  plotdim = c(1000, 500),
  type = 'ba'
  )

totalBAperquad = colSums(BA$ba)
summary(totalBAperquad)
matrixdata <- coldata.to.imagemat(
  z = totalBAperquad,
  x = NULL,
  y = NULL,
  gridsize = 20,
  plotmax = c(1000, 500)
)
length(totalBAperquad)
dim(matrixdata$mat)
length(matrixdata$x)

## End(Not run)