Learn R Programming

terra (version 1.7-83)

roll: Rolling (moving) functions

Description

Compute "rolling" or "moving" values, such as the "rolling average" for each cell in a SpatRaster.

See focal for spatially moving averages and similar computations. And see cumsum and other cum* functions to compute cumulate values.

Usage

# S4 method for SpatRaster
roll(x, n, fun=mean, type="around", circular=FALSE,
			na.rm=FALSE, filename="", ..., wopt=list()) 

# S4 method for numeric roll(x, n, fun=mean, type="around", circular=FALSE, na.rm=FALSE, ...)

Value

Same as x

Arguments

x

SpatRaster or numeric

n

integer > 1. The size of the "window", that is, the number of sequential cells to use in fun

fun

a function like mean, min, max, sum

type

character. One of "around", "to", or "from". The choice indicates which values should be used in the computation. The focal cell is always used. If type is "around", (n-1)/2 before and after the focal cell are also included. If type = "from", n-1 cells are after the focal cell are included. If type = "to", n-1 cells before the focal cell are included. For example, when using n=3 for element 5 of a vector; "around" used elements 4,5,6; "to" used elements 3,4,5, and "from" uses elements 5,6,7

circular

logical. If TRUE, the data are considered to have a circular nature (e.g. days or months of the year), such that there are no missing values before first or after the last value.

na.rm

logical. If TRUE, NA values should be ignored (by fun)

filename

character. Output filename

...

additional arguments for fun

wopt

list with named options for writing files as in writeRaster

See Also

cumsum, focal

Examples

Run this code
## numeric
roll(1:12, 3, mean)
roll(1:12, 3, mean, "to")
roll(1:12, 3, mean, circular=TRUE)

## SpatRaster
r <- rast(ncol=2, nrow=2, nlyr=10, vals=1)
r[1,2] = 2
r[2,2] = 4

roll(r, n=3, "sum", "from", na.rm=FALSE) |> values()
roll(r, n=3, "sum", "from", na.rm=TRUE) |> values()
roll(r, n=3, "sum", "from", circular=TRUE) |> values()

roll(r, n=3, "sum", "to", na.rm=TRUE) |> values()

roll(r, n=3, "sum", "around", circular=TRUE) |> values()

Run the code above in your browser using DataLab