R/NeighborDensityFun.R
In forestgeo/ctfs: Tools for the Analysis of Forest Dynamics
# Roxygen documentation generated programatically -------------------
#' Total number or basal area of conspecific and heterospecific neighbors.
#'
#' @description Calculates the total number or basal area of conspecific and
#' heterospecific neighbors at a radius r with edge effect correction. Most
#' arguments aim to describe the type of neighbors to count.
#' [NeighborDensities()] counts neighbors for each of the individuals in
#' censdata.
#'
#'
#' @return
#' A data.frame with two columns listing the density of consp and hetsp, and
#' rows equal to the size of censdata if `censdata2` is absent, or `censdata2`
#' if it is present.
#'
#' @details
#' * Arguments `mindbh` and `r` refer to neighbours
#' * `censdata` should have the coordinates gx, gy, species id `sp`, `status`,
#' `tag` and `dbh`.
#'
#' Dependencies: CalcRingArea and circlearea from the CTFS package, and splancs
#' package.
#'
#' @author Tania Brenes
#'
#' @template plotdim
#' @template mindbh
#' @template censdata
#' @param censdata2 (NULL) an optional dataset for focal individuals, for
#' example a subset of only one species or a list of seedling coordinates that
#' are not part of census data. It must contain coordinates (gx,gy), species
#' id (sp) and tag. It will count the neighbors in censdata but only for these
#' focal coordinates.
#' @template r
#' @param type (count) calculates count of stems ('count') or sum of basal areas
#' ('basal'). Note that for the sum of basal areas is better to use the plot
#' stem table, otherwise only one stem per tree will be included .
#' @param include (c("A")) a vector of status to include in the neighbors, by
#' default it counts only alive individuals, for all status use
#' c("A","P","D","M").
#'
#' @examples
#' \dontrun{
#' # sum consp and hetsp neighbors for all stems in the plot
#' neighbor.counts <- NeighborDensities(bciex::bci12t7mini, r=20, type='count')
#'
#' # sum consp and hetsp neighbors for only one species:
#' one.sp = subset(bciex::bci12t7mini, sp=="quaras")
#' neighbor.counts <- NeighborDensities(bciex::bci12t7mini, one.sp, type='count')
#' }
#'
'NeighborDensities'
#' Count of neighbors within radius r with edge effect correction.
#'
#' @description
#' Calculates the count of neighbors within radius `r` with edge effect
#' correction. Is a quick version of [NeighborDensities()] with limited
#' capabilities.
#'
#' @details
#' `censdata` should have the coordinates `gx`, `gy`.
#'
#' @return A single vector with neighbor counts within radius r for each
#' individual in censdata.
#'
#' @author Tania Brenes
#'
#' @template plotdim
#' @template censdata
#' @template r
#'
#' @examples
#' \dontrun{
#' # Count all the consp neighbors of one species
#' one.sp = subset(bciex::bci12t7mini, sp == "ingasa" & status == "A")
#' neighbor.counts <- NDcount(one.sp)
#' }
#'
'NDcount'
# Source code and original documentation ----------------------------
# <function>
# <name>
# NeighborDensities
# </name>
# <description>
# Calculates the total number or basal area of conspecific and heterospecific neighbors at a radius r
# with edge effect correction. Most arguments aim to describe the type of neighbors to count <br>
# Dependencies: CalcRingArea and circlearea from the CTFS package, and splancs package.<br>
# Output: a data.frame with two columns listing the density of consp and hetsp, and rows equal to the size of censdata if censdata2 is absent, or censdata2 if it is present.
# Author: Tania Brenes<br>
# </description>
# <arguments>
# <ul>
# <li> censdata: should have the coordinates (gx, gy), species id (sp), status, tag and dbh. It counts neighbors for each of the individuals in censdata.
# <li> censdata2 (NULL): an optional dataset for focal individuals, for example a subset of only one species or a list of seedling coordinates that are not part of census data. It must contain coordinates (gx,gy), species id (sp) and tag. It will count the neighbors in censdata but only for these focal coordinates.
# <li> r (20): radius to count neighbors
# <li> plotdim (c(1000,500)): vector with two numbers indicating the xy-plot size;
# <li> mindbh (10): minimum size of neighbors to be counted
# <li> type (count): calculates count of stems ('count') or sum of basal areas ('basal'). Note that for the sum of basal areas is better to use the plot stem table, otherwise only one stem per tree will be included .
# <li> include (c("A")): a vector of status to include in the neighbors, by default it counts only alive individuals, for all status use c("A","P","D","M").
# </ul>
# </arguments>
# <sample>
# > load("bci.full7.rdata")<br>
# > attach("CTFSRPackage.rdata")<br>
# ## sum consp and hetsp neighbors for all stems in the plot <br>
# > neighbor.counts <- NeighborDensities(bci.full7, r=20, type='count')<br>
# ## sum consp and hetsp neighbors for only one species:<br>
# > one.sp = subset(bci.full7, sp=="quaras")<br>
# > neighbor.counts <- NeighborDensities(bci.full7, one.sp, type='count')<br>
# </sample>
# <source>
#' @export
NeighborDensities <- function(censdata,
censdata2 = NULL,
r = 20,
plotdim = c(1000, 500),
mindbh = 10,
type = 'count',
include = c("A")) {
ptm <- proc.time()
if (is.null(censdata2)) censdata2 <- censdata
n = nrow(censdata2)
output = matrix(NA, ncol=2, nrow=n)
dimnames(output)[[2]] <- c("con.sp","het.sp")
if (type =='count') {
# spd = subset( censdata, !is.na(gx) & !is.na(gy) & !duplicated(tag) & dbh>=mindbh & status %in% include)
cond_1 <- !is.na(censdata$gx) &
!is.na(censdata$gy) &
!duplicated(censdata$tag) &
censdata$dbh >= mindbh &
censdata$status %in% include
spd <- censdata[cond_1, , drop = FALSE]
spd <- rm_na_row(spd)
for (i in 1:n) {
focal = censdata2[i,]
if (is.na(focal$gx) | is.na(focal$gy) | duplicated(focal$tag)) output[i,1:2] = NA else {
poly= with(focal, splancs::spoints(c(gx-r, gy-r, gx-r, gy+r, gx+r, gy+r, gx+r, gy-r)))
use = splancs::inpip(splancs::spoints(rbind(spd$gx,spd$gy)), poly)
if (length(use)==0) output[i,1:2]=0 else {
incircle = sqrt(splancs::dsquare(splancs::spoints(rbind(spd$gx[use],spd$gy[use])), splancs::spoints(rbind(focal$gx,focal$gy)) )) <= r
nn = use[incircle] # & spd$tag[use] != focal$tag[i]
consp <- spd$sp[nn]==focal$sp
area = CalcRingArea(data.frame(gx=focal$gx,gy=focal$gy), r, plotdim)$each
output[i,1] = length(nn[consp]) * (pi*r^2 /area)
output[i,2] = length(nn[!consp]) * (pi*r^2 /area)
} }
if (i %in% seq(5000,n+5000,5000)) cat( i, "of", n, " elapsed time = ", (proc.time()-ptm)[3], "seconds" , "\n" )
}
} # end of count subrutine
if (type =='basal') {
# spd = subset( censdata, !is.na(gx) & !is.na(gy) & !is.na(dbh) & dbh>=mindbh & status %in% include)
cond_2 <- !is.na(censdata$gx) &
!is.na(censdata$gy) &
!is.na(censdata$dbh) &
censdata$dbh >= mindbh &
censdata$status %in% include
spd <- censdata[cond_2, , drop = FALSE]
spd <- rm_na_row(spd)
for (i in 1:n)
{
focal = censdata2[i,]
if(is.na(focal$gx) | is.na(focal$gy) ) output[i,1:2] = NA else
{
poly= with(focal, splancs::spoints(c(gx-r, gy-r, gx-r, gy+r, gx+r, gy+r, gx+r, gy-r)))
use = splancs::inpip(splancs::spoints(rbind(spd$gx,spd$gy)), poly)
if (length(use)==0) output[i,1:2]=0 else
{
incircle = sqrt(splancs::dsquare(splancs::spoints(rbind(spd$gx[use],spd$gy[use])), splancs::spoints(rbind(focal$gx,focal$gy)) )) <= r
nn = use[incircle] # & spd$tag[use] != focal$tag[i]
area = CalcRingArea(data.frame(gx=focal$gx,gy=focal$gy), r, plotdim)$each
BA = circlearea(spd$dbh[nn]/20) # div by 2 for radius * 10 to convert mm to cm
consp <- spd$sp[nn]==focal$sp
output[i,1] = sum(BA[consp], na.rm=TRUE) * (pi*r^2 /area)
output[i,2] = sum(BA[!consp], na.rm=TRUE) * (pi*r^2 /area)
}
}
if (i %in% seq(5000,n+5000,5000)) cat( i, "of", n, " elapsed time = ", (proc.time()-ptm)[3], "seconds" , "\n")
}
} # end of basal subrutine
cat( "Total elapsed time = ", (proc.time()-ptm)[3], "seconds" , "\n")
return(output)
}
# </source>
# </function>
# <function>
# <name>
# NDcount
# </name>
# <description>
# Calculates the count of neighbors within radius r with edge effect correction. Is a quick version of NeighborDensities with limited capabilities. <br>
# Dependencies: CalcRingArea from the CTFS package and splancs. <br>
# Output: a single vector with neighbor counts within radius r for each individual in censdata. <br>
# Author: Tania Brenes
# </description>
# <arguments>
# <ul>
# <li> censdata: Is the coordinates dataset, should have the coordinates (gx, gy)
# <li> r (20): radius
# <li> plotdim (c(1000,500)): vector with two numbers indicating the xy-plot size;
# </ul>
# </arguments>
# <sample>
# ## count all the consp neighbors of one species <br>
# > one.sp = subset(bci.full7, sp=="ingasa" & status=="A")<br>
# > neighbor.counts <- NDcount(one.sp)<br>
# </sample>
# <source>
#' @export
NDcount=function(censdata, r=20, plotdim=c(1000,500))
{
sppi.pts <- with(censdata, data.frame(gx,gy))
areas <- CalcRingArea(sppi.pts, r, plotdim)$each
poly=splancs::spoints(c(0,0,plotdim[1],0,plotdim[1],plotdim[2],0,plotdim[2]))
counts <- khat(sppi.pts, poly=poly,r, newstyle=TRUE)$counts
output = counts * pi * r^2 /areas
return(as.vector(output))
}
# </source>
# </function>
forestgeo/ctfs documentation built on May 3, 2019, 6:44 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
# Roxygen documentation generated programatically -------------------
#' Total number or basal area of conspecific and heterospecific neighbors.
#'
#' @description Calculates the total number or basal area of conspecific and
#' heterospecific neighbors at a radius r with edge effect correction. Most
#' arguments aim to describe the type of neighbors to count.
#' [NeighborDensities()] counts neighbors for each of the individuals in
#' censdata.
#'
#'
#' @return
#' A data.frame with two columns listing the density of consp and hetsp, and
#' rows equal to the size of censdata if `censdata2` is absent, or `censdata2`
#' if it is present.
#'
#' @details
#' * Arguments `mindbh` and `r` refer to neighbours
#' * `censdata` should have the coordinates gx, gy, species id `sp`, `status`,
#' `tag` and `dbh`.
#'
#' Dependencies: CalcRingArea and circlearea from the CTFS package, and splancs
#' package.
#'
#' @author Tania Brenes
#'
#' @template plotdim
#' @template mindbh
#' @template censdata
#' @param censdata2 (NULL) an optional dataset for focal individuals, for
#' example a subset of only one species or a list of seedling coordinates that
#' are not part of census data. It must contain coordinates (gx,gy), species
#' id (sp) and tag. It will count the neighbors in censdata but only for these
#' focal coordinates.
#' @template r
#' @param type (count) calculates count of stems ('count') or sum of basal areas
#' ('basal'). Note that for the sum of basal areas is better to use the plot
#' stem table, otherwise only one stem per tree will be included .
#' @param include (c("A")) a vector of status to include in the neighbors, by
#' default it counts only alive individuals, for all status use
#' c("A","P","D","M").
#'
#' @examples
#' \dontrun{
#' # sum consp and hetsp neighbors for all stems in the plot
#' neighbor.counts <- NeighborDensities(bciex::bci12t7mini, r=20, type='count')
#'
#' # sum consp and hetsp neighbors for only one species:
#' one.sp = subset(bciex::bci12t7mini, sp=="quaras")
#' neighbor.counts <- NeighborDensities(bciex::bci12t7mini, one.sp, type='count')
#' }
#'
'NeighborDensities'
#' Count of neighbors within radius r with edge effect correction.
#'
#' @description
#' Calculates the count of neighbors within radius `r` with edge effect
#' correction. Is a quick version of [NeighborDensities()] with limited
#' capabilities.
#'
#' @details
#' `censdata` should have the coordinates `gx`, `gy`.
#'
#' @return A single vector with neighbor counts within radius r for each
#' individual in censdata.
#'
#' @author Tania Brenes
#'
#' @template plotdim
#' @template censdata
#' @template r
#'
#' @examples
#' \dontrun{
#' # Count all the consp neighbors of one species
#' one.sp = subset(bciex::bci12t7mini, sp == "ingasa" & status == "A")
#' neighbor.counts <- NDcount(one.sp)
#' }
#'
'NDcount'
# Source code and original documentation ----------------------------
# <function>
# <name>
# NeighborDensities
# </name>
# <description>
# Calculates the total number or basal area of conspecific and heterospecific neighbors at a radius r
# with edge effect correction. Most arguments aim to describe the type of neighbors to count <br>
# Dependencies: CalcRingArea and circlearea from the CTFS package, and splancs package.<br>
# Output: a data.frame with two columns listing the density of consp and hetsp, and rows equal to the size of censdata if censdata2 is absent, or censdata2 if it is present.
# Author: Tania Brenes<br>
# </description>
# <arguments>
# <ul>
# <li> censdata: should have the coordinates (gx, gy), species id (sp), status, tag and dbh. It counts neighbors for each of the individuals in censdata.
# <li> censdata2 (NULL): an optional dataset for focal individuals, for example a subset of only one species or a list of seedling coordinates that are not part of census data. It must contain coordinates (gx,gy), species id (sp) and tag. It will count the neighbors in censdata but only for these focal coordinates.
# <li> r (20): radius to count neighbors
# <li> plotdim (c(1000,500)): vector with two numbers indicating the xy-plot size;
# <li> mindbh (10): minimum size of neighbors to be counted
# <li> type (count): calculates count of stems ('count') or sum of basal areas ('basal'). Note that for the sum of basal areas is better to use the plot stem table, otherwise only one stem per tree will be included .
# <li> include (c("A")): a vector of status to include in the neighbors, by default it counts only alive individuals, for all status use c("A","P","D","M").
# </ul>
# </arguments>
# <sample>
# > load("bci.full7.rdata")<br>
# > attach("CTFSRPackage.rdata")<br>
# ## sum consp and hetsp neighbors for all stems in the plot <br>
# > neighbor.counts <- NeighborDensities(bci.full7, r=20, type='count')<br>
# ## sum consp and hetsp neighbors for only one species:<br>
# > one.sp = subset(bci.full7, sp=="quaras")<br>
# > neighbor.counts <- NeighborDensities(bci.full7, one.sp, type='count')<br>
# </sample>
# <source>
#' @export
NeighborDensities <- function(censdata,
censdata2 = NULL,
r = 20,
plotdim = c(1000, 500),
mindbh = 10,
type = 'count',
include = c("A")) {
ptm <- proc.time()
if (is.null(censdata2)) censdata2 <- censdata
n = nrow(censdata2)
output = matrix(NA, ncol=2, nrow=n)
dimnames(output)[[2]] <- c("con.sp","het.sp")
if (type =='count') {
# spd = subset( censdata, !is.na(gx) & !is.na(gy) & !duplicated(tag) & dbh>=mindbh & status %in% include)
cond_1 <- !is.na(censdata$gx) &
!is.na(censdata$gy) &
!duplicated(censdata$tag) &
censdata$dbh >= mindbh &
censdata$status %in% include
spd <- censdata[cond_1, , drop = FALSE]
spd <- rm_na_row(spd)
for (i in 1:n) {
focal = censdata2[i,]
if (is.na(focal$gx) | is.na(focal$gy) | duplicated(focal$tag)) output[i,1:2] = NA else {
poly= with(focal, splancs::spoints(c(gx-r, gy-r, gx-r, gy+r, gx+r, gy+r, gx+r, gy-r)))
use = splancs::inpip(splancs::spoints(rbind(spd$gx,spd$gy)), poly)
if (length(use)==0) output[i,1:2]=0 else {
incircle = sqrt(splancs::dsquare(splancs::spoints(rbind(spd$gx[use],spd$gy[use])), splancs::spoints(rbind(focal$gx,focal$gy)) )) <= r
nn = use[incircle] # & spd$tag[use] != focal$tag[i]
consp <- spd$sp[nn]==focal$sp
area = CalcRingArea(data.frame(gx=focal$gx,gy=focal$gy), r, plotdim)$each
output[i,1] = length(nn[consp]) * (pi*r^2 /area)
output[i,2] = length(nn[!consp]) * (pi*r^2 /area)
} }
if (i %in% seq(5000,n+5000,5000)) cat( i, "of", n, " elapsed time = ", (proc.time()-ptm)[3], "seconds" , "\n" )
}
} # end of count subrutine
if (type =='basal') {
# spd = subset( censdata, !is.na(gx) & !is.na(gy) & !is.na(dbh) & dbh>=mindbh & status %in% include)
cond_2 <- !is.na(censdata$gx) &
!is.na(censdata$gy) &
!is.na(censdata$dbh) &
censdata$dbh >= mindbh &
censdata$status %in% include
spd <- censdata[cond_2, , drop = FALSE]
spd <- rm_na_row(spd)
for (i in 1:n)
{
focal = censdata2[i,]
if(is.na(focal$gx) | is.na(focal$gy) ) output[i,1:2] = NA else
{
poly= with(focal, splancs::spoints(c(gx-r, gy-r, gx-r, gy+r, gx+r, gy+r, gx+r, gy-r)))
use = splancs::inpip(splancs::spoints(rbind(spd$gx,spd$gy)), poly)
if (length(use)==0) output[i,1:2]=0 else
{
incircle = sqrt(splancs::dsquare(splancs::spoints(rbind(spd$gx[use],spd$gy[use])), splancs::spoints(rbind(focal$gx,focal$gy)) )) <= r
nn = use[incircle] # & spd$tag[use] != focal$tag[i]
area = CalcRingArea(data.frame(gx=focal$gx,gy=focal$gy), r, plotdim)$each
BA = circlearea(spd$dbh[nn]/20) # div by 2 for radius * 10 to convert mm to cm
consp <- spd$sp[nn]==focal$sp
output[i,1] = sum(BA[consp], na.rm=TRUE) * (pi*r^2 /area)
output[i,2] = sum(BA[!consp], na.rm=TRUE) * (pi*r^2 /area)
}
}
if (i %in% seq(5000,n+5000,5000)) cat( i, "of", n, " elapsed time = ", (proc.time()-ptm)[3], "seconds" , "\n")
}
} # end of basal subrutine
cat( "Total elapsed time = ", (proc.time()-ptm)[3], "seconds" , "\n")
return(output)
}
# </source>
# </function>
# <function>
# <name>
# NDcount
# </name>
# <description>
# Calculates the count of neighbors within radius r with edge effect correction. Is a quick version of NeighborDensities with limited capabilities. <br>
# Dependencies: CalcRingArea from the CTFS package and splancs. <br>
# Output: a single vector with neighbor counts within radius r for each individual in censdata. <br>
# Author: Tania Brenes
# </description>
# <arguments>
# <ul>
# <li> censdata: Is the coordinates dataset, should have the coordinates (gx, gy)
# <li> r (20): radius
# <li> plotdim (c(1000,500)): vector with two numbers indicating the xy-plot size;
# </ul>
# </arguments>
# <sample>
# ## count all the consp neighbors of one species <br>
# > one.sp = subset(bci.full7, sp=="ingasa" & status=="A")<br>
# > neighbor.counts <- NDcount(one.sp)<br>
# </sample>
# <source>
#' @export
NDcount=function(censdata, r=20, plotdim=c(1000,500))
{
sppi.pts <- with(censdata, data.frame(gx,gy))
areas <- CalcRingArea(sppi.pts, r, plotdim)$each
poly=splancs::spoints(c(0,0,plotdim[1],0,plotdim[1],plotdim[2],0,plotdim[2]))
counts <- khat(sppi.pts, poly=poly,r, newstyle=TRUE)$counts
output = counts * pi * r^2 /areas
return(as.vector(output))
}
# </source>
# </function>
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.