R/PlotAOR.r

In LDalby/ralmass: Functions to Prepare and Analyze Data for and from ALMaSS

#' Plot the AOR index
#'
#' Plot the AOR index (Hoye et al. 2012). The index is based on the 
#' abundance-occupancy relationship and facilitates the interpretation of 
#' outputs from individual based models.
#' 
#' @param data data.frame Data.frame with x, y and scenario column.
#' @param x character The name of the column with x coordinates (occupancy).
#' @param y character The name of the column with y coordinates (abundance).
#' @param scenarios character The name of the column indicating which scenario
#' a given set of coordinates belong to.
#' @param title character The title of the plot (optional).
#' @param fixed logical If TRUE (the default) plot has xlim and ylim -1,1.
#' @param legendtitle character Title of the legend.
#' @param triangle logical Should the area of decrease on both axes be highlighted?
#' default is FALSE.
#' @param trigger numeric The trigger value for accepted decrease in the
#' summed-AOR indexs score. Mostly used when assessing impact of pesticides.
#' Any points falling beyond the area highlighted by the shaded triangle are an
#' unacceptable level of impact.
#' @param brewerpal character Name of a RColorBrewer palette. If left blank 
#' ggplot2 default colors are used.
#' @param shape logical Should scenarios be distinguished by shape (instead of 
#' colors)? Default is FALSE.
#' @return An AOR plot
#' @references Hoye, T. T., Skov, F. & Topping, C.J. (2012). Interpreting
#' outputs of agent-based models using abundance-occupancy relationships. 
#' Ecological Indicators 20: 221-227.
#' @export
#' @examples
#' df = data.frame('x' = rnorm(8, 0, .5), 'y' = rnorm(8, 0, .5), 'scenario' = letters[1:8])
#' PlotAOR(df, x = 'x', y = 'y', scenarios = 'scenario')
PlotAOR = function(data, x = NULL, y = NULL, scenarios = NULL, fixed = TRUE,
 title = NULL, legendtitle = NULL, triangle = FALSE, trigger = 0,
 brewerpal = NULL, shape = FALSE)
{
  # Append the origo point to the data:
  if(data.table::is.data.table(data))
  {
    data = data[,c(x, y, scenarios), with = FALSE]
    dataorigo = data.frame(x = rep(0, nrow(data)), y = rep(0, nrow(data)),
     'scenario' = unique(data[,scenarios, with = FALSE]))
    setnames(dataorigo, c(x,y,scenarios))
  }
  if(!data.table::is.data.table(data))
  {
    data = data[,match(c(x, y, scenarios), names(data))]
    dataorigo = data.frame(x = rep(0, nrow(data)), y = rep(0, nrow(data)),
     'scenario' = unique(data[,scenarios]))
    names(dataorigo) = c(x,y,scenarios)
  }
  dataorigo = rbind(data, dataorigo)
  ltitle = ''
  if(!is.null(legendtitle))
  {
    ltitle = legendtitle
  }
# Setup the plot
  p = ggplot2::ggplot(dataorigo, ggplot2::aes_string(x,y))
  if(triangle)
  {
    trigger = abs(trigger)
    triangle = data.frame(x = c(-1,-1,1-trigger), y = c(1-trigger,-1,-1))
    p = p + ggplot2::geom_polygon(data = triangle, ggplot2::aes(x=x, y=y), alpha = 0.10)
  }
  p = p + ggplot2::geom_vline(xintercept = 0, size = 0.4) +
  ggplot2::geom_hline(yintercept = 0, size = 0.4)
  if(shape)
  {
    p = p + ggplot2::geom_line(ggplot2::aes_string(group = scenarios), size = 0.3) + 
    ggplot2::geom_point(data = data, ggplot2::aes_string(x, y,
     shape = scenarios), size = 2) + scale_shape(name = ltitle)    
  }
  if(!shape)
  {
    p = p + ggplot2::geom_line(ggplot2::aes_string(color = scenarios)) + 
    ggplot2::geom_point(data = data, ggplot2::aes_string(x,y,
     color = scenarios), size = 2)
  }
  if(!is.null(brewerpal))
  {
    p = p + ggplot2::scale_colour_brewer(palette=brewerpal, name = ltitle)
  }
  p = p + ggplot2::labs(y = expression(paste(Delta, ' Abundance')), 
    x = expression(paste(Delta, ' Occupancy'))) + ggplot2::theme_bw(base_size = 10)
  if(!is.null(title))
  {
    p = p + ggplot2::ggtitle(title)
  }
  if(fixed)
  {
   p = p + ggplot2::coord_cartesian(ylim = c(-1,1), xlim = c(-1,1))
   return(p)
 }
 if(!fixed)
 {
  return(p)
}
}