Advanced Analysis Panel (rank/occupancy/SAR/decay/rarefaction + optional PPC)
Source:R/panel.R
generate_advanced_panel.RdBuild a multi‑panel summary of key ecological diagnostics from a simulation run: rank–abundance (SAD), occupancy–abundance, species–area (accumulation), distance–decay, and per‑quadrat rarefaction curves. If the point‑process packages spatstat.geom and spatstat.explore are available, an additional row of point–process diagnostics is appended, showing Ripley’s K (border correction), \(L(r)-r\), and the pair‑correlation function \(g(r)\).
Arguments
- res
A list produced by the main simulator containing at least:
PParameter list (used for labels/themes if needed).
species_distAn
sfPOINT layer of individuals with aspeciescolumn.abund_matrixSite \(\times\) species abundance data frame (first column
site).site_coordsData frame with
site, x, yfor quadrat centroids.domainAn
sfpolygon/multipolygon of the study area (used for PPC window).
Value
A patchwork object (a ggplot layout). You can print it
or save it with ggplot2::ggsave().
Details
The optional point–process diagnostics are computed only when both
spatstat.geom (for spatial windows and point patterns) and
spatstat.explore (for summary functions such as Kest and
pcf) are installed. The domain is converted to a window (owin)
using spatstat.geom::as.owin() on the sf polygon; the
individuals are converted to a ppp with that window. We then compute:
Kest(ppp, correction = "border")and plot the border‑corrected K.\(L(r) = \sqrt{K(r)/\pi}\) and \(L(r)-r\) (\(>0\) suggests clustering).
pcf(ppp)as an estimate of the pair‑correlation \(g(r)\) (\(>1\) suggests clustering; \(<1\) inhibition).
If conversion to owin/ppp fails, the PPC row is omitted gracefully.
Examples
if (FALSE) { # \dontrun{
panel <- generate_advanced_panel(results_list)
ggplot2::ggsave("advanced_panel.png", panel, width = 12, height = 14, dpi = 300)
} # }