siplab-package: Spatial Individual-Plant Simulation

siplab-packageR Documentation

Spatial Individual-Plant Simulation

Description

A platform for experimenting with spatially explicit individual-based plant modelling

Details

Package: siplab
Type: Package
Version: 1.6
License: GPL

The main top level functions are pairwise(), and assimilation().

pairwise() computes the competition indices most commonly used in individual-tree distance-dependent (or spatially explicit) forest growth models. These indices are based on a sum of functions of size and distance between the subject plant and each of its competitors. They represent an aggregate of pairwise interactions, the angular configuration of competitors and any higher-order interactions are ignored. Each index is characterized by a specific interaction function, here called a kernel, and by a definition of competitors.

assimilation() deals with “fully spatial” models, computing “assimilation indices” that aim at a mechanistic approximation of effective resource capture. One starts with a spatial resource distribution that is typically assumed to be uniform, Plants exert competitive pressure depending on size and distance, described by influence functions. The resource available at each point is allocated to plants according to their local influence and to a partition rule. Finally, the resource uptake may be weighted by an efficiency function that depends on size and distance, and is spatially integrated to obtain the plant's assimilation index. Several examples of influence and efficiency functions are pre-programmed, and others can be easily produced.

The edges() function is useful for handling edge effects.

Some sample data sets are included, see links below.

The package is built on top of the spatstat library (http://spatstat.org/), which needs to be installed first.

Author(s)

Oscar García

Maintainer: O. Garcia <garcia@dasometrics.net>

References

García, O. “Siplab, a spatial individual-based plant modelling system”. Computational Ecology and Software 4(4), 215-222. 2014. (http://www.iaees.org/publications/journals/ces/articles/2014-4(4)/2014-4(4).asp).

García, O. “A generic approach to spatial individual-based modelling and simulation of plant communities”. Mathematical and Computational Forestry and Nat.-Res. Sci. (MCFNS) 6(1), 36-47. 2014. (http://mcfns.net/index.php/Journal/article/view/6_36).

https://github.com/ogarciav/siplab.

http://forestgrowth.unbc.ca/siplab/ (no longer maintained).

See Also

Example siplab data sets: boreasNP, boreasNS, boreasSA, boreasSP.

Some spatstat standard data sets may also be of interest: finpines, longleaf, spruces, waka.

For tutorials try the vignettes. E. g., in R type help.start() to open the help browser, and navigate to Packages > siplab > Vignettes.

Examples


# Pretend that the data is given as a simple data frame
data <- as.data.frame(spruces) #  from a spatstat data set
head(data) # x-y coordinates in a 56x38 m plot, marks are dbh in meters
data$marks = data$marks * 100  # dbh in cm
# Convert to a point pattern object
datap <- as.ppp(data, c(0, 56, 0, 38))  # plot limits (minx, maxx, miny, maxy)
# or datap <- ppp(data$x, data$y, c(0, 56), c(0, 38), marks = data$marks)
# Hegyi (1974) index (as usual without his original 1-foot distance offset)
hegyi <- pairwise(datap, maxR = 6, kernel = powers_ker, kerpar = list(pi=1,
    pj=1, pr=1, smark=1))
head(marks(hegyi))
# ZOI model
zoi <- assimilation(datap, influence=zoi_inf, infpar=c(k=0.2, smark=1),
    asym=1)

siplab documentation built on March 18, 2022, 6:53 p.m.