sars-package: sars: Fit and compare species-area relationship models using...

In txm676/mmSAR2: Fit and Compare Species-Area Relationship Models Using Multimodel Inference

sars: Fit and compare species-area relationship models using multimodel inference

Description

This package provides functions to fit twenty models to species-area relationship (SAR) data (see Triantis et al. 2012), plot the model fits, and to construct a multimodel SAR curve using information criterion weights. A number of additional SAR functions are provided, e.g. to fit the log-log power model, the general dynamic model of island biogeography (GDM), Coleman's Random Placement model, and piecewise ISAR models (i.e. models with thresholds in the ISAR).

Details

Functions are provided to fit 20 individual SAR models. Nineteen are fitted using non-linear regression, whilst a single model (the linear model) is fitted using linear regression. Each model has its own function (e.g. sar_power). A set of multiple model fits can be combined into a fit collection (sar_multi). Plotting functions (plot.sars) are provided that enable individual model fits to be plotted on their own, or the fits of multiple models to be overlayed on the same plot. Model fits can be validated using a number of checks, e.g. the normality and homogeneity of the model residuals can be assessed.

A multimodel SAR curve can be constructed using the sar_average function. This fits up to twenty SAR models and constructs the multimodel curve (with confidence intervals) using information criterion weights (see summary.sars to calculate a table of models ranked by information criterion weight). The plot.multi functions enables the multimodel SAR curve to be plotted with or without the fits of the individual models.

Other SAR related functions include: (i) lin_pow, which fits the log-log power model and enables comparison of the model parameters with those calculated using the non-linear power model, (ii) gdm, which fits the general dynamic model of island biogeography (Whittaker et al. 2008) using several different functions, and (iii) coleman, which fits Coleman's (1981) random placement model to a species-site abundance matrix. Version 1.3.0 has added functions for fitting, evaluating and plotting a range of commonly used piecewise SAR models (sar_threshold).

Author(s)

Thomas J. Matthews and Francois Guilhaumon

References

Coleman, B. D. (1981). On random placement and species-area relations. Mathematical Biosciences, 54, 191-215.

Guilhaumon, F., Mouillot, D., & Gimenez, O. (2010). mmSAR: an R-package for multimodel species–area relationship inference. Ecography, 33, 420-424.

Matthews, T.J., Guilhaumon, F., Triantis, K.A, Borregaard, M.K., & Whittaker, R.J. (2015b) On the form of species–area relationships in habitat islands and true islands. Global Ecology & Biogeography. DOI: 10.1111/geb.12269.

Triantis, K.A., Guilhaumon, F. & Whittaker, R.J. (2012) The island species–area relationship: biology and statistics. Journal of Biogeography, 39, 215-231.

Whittaker, R.J., Triantis, K.A. & Ladle, R.J. (2008) A general dynamic theory of oceanic island biogeography. Journal of Biogeography, 35, 977-994.

See Also

https://github.com/txm676/sars

Examples

data(galap, package = "sars")
#fit the power model
fit <- sar_power(galap)
summary(fit)
plot(fit)

#Construct a multimodel averaged SAR curve, using no grid_start simply
#for speed (not recommended - see documentation for sar_average())
fit_multi <- sar_average(data = galap, grid_start = "none")
summary(fit_multi)
plot(fit_multi)

txm676/mmSAR2 documentation built on Nov. 21, 2024, 5:03 a.m.