Mario J. Marques-Azevedo
nicheApport package simulate and fit the stochastic niche-oriented
species abundance distribution models proposed by Tokeshi (1990, 1996).
This package simulate species rank abundance distributions (RADs) and
use Monte Carlo approach, proposed by Bersier & Sugihara (1997) and
improved by Cassey & King (2001) and Mouillot et al. (2003), to fit
models to replicated data.
The nicheApport package provides:
-
Simulate species abundance models: Dominance Decay, Dominance Preemption, MacArthur Fraction, Random Fraction, Random Assortment and Power Fraction;
-
Fit simulated rank abundance distribution to observed rank abundance data.
Bersier, L.-F. & Sugihara, G. 1997. Species abundance patterns: the problem of testing stochastic models. J. Anim. Ecol. 66: 769-774.
Cassey, P. & King, R. A. R. 2001. The problem of testing the goodness-of-fit of stochastic resource apportionment models. Environmetrics 12: 691-698.
Mouillot, D. et al. 2003. How parasites divide resources: a test of the niche apportionment hypothesis. J. Anim. Ecol. 72: 757-764.
Tokeshi, M. 1990. Niche apportionment or random assortment: species abundance patterns revisited. J. Anim. Ecol. 59: 1129-1146.
Tokeshi, M. 1996. Power fraction: a new explanation of relative abundance patterns in species-rich assemblages. Oikos 75: 543-550.
library(devtools)
install_github(repo = "mariojose/nicheApport", ref = "master", build_vignettes = TRUE)
This version have the last updates waiting to release.
library(devtools)
install_github(repo = "mariojose/nicheApport", ref = "dev", build_vignettes = TRUE)
Note that if you are running Microsoft Windows, you need to install the
Rtools
package to build vignettes locally. If you do not want build vignettes,
remove 'build_vignettes = TRUE' argument.
We can simulate one of the six models informing number of species, total abundance and type of data:
# Load library
library(nicheApport)
# Dominance Decay model with 10 species and 100 individuals
# ('count = TRUE' define that data is discrete)
dominanceDecay(N = 100, S = 10, count = TRUE)
> [1] 17 17 12 12 11 9 7 7 6 2
# Dominance Preemption model with 10 species and 100 kg,
# for instance, as total abundance.
# ('count = FALSE' define that data is continuous - default value)
dominancePreemp(N = 100, S = 10, count = FALSE)
> [1] 7.288709e+01 2.330507e+01 3.683463e+00 7.807442e-02 3.385535e-02
> [6] 1.207574e-02 3.693183e-04 2.271262e-06 1.322838e-06 4.708442e-07
# Power Fraction model with 10 species, 100 individuals and
# 'w' parameter iqual 0.2
powerFraction(N = 100, S = 10, w = 0.2, TRUE)
> [1] 42 22 16 9 5 4 1 1 0 0
For fitting model to data we can use fitmodel function or fitmodelCl function to run simulation in cluster mode.
# Creating an abstract data with 10 replicates and 40 ranks (species)
dt <- matrix(nrow = 10, ncol = 40)
for(i in 1:length(dt[ ,1])){
dt[i, ] <- randFraction(N = 100, S = 40, count = FALSE)
}
# Fitting model to data
m_rf<- fitmodel(x = dt, model = "randFraction", count = FALSE, nRand = 99)
# Printing statistics for fitting
m_rf
> Niche Apportionment fitting
> Data type: continuous
> Model: randFraction
> Number of randomization: 99
> Number of ranks: 40
> Number of replicates: 10
>
> Fitting for mean
> Min. 1st Qu. Median Mean 3rd Qu. Max.
> 18.87 44.58 59.83 75.17 96.92 249.29
> Observed T: 98.1
> P-value for Tobs greater than simulated: 0.24
>
> Fitting for variance
> Min. 1st Qu. Median Mean 3rd Qu. Max.
> 27.82 53.16 68.61 74.21 87.54 182.54
> Observed T: 192.93
> P-value for Tobs greater than simulated: 0
# Get simulated data statistics
m_rf@sim.stats
> rank1 rank2 rank3 rank4 rank5
> mean 0.35346394 0.172064019 0.107631889 0.0746136716 0.0554965529
> variance 0.02729775 0.004558392 0.001721746 0.0008838744 0.0005735733
> rank6 rank7 rank8 rank9 rank10
> mean 0.0424598395 0.033592337 0.0269939667 0.0218551241 0.0178165851
> variance 0.0003837577 0.000271175 0.0002051491 0.0001543256 0.0001167225
> rank11 rank12 rank13 rank14 rank15
> mean 1.488136e-02 1.232765e-02 1.038316e-02 8.910712e-03 7.531478e-03
> variance 8.745819e-05 6.451295e-05 5.007444e-05 4.061548e-05 3.232488e-05
> rank16 rank17 rank18 rank19 rank20
> mean 6.344476e-03 5.425695e-03 4.641984e-03 3.946289e-03 3.339890e-03
> variance 2.450332e-05 1.950282e-05 1.577709e-05 1.242384e-05 9.697496e-06
> rank21 rank22 rank23 rank24 rank25
> mean 2.825188e-03 2.376024e-03 1.984426e-03 1.681094e-03 1.409751e-03
> variance 7.420778e-06 5.485287e-06 4.229102e-06 3.302857e-06 2.467756e-06
> rank26 rank27 rank28 rank29 rank30
> mean 1.172788e-03 9.793651e-04 8.238802e-04 6.752213e-04 5.600907e-04
> variance 1.752866e-06 1.333184e-06 1.037953e-06 7.532167e-07 5.469703e-07
> rank31 rank32 rank33 rank34 rank35
> mean 4.544333e-04 3.665674e-04 2.881572e-04 2.181292e-04 1.641016e-04
> variance 3.725015e-07 2.569343e-07 1.701734e-07 1.001785e-07 6.321146e-08
> rank36 rank37 rank38 rank39 rank40
> mean 1.190960e-04 8.155040e-05 5.499667e-05 3.070667e-05 1.380774e-05
> variance 3.720048e-08 2.038689e-08 1.202986e-08 4.048009e-09 1.383442e-09
The nicheApport project are hosted in GitHub and issues or bugs report are welcome.
For citing nicheApport use citation(package = 'nicheApport') function in R.