calc_QSS: Calc the Quasi Sign Stability measure for antagonistic...

In lsaravia/EcoNetwork: Ecological network analysis includind multiplex networks

Description

The proportion of matrices that are locally stable, these matrices are created by sampling the values of the community matrix (the Jacobian) from a uniform distribution, preserving the sign structure (the links) 1. It also calculates the mean of the real part of the maximum eingenvalue, which is also a measure of stability 2 It uses a uniform distribution between 0 and maximum values given by the parameters negative, positive and selfDamping corresponding to the sign of interactions and self-limitation effect3. If the values of these parameters are 0 then there is no interaction of that kind.

Usage

calc_QSS(ig, nsim = 1000, ncores = 0, negative = -10,
  positive = 0.1, selfDamping = -1)

Arguments

ig	igraph or a list of igraph networks
ncores	number of cores to use in parallel comutation if 0 it uses sequential processing
negative	the maximum magnitude of the negative interaction (the effect of the predator on the prey) must be <= 0
positive	the maximum magnitude of the positive interaction (the effect of the prey on the predator) must be >= 0
selfDamping	the maximum magnitude of the self-limitation (the effect of the species on itself) must be <= 0
sims	number of simulations to calculate QSS

Value

a data.frame with the QSS, and MEing, the mean of the real part of the maximum eingenvalue

References

1. Allesina, S. & Pascual, M. (2008). Network structure, predator - Prey modules, and stability in large food webs. Theor. Ecol., 1, 55–64.

2. Grilli, J., Rogers, T. & Allesina, S. (2016). Modularity and stability in ecological communities. Nat. Commun., 7, 12031

3. Monteiro, A.B. & Del Bianco Faria, L. (2017). Causal relationships between population stability and food-web topology. Functional Ecology, 31, 1294–1300.

Examples

## Not run: 

g <- netData[[2]]

tp <- calc_QSS(g)


## End(Not run)

lsaravia/EcoNetwork documentation built on Oct. 26, 2020, 11:09 p.m.