R help - Dec 2010 - Predator Prey Models

Dear R-users,

   I am currently modifying a previously developed predator prey model and
was curious if there was a way to add in a disturbance to the model (let's
say at time t=100).  The disturbance can be the introduction of 40 prey
(N=40) and 10 predators (Pred = 10).  I would like to see my model go from a
state of equilibrium (up to t = 99), show this disturbance (at t = 100) and
then slowly work its way back to equilibrium.  Does anybody know if this
could be done?


LVmod0D <- function(Time, State, Pars) {

with(as.list(c(State, Pars)), {

IngestPred <- rI * N * Pred

GrowthN <- rG * N * (1 - N/K)

MortPred <- rM * Pred



dN <- GrowthN - IngestPred

dPred <- IngestPred * AE - MortPred



return(list(c(dN, dPred)))

})

}



pars <- c(rI = 0.1, rG = 0.9, rM = 0.8, AE = 0.9, K = 20)

yini <- c(N = 20, Pred = 20)

times <- seq(0, 200, by = 1)

y(time=100)<-c(N=10, Pred=5)

print(system.time)

zout <- as.data.frame(lsoda(yini, y(time=100), times, LVmod0D, pars))

plot(zout[,1],zout[,2], type="l",col="black",
ylim=range(0,25),
xlab="time",ylab="population")

points(zout[,1],zout[,3],type="l",col="red")

title(main="prey (black) predator (red)")



Thanks much for your help!

	[[alternative HTML version deleted]]

Craig O'Connell <craigpoconnell <at> gmail.com> writes:
>    I am currently modifying a previously developed predator prey model and
> was curious if there was a way to add in a disturbance to the model
(let's
> say at time t=100).  The disturbance can be the introduction of 40 prey
> (N=40) and 10 predators (Pred = 10).  I would like to see my model go from
a
> state of equilibrium (up to t = 99), show this disturbance (at t = 100) and
> then slowly work its way back to equilibrium.  Does anybody know if this
> could be done?
library(deSolve)
?events

lsoda(...,
 
events=list(data=data.frame(var=c("N","Pred"),time=rep(100,2),
                  value=c(40,10),method=rep("add",2))))

or something like that ...

Craig O'Connell-2 wrote:
> 
> Dear R-users,
> 
>    I am currently modifying a previously developed predator prey model and
> was curious if there was a way to add in a disturbance to the model
(let's
> say at time t=100).  The disturbance can be the introduction of 40 prey
> (N=40) and 10 predators (Pred = 10).  I would like to see my model go from
> a
> state of equilibrium (up to t = 99), show this disturbance (at t = 100)
> and
> then slowly work its way back to equilibrium.  Does anybody know if this
> could be done?
> 
You can also use package simecol, (imports deSolve), as follows:

library(simecol)

LVmod0D <- new("odeModel",
    main = function(time, state, pars) { 
        with(as.list(c(state,pars)), {
            IngestPred <- rI * N * Pred
            GrowthN    <- rG * N * (1 - N/K)
            MortPred   <- rM * Pred
            
            dN    <- GrowthN - IngestPred
            dPred <- IngestPred * AE - MortPred
            
            return(list(c(dN, dPred)))
        })
    },
    parms = c(rI = 0.1, rG = 0.9, rM = 0.8, AE = 0.9, K = 20),
    times = seq(0, 200, by = 1),
    init = c(N = 20, Pred = 20) ,
    solver="lsoda"
) 
 
# from reply of previous poster
Levents <-
list(data=data.frame(var=c("N","Pred"),time=rep(100,2),
                  value=c(40,10),method=rep("add",2))) 
Levents

lv.sim <- sim(LVmod0D,events=Levents)
lv.out <- out(lv.sim)
lv.out

plot(lv.sim)

I'm not quite sure if the specification of the events is what you need. If I
understood correctly this would be more what you want:

Levents <-
list(data=data.frame(var=c("N","Pred"),time=rep(100,2),
                  value=c(40,10),method=rep("rep",2))) 


best

Berend
-- 
View this message in context:
http://r.789695.n4.nabble.com/Predator-Prey-Models-tp3083759p3084165.html
Sent from the R help mailing list archive at Nabble.com.