Set resource dynamics — setResource • mizer

Sets the intrinsic resource growth rate and the intrinsic resource carrying capacity as well as the name of the function used to simulate the resource dynamics. By default this function changes both the rate and the capacity together in such a way that the resource replenishes at the same rate at which it is consumed.

Usage

setResource(
  params,
  resource_rate = NULL,
  resource_capacity = NULL,
  resource_level = NULL,
  resource_dynamics = NULL,
  balance = NULL,
  lambda = resource_params(params)[["lambda"]],
  n = resource_params(params)[["n"]],
  w_pp_cutoff = resource_params(params)[["w_pp_cutoff"]],
  r_pp = deprecated(),
  kappa = deprecated(),
  ...
)

resource_rate(params)

resource_rate(params) <- value

resource_capacity(params)

resource_capacity(params) <- value

resource_level(params)

resource_level(params) <- value

resource_dynamics(params)

resource_dynamics(params) <- value

Arguments

params: A MizerParams object
resource_rate: Optional. Vector of resource intrinsic birth rates or coefficient in the power-law for the birth rate, see Details. Must be strictly positive.
resource_capacity: Optional. Vector of resource intrinsic carrying capacities or coefficient in the power-law for the capacity, see Details. The resource capacity must be larger than the resource abundance.
resource_level: Optional. The ratio between the current resource number density and the resource capacity. Either a number used at all sizes or a vector specifying a value for each size. Must be strictly between 0 and 1, except at sizes where the resource is zero, where it can be NaN. This determines the resource capacity, so do not specify both this and resource_capacity.
resource_dynamics: Optional. Name of the function that determines the resource dynamics by calculating the resource spectrum at the next time step from the current state.
balance: By default, if possible, the resource parameters are set so that the resource replenishes at the same rate at which it is consumed. In this case you should only specify either the resource rate or the resource capacity (or resource level) because the other is then determined automatically. Set to FALSE if you do not want the balancing.
lambda: Used to set power-law exponent for resource capacity if the resource_capacity argument is given as a single number.
n: Used to set power-law exponent for resource rate if the resource_rate argument is given as a single number.
w_pp_cutoff: The upper cut off size of the resource spectrum power law used only if resource_capacity is given as a single number.
r_pp: . Use resource_rate argument instead.
kappa: . Use resource_capacity argument instead.
...: Unused
value: The desired new value for the respective parameter.

Value

setResource: A MizerParams object with updated resource parameters

Setting resource dynamics

You would usually set the resource dynamics only after having finished the calibration of the steady state. Then setting the resource dynamics with this function will preserve that steady state, unless you explicitly choose to set balance = FALSE. Your choice of the resource dynamics only affects the dynamics around the steady state. The higher the resource rate or the lower the resource capacity the less sensitive the model will be to changes in the competition for resource.

The resource_dynamics argument allows you to choose the resource dynamics function. By default, mizer uses a semichemostat model to describe the resource dynamics in each size class independently. This semichemostat dynamics is implemented by the function resource_semichemostat(). You can change that to use a logistic model implemented by resource_logistic() or you can use resource_constant() which keeps the resource constant or you can write your own function.

Both the resource_semichemostat() and the resource_logistic() dynamics are parametrised in terms of a size-dependent rate $r_R(w)$ and a size-dependent capacity $c_R$. The help pages of these functions give the details.

The resource_rate argument can be a vector (with the same length as w_full(params)) specifying the intrinsic resource growth rate for each size class. Alternatively it can be a single number, which is then used as the coefficient in a power law: then the intrinsic growth rate $r_R(w)$ at size $w$ is set to $$r_R(w) = r_R w^{n-1}.$$ The power-law exponent $n$ is taken from the n argument.

The resource_capacity argument can be a vector specifying the intrinsic resource carrying capacity for each size class. Alternatively it can be a single number, which is then used as the coefficient in a truncated power law: then the intrinsic growth rate $c_R(w)$ at size $w$ is set to $$c(w) = \kappa\, w^{-\lambda}$$ for all $w$ less than w_pp_cutoff and zero for larger sizes. The power-law exponent $\lambda$ is taken from the lambda argument.

The values for lambda, n and w_pp_cutoff are stored in a list in the resource_params slot of the MizerParams object so that they can be re-used automatically in the future. That list can be accessed with resource_params(). It also holds the coefficient kappa that describes the steady-state resource abundance.

Examples

params <- NS_params
resource_dynamics(params)
#> [1] "resource_semichemostat"
resource_dynamics(params) <- "resource_constant"