Accumulation rate

Description

Dataframe, output of simulateAccumulationRate.

Usage

data(accumulationRate)

Format

Dataframe with 5000 rows and the following columns: #' @return A dataframe with the following columns.

• time: numeric, time or age of the given case. Important: the time column goes from "left to right", meaning that oldest samples have the lowest values of age/time, and viceversa.

• accumulation.rate: numeric, in years per centimetre, simulated accumulation rate.

• grouping: integer, grouping variable to aggregate together (with aggregateSimulation) samples deposited in the same centimetre according accumulation.rate.

Author(s)

Blas M. Benito <blasbenito@gmail.com>

See Also

simulateAccumulationRate, aggregateSimulation

Computes temporal autocorrelation in a vector, and returns a dataframe for easy plotting.

Description

It reads a vector representing a time series, applies acf for a given number of lags

Usage

acfToDf(
  x = NULL,
  lag.max = 100,
  length.out = 10
  )

Arguments

Details

This function computes temporal autocorrelation of a given vector using acf, and returns a dataframe ready for easy plotting with plotAcf.

Value

A dataframe with the columns: #'

• lag: numeric, lag in the time units of x with a maximum determined by lag.max, and a number of unique values determined by length.out

• acf: Pearson correlation index returned by the acf for a given number of lags for the given lag.

• ci.max: Maximum value of the confidence interval of acf.

• ci.min: Minimum value of the confidence interval of acf.

Author(s)

Blas M. Benito <blasbenito@gmail.com>

See Also

acf, plotAcf

Examples

#getting a driver
data(driverA)

#computing temporal autocorrelations
x.df <- acfToDf(
  x = driverA,
  lag.max = 1000,
  length.out = 100
)
str(x.df)

#plotting output
plotAcf(x.df)

Aggregates the output of simulatePopulation.

Description

Takes the output of simulatePopulation and aggregates it into centimetres by following a sediment accumulation rate produced by simulateAccumulationRate. It further samples it at given depth intervals. It intends to simulate a pseudo-realistic sedimentation of the pollen produced by the simulation, and to apply a pollen-sampling pattern to a virtual pollen core.

Usage

aggregateSimulation(
  simulation.output=NULL,
  accumulation.rate=NULL,
  sampling.intervals=1
  )

Arguments

Details

The function uses the values in the grouping column of the simulateAccumulationRate output to aggregate together (by computing the mean) as many samples as cases in grouping have the same identificator. Output samples are identified by the average age of the samples within the given centimetre.

Value

A list of dataframes with as many rows as virtual taxa were produced by simulatePopulation, and the following columns: column 1 is the original data, column 2 is the original data aggregated by the accumulation rate, columns 3 to n are the different sampling intervals defined by the user.

Author(s)

Blas M. Benito <blasbenito@gmail.com>

See Also

simulateAccumulationRate, simulatePopulation

Examples

#getting example data
data(simulation)
data(accumulationRate)

#aggregating first simulation outcome
sim.output.aggregated <- aggregateSimulation(
 simulation.output = simulation[1],
 accumulation.rate = accumulationRate,
 sampling.intervals = c(2,6))

#comparing simulations
par(mfrow = c(3,1))
#notice the subsetting of the given column of the input list
plot(sim.output.aggregated[[1,1]]$Time,
 sim.output.aggregated[[1,1]]$Pollen,
 type = "l",
 xlim = c(500, 1000),
 main = "Annual"
 )
plot(sim.output.aggregated[[1,2]]$Time,
 sim.output.aggregated[[1,2]]$Pollen,
 type = "l",
 xlim = c(500, 1000),
 main = "2cm"
 )
plot(sim.output.aggregated[[1,3]]$Time,
 sim.output.aggregated[[1,3]]$Pollen,
 type = "l",
 xlim = c(500, 1000),
 main = "6cm"
 )

#check differences in nrow
nrow(sim.output.aggregated[[1,1]]) #original data
nrow(sim.output.aggregated[[1,2]]) #2cm
nrow(sim.output.aggregated[[1,3]]) #6cm intervals

Compares different simulations produced by simulatePopulation.

Description

Plots together the results of different virtual taxa produced by a single run of simulatePopulation.

Usage

compareSimulations(
  simulation.output = NULL,
  species = "all",
  filename = NULL,
  columns = "Pollen",
  time.zoom = NULL,
  width = 12,
  text.size = 20,
  title.size = 25,
  plot.title = ""
  )

Arguments

Details

The user can decide what virtual taxa to plot through the species argument, and what information to show throught the columns argument. Output is plotted on screen by default, and printed to pdf if the filename argument is filled.

Author(s)

Blas M. Benito <blasbenito@gmail.com>

See Also

simulatePopulation, plotSimulation

Examples

#getting example data
data(simulation)

#compare taxa 1, 2, and 3.
compareSimulations(simulation.output = simulation,
 species = c(1, 2, 3),
 columns = c("Pollen", "Suitability"),
 time.zoom = c(1000, 2000)
)

Driver A

Description

A vector of 5000 values (years) between 0 and 100 generated by simulateDriver with a temporal autocorrelation significant for 600 years. It is meant to be used as input for simulatePopulation.

Usage

data(driverA)

Format

Numeric vector of length 5000.

Author(s)

Blas M. Benito <blasbenito@gmail.com>

See Also

simulateDriver

Driver B

Description

A vector of 5000 values (years) between 0 and 100 generated by simulateDriver with a temporal autocorrelation significant for 600 years. It is meant to be used as input for simulatePopulation.

Usage

data(driverB)

Format

Numeric vector of length 5000.

Author(s)

Blas M. Benito <blasbenito@gmail.com>

See Also

simulateDriver

Drivers with different temporal autocorrelation.

Description

A dataframe with 20000 rows and 4 columns (years) containing two drivers (A and B) generated by simulateDriver with different temporal autocorrelations (200 and 600). Each driver represents a period of 5000 years. This dataset is is meant to be used as input for simulatePopulation

Usage

data(drivers)

Format

Dataframe with 4 columns and 20000 rows.

Details

• time integer, represents time from 0 to 10000, where 0 is the oldest sample, and 10000 is the newest one (opposite to the general interpretation of age in palaeoecology!).

• driver character, values are A and B

• autocorrelation.length numeric, values are 200, 600, and 1800.

• value numeric, value of the driver for the given time, driver, and autocorrelation.length.

Author(s)

Blas M. Benito <blasbenito@gmail.com>

See Also

simulateDriver

Fix data types in a parameters dataframe.

Description

It converts all columns (but the label one) of a parameters dataframe created by parametersDataframe and filled by the user into type numeric, and checks the coherence of the parameters for each taxon. It provides feedback on the check results on screen for each taxon.

Usage

fixParametersTypes(x)

Arguments

Value

Same dataframe provided in argument x but with fixed data types.

Author(s)

Blas M. Benito <blasbenito@gmail.com>

See Also

parametersDataframe

Examples

parameters <- parametersDataframe(rows=1)
parameters[1,] <- c("Species 1", 50, 20, 2, 0.2, 0, 100, 1000, 1, 0, 50, 10, 0, 0, 600, 600)
parameters <- fixParametersTypes(x=parameters)

Parameters of 4 virtual taxa.

Description

A dataframe with the parameters of 4 virtual taxa. This dataset was generated by parametersDataframe and fixParametersTypes. It is meant to be used as input for simulatePopulation. Its columns are:

Usage

data(parameters)

Format

Dataframe with 16 columns and 4 rows.

Details

• label: to store names (character string) of the virtual taxa.

• maximum.age: integer, maximum possible age of the individuals in years.

• reproductive.age: integer, age of sexual maturity in years.

• fecundity: integer, number of maximum viable seeds produced by a mature individual under fully suitable conditions.

• growth.rate: numeric, parameter of the logistic growth function.

• pollen.control: numeric in the interval [0, 1]. If 0, pollen productivity depends on environmental suitability only. The larger the number, biomass takes over environmental suitability in determining pollen productivity.

• maximum.biomass: integer, maximum biomass of the individuals.

• carrying.capacity: integer, maximum sum of biomass of the individuals. Very large carrying capacities plus a low maximum.biomass generates too many individuals for the simulation to remain efficient. Try to set carrying.capacity and maximum.biomass to carrying.capacity divided by biomass returns a number lower than 1000 (and even better if it is closer to 100).

• driver.A.weight: numeric in the interval [0, 1], represents the relative influence of the driver on environmental suitability.

• driver.B.weight: numeric in the interval [0, 1], represents the relative influence of the driver on environmental suitability. The sum of weights of drivers A and B should be 1.

• niche.A.mean: numeric, in the same units as driver A. It is the mean of the normal function defining the response of the virtual taxa to driver A.

• niche.A.sd: numeric, in the same units as driver A. It is the standard deviation of the normal function defining the response of the virtual taxa to driver A.

• niche.B.mean: as above, but for driver B.

• niche.B.sd: as above, but for driver B.

• autocorrelation.length.A: numeric, only useful if several drivers generated with different autocorrelation lengths are available (and identified by the column autocorrelation.length) in the drivers argument provided to the simulatePopulation function.

• autocorrelation.length.B: same as above.

Author(s)

Blas M. Benito <blasbenito@gmail.com>

See Also

parametersCheck, parametersDataframe, simulatePopulation

Plots main simulation parameters.

Description

Plots the environmental niche, fecundity, growth curve, and maturity age, of each virtual taxa in a parameters dataframe for simulatePopulation, to help the user in making choices while adjusting them.

Usage

parametersCheck(
  parameters,
  species = "all",
  driver.A = NULL,
  driver.B = NULL,
  drivers = NULL,
  filename = NULL
  )

Arguments

Details

The function prints the plot, can save it to a pdf file if filename is provided, and returns a ggplot2 object. Priority is given to drivers introduced through the drivers argument.

Value

A ggplot2 object.

Author(s)

Blas M. Benito <blasbenito@gmail.com>

See Also

parametersDataframe, fixParametersTypes

Examples

#getting data
data(parameters)
data(drivers)

#plotting parameters
parametersCheck(
 parameters = parameters,
 drivers = drivers
 )

Generates a template dataframe to contain simulation parameters.

Description

Generates the dataframe structure needed to contain the parameters used as input for the simulatePopulation function.

Usage

parametersDataframe(rows=1)

Arguments

Details

The resulting dataframe can either be filled manually through vectors, as shown in the example (but this requires to use the function fixParametersTypes once the dataframe is completed), or can be edited manually in Rstudio by installing the editData package.

Value

A dataframe filled with NA values and the columns:

• label: to store names (character string) of the virtual taxa.

• maximum.age: integer, maximum possible age of the individuals in years.

• reproductive.age: integer, age of sexual maturity in years.

• fecundity: integer, number of maximum viable seeds produced by a mature individual under fully suitable conditions.

• growth.rate: numeric, parameter of the logistic growth function.

• pollen.control: numeric in the interval [0, 1]. If 0, pollen productivity depends on environmental suitability only. The larger the number, biomass takes over environmental suitability in determining pollen productivity.

• maximum.biomass: integer, maximum biomass of the individuals.

• carrying.capacity: integer, maximum sum of biomass of the individuals. Very large carrying capacities plus a low maximum.biomass generates too many individuals for the simulation to remain efficient. Try to set carrying.capacity and maximum.biomass to carrying.capacity divided by biomass returns a number lower than 1000 (and even better if it is closer to 100).

• driver.A.weight: numeric in the interval [0, 1], represents the relative influence of the driver on environmental suitability.

• driver.B.weight: numeric in the interval [0, 1], represents the relative influence of the driver on environmental suitability. The sum of weights of drivers A and B should be 1.

• niche.A.mean: numeric, in the same units as driver A. It is the mean of the normal function defining the response of the virtual taxa to driver A.

• niche.A.sd: numeric, in the same units as driver A. It is the standard deviation of the normal function defining the response of the virtual taxa to driver A.

• niche.B.mean: as above, but for driver B.

• niche.B.sd: as above, but for driver B.

• autocorrelation.length.A: numeric, only useful if several drivers generated with different autocorrelation lengths are available (and identified by the column autocorrelation.length) in the drivers argument provided to the simulatePopulation function.

• autocorrelation.length.B: same as above.

Author(s)

Blas M. Benito <blasbenito@gmail.com>

See Also

simulatePopulation, fixParametersTypes

Examples

#generating the template
parameters <- parametersDataframe(rows=1)

#filling it with a vector
parameters[1,] <- c("Species 1", 50, 20, 2, 0.2, 0, 100, 1000, 1, 0, 50, 10, 0, 0, 600, 600)

Plots results of acfToDf.

Description

Plots a dataframe resulting from acfToDf by using ggplot2.

Usage

plotAcf(
  x = NULL,
  plot.title = ""
  )

Arguments

Value

A ggplot object

Author(s)

Blas M. Benito <blasbenito@gmail.com>

See Also

acfToDf

Examples

#getting a driver
data(driverA)

#computing temporal autocorrelations
x.df <- acfToDf(
  x = driverA,
  lag.max = 1000,
  length.out = 100
)
str(x.df)

#plotting output
plotAcf(x.df)

Plots results of simulatePopulation.

Description

This function takes as input the output of simulatePopulation, and plots the pollen abundance, number of individuals, biomass, driver, and environmnetal suitability of each simulation outcome.

Usage

plotSimulation(
  simulation.output = NULL,
  species = "all",
  burnin = FALSE,
  filename = NULL,
  time.zoom = NULL,
  panels = c("Driver A",
           "Driver B",
           "Suitability",
           "Population",
           "Mortality",
           "Biomass",
           "Pollen"
           ),
 plot.title = NULL,
 width = 12,
 text.size = 20,
 title.size = 25,
 line.size = 1
 )

Arguments

Details

The user can decide what virtual taxa to plot (argument species), and what information to show throught the panels argument. Output is plotted on screen by default, and printed to pdf if the filename argument is filled.

Author(s)

Blas M. Benito <blasbenito@gmail.com>

See Also

simulatePopulation, compareSimulations

Examples

#getting example data
data(simulation)

#plot first simulation
plotSimulation(simulation.output = simulation[[1]])

Rescales a vector within given bounds.

Description

Takes a numeric vector x and rescales it within the values given by new.min and new.max.

Usage

rescaleVector(
  x = rnorm(100),
  new.min = 0,
  new.max = 100,
  integer = FALSE
  )

Arguments

Value

A vector of the same length as x rescaled between output.min and output.max.

Author(s)

Blas M. Benito <blasbenito@gmail.com>

Examples

#generating example data
x = rnorm(100)

#as float
x.float <- rescaleVector(
  x = x,
  new.min = 0,
  new.max = 100,
  integer = FALSE
  )

#as integer
x.integer <- rescaleVector(
  x = x,
  new.min = 0,
  new.max = 100,
  integer = TRUE
  )

Simulates a virtual sediment accumulation rate.

Description

Generates a virtual sediment accumulation rate to be applied to the results of simulatePopulation.

Usage

simulateAccumulationRate(
  seed=50,
  time=1:1000,
  output.min=10,
  output.max=40,
  direction=1,
  plot=TRUE
  )

Arguments

Details

The accumulation rate curve is generated through a random walk smoothed by a GAM model. The value of the seed argument changes the shape of the curve, but the user has no more control than trying different values to achieve a curve closer to the desired one. If plot is set to TRUE, the accumulation rate curve is printed on screen, but not exported to pdf.

Value

A dataframe like accumulationRate, with the following columns:

• time: numeric, time or age of the given case.

• accumulation.rate: numeric, in years per centimetre, simulated accumulation rate.

• grouping: integer, grouping variable to aggregate together (with aggregateSimulation) samples deposited in the same centimetre according accumulation.rate.

Author(s)

Blas M. Benito <blasbenito@gmail.com>

See Also

simulatePopulation, aggregateSimulation

Examples

acc.rate <- simulateAccumulationRate(
 seed = 50,
 time = 1:1000,
 output.min = 10,
 output.max = 40,
 direction = 1,
 plot = TRUE
 )

str(acc.rate)

Generates a random time series with temporal autocorrelation.

Description

Generates a vector of the same legnth as the time argument, with a temporal autocorrelation length close to the defined by autocorrelation.length, and a range within output.min and output.max. The output of this function is intended to be used as an input to the function simulatePopulation. Important: note that the variable time runs from left to right in simulatePopulation, with lower values representing older samples.

Usage

simulateDriver(
  random.seed = 30,
  time = 1:10000,
  autocorrelation.length = 100,
  output.min = 0,
  output.max = 100,
  rescale = TRUE
  )

Arguments

Details

It is recommended to use time vectors with a time step of 1 between consecutive values when the output is to be used as input for simulatePopulation, which considers annual time-steps while simulating virtual pollen curves. The initial random sequence of numers is generated by rnorm. Desired temporal autocorrelation are approximate, but deviation becomes higher if autocorrelation.length is larger than half the length of time. Consequently, the function limits autocorrelation.length to length(time)/2.

Value

A vector of the same length as time. Datasets driverA and driverB are outputs of this function.

Author(s)

Blas M. Benito <blasbenito@gmail.com>

See Also

rescaleVector, driverA, driverB, set.seed

Examples

x <- simulateDriver(
  random.seed = 30,
  time = 1:10000,
  autocorrelation.length = 100,
  output.min = -10,
  output.max = 20,
  rescale = TRUE
  )

#plots output
plot(x, type = "l")

#checks temporal autocorrelation
acf(x, lag.max = 300)

Generates drivers for simulatePopulation.

Description

Wrapper of simulateDriver to generate several drivers with different autocorrelation lengths, and return a long format dataframe to be used as input for simulatePopulation. It also produces a plot of the generated drivers. Important: note that the variable time runs from left to right, with lower values representing older samples.

Usage

simulateDriverS(
 random.seeds=c(60, 120),
 time=1:10000,
 autocorrelation.lengths=c(200, 600),
 output.min=c(0,0),
 output.max=c(100, 100),
 driver.names=c("A", "B"),
 filename=NULL)

Arguments

Details

It is recommended to use time vectors with a time step of 1 between consecutive values when the output is to be used as input for simulatePopulation, which considers annual time-steps while simulating virtual pollen curves. Initial random sequence is generated by rnorm. Desired temporal autocorrelation are approximate, but deviation becomes higher if autocorrelation.length is larger than half the length of time. Consequently, the function limits autocorrelation.length to length(time)/2.

Value

A long format dataframe (see dataset drivers) with the following columns:

• time: integer.

• driver: character, values are A and B

• autocorrelation.length: numeric, default values are 200, 600, and 1800.

• value: numeric, value of the driver for the given time.

Author(s)

Blas M. Benito <blasbenito@gmail.com>

See Also

drivers, simulateDriver, simulatePopulation, drivers

Examples

drivers <- simulateDriverS(
 random.seeds=c(60, 120),
 time=1:10000,
 autocorrelation.lengths=c(200, 600),
 output.min=c(0,0),
 output.max=c(100, 100),
 driver.names=c("A", "B"),
 filename=NULL
)

str(drivers)

Simulates population dynamics for virtual species with different traits.

Description

This function takes as input a dataframe of parameters defining virtual taxa produced by parametersDataframe and fixParametersTypes, a driver or drivers generated with simulateDriver or simulateDriverS, and simulates population dynamics for the given virtual taxa at yearly resolution for the time-length defined by the driver or drivers. Important: note that the variable time runs from left to right, with lower values representing older samples. The model relies on the following set of assumptions:

• The spatial structure of the population is not important to explain its pollen productivity. This is an operative assumption, to speed-up model execution.

• The environmental niche of the species follows a Gaussian distribution, characterized by a mean (niche optimum, also niche position) and a standard deviation (niche breadth or tolerance).

• Different drivers can have a different influence on the species dynamics, and that influence can be defined by the user by tuning the weights of each driver.

• Environmental suitability, expressed in the range [0, 1], is the result of an additive function of the species niches (normal function defined by the species' mean and standard deviation for each driver), the drivers' values, and the relative influence of each driver (driver weights).

• Pollen productivity is a function of the individual's biomass and environmental suitability, so under a hypothetical constant individual's biomass, its pollen production depends linearly on environmental suitability values.

• Effective fecundity is limited by environmental suitability. Low environmental suitability values limit recruitment, acting as an environmental filter. Therefore, even though the fecundity of the individuals is fixed by the fecundity parameter, the overall population fecundity is limited by environmental suitability.

Usage

simulatePopulation(
  parameters=NULL,
  species="all",
  driver.A=NULL,
  driver.B=NULL,
  drivers=NULL,
  burnin=TRUE
  )

Arguments

Details

The model starts with a population of 100 individuals with random ages, in the range [1, maximum age], taken from a uniform distribution (all ages are equiprobable). For each environmental suitability value, including the burn-in period, the model performs the following operations:

• Aging: adds one year to the age of the individuals.

• Mortality due to senescence: individuals reaching the maximum age are removed from the simulation.

• Local extinction and immigration: If the number of individuals drops to zero, the population is replaced by a "seed bank" of #' 100 individuals with age zero, and the simulation jumps to step 7.. This is intended to simulate the arrival of seeds from nearby regions, and will only lead to population growth if environmental suitability is higher than zero.

• Plant growth: Applies a plant growth equation to compute the biomass of every individual.

• Carrying capacity: If maximum population biomass is reached, individuals are iteratively selected for removal according to a mortality risk curve computed by the equation P_{m} = 1 - sqrt(a/A), were Pm is the probability of mortality, a is the age of the given individual, and A is the maximum age reached by the virtual taxa. This curve gives removal preference to younger individuals, matching observed patterns in natural populations.

• Pollen productivity: In each time step the model computes the pollen productivity (in relative values) of the population using the equation P_{t} = \sum x_{it} \times max(S_{t}, B), where t is time (a given simulation time step), P is the pollen productivity of the population at a given time, x_i represents the biomass of every adult individual, S is the environmental suitability at the given time, B is the contribution of biomass to pollen productivity regardless of environmental suitability (pollen.control parameter in the simulation, 0 by default). If B equals 1, P is equal to the total biomass sum of the adult population, regardless of the environmental suitability. If B equals 0, pollen productivity depends entirely on environmental suitability values.

• Reproduction: Generates as many seeds as reproductive individuals are available multiplied by the maximum fecundity and the environmental suitability of the given time.

The model returns a table with climatic suitability, pollen production, and population size (reproductive individuals only) per simulation year. Figure 10 shows the results of the population model when applied to the example virtual species.

Value

A list of dataframes, each one of them with the results of one simulation. The dataset simulation exemplifies the output of this function. Each dataframe in the output list has the columns:

• Time integer, ages in years. Negative ages indicate the burn-in period.

• Pollen numeric, pollen counts

• Population.mature numeric, number of mature individuals.

• Population.immatre numeric, number of immature individuals.

• Population.viable.seeds numeric, number of viable seeds generated each year.

• Suitability numeric, environmental suitability computed from the driver by the normal function/s defining the taxon niche.

• Biomass.total numeric, overall biomass of the population.

• Biomass.mature numeric, sum of biomass of mature individuals.

• Biomass.immature numeric, sum of biomass of immature individuals.

• Mortality.mature numeric, number of mature individuals dead each year.

• Mortality.immature numeric, same as above for immature individuals.

• Driver.A numeric, values of driver A.

• Driver.B numeric, values of driver B, if available, and NA otherwise.

• Period qualitative, with value "Burn-in" for burn-in period, and "Simulation" otherwise.

Author(s)

Blas M. Benito <blasbenito@gmail.com>

See Also

parametersDataframe, fixParametersTypes, plotSimulation

Examples

#getting data
data(parameters)
data(driverA)

#simulating population dynamics
# of first taxon in parameters
# for first 500 values of driverA
sim.output <- simulatePopulation(
 parameters=parameters[1,],
 driver.A=driverA[1:500]
 )

#checking output
str(sim.output)

List with simulation outputs for all virtual taxa in parameters.

Description

A list of dataframes with 4 slots, output of simulatePopulation, taking parameters and drivers as inputs. Each dataframe in the list has the following columns:

Usage

data(simulation)

Format

List with 4 dataframes with outputs of simulatePopulation.

Details

• Time: integer, ages in years. Negative ages indicate the burn-in period.

• Pollen: numeric, pollen counts

• Population.mature: numeric, number of mature individuals.

• Population.immatre: numeric, number of immature individuals.

• Population.viable.seeds: numeric, number of viable seeds generated each year.

• Suitability: numeric, environmental suitability computed from the driver by the normal function/s defining the taxon niche.

• Biomass.total: numeric, overall biomass of the population.

• Biomass.mature: numeric, sum of biomass of mature individuals.

• Biomass.immature: numeric, sum of biomass of immature individuals.

• Mortality.mature: numeric, number of mature individuals dead each year.

• Mortality.immature: numeric, same as above for immature individuals.

• Driver.A: numeric, values of driver A.

• Driver.B: numeric, values of driver B, if available, and NA otherwise.

• Period: qualitative, with value "Burn-in" for burn-in period, and "Simulation" otherwise.

Author(s)

Blas M. Benito <blasbenito@gmail.com>

See Also

simulatePopulation, plotSimulation

Reinterpolates aggregated simulations into regular time.

Description

Takes the output of aggregateSimulation, and interpolates it into a regular time grid.

Usage

toRegularTime(
  x = NULL,
  time.column = "Time",
  interpolation.interval = NULL,
  columns.to.interpolate = c("Suitability",
                             "Driver.A",
                             "Pollen")
  )

Arguments

Details

This function fits a loess model of the form y ~ x, where y is any column given by columns.to.interpolate and x is the column given by the time.column argument. The model is used to interpolate column y on a regular time series of intervals equal to interpolation.interval. If x is a matrix-like list returned by aggregateSimulation (on results of simulateAccumulationRate and simulatePopulation), the first column of the matrix will already have a regular time column, and therefore nothing will be done with this column of the list.

Value

If x is a list of dataframes, the function returns a list with the same structure as the input list. If x is a dataframe, the function returns a dataframe. In any case, output dataframes have the columns "Time" (now regular), and any column listed in columns.to.interpolate. Important: as in the input data, the time column of the output data has lower time for oldest samples and higher time for newest samples.

Author(s)

Blas M. Benito <blasbenito@gmail.com>

See Also

simulateAccumulationRate, aggregateSimulation

Examples

## Not run: 
#getting example data
data(simulation)
data(accumulationRate)

#aggregating first simulation outcome
sim.output.aggregated <- aggregateSimulation(
 simulation.output = simulation[1],
 accumulation.rate = accumulationRate,
 sampling.intervals = c(2,6))

#to regular time
sim.output.aggregated <- toRegularTime(
 x=sim.output.aggregated,
 time.column ="Time",
 interpolation.interval = 10,
 columns.to.interpolate = c("Suitability", "Pollen")
)

#comparing simulations
par(mfrow = c(3,1))
#notice the subsetting of the given column of the input list
plot(sim.output.aggregated[[1,1]]$Time,
    sim.output.aggregated[[1,1]]$Pollen,
    type = "l",
    xlim = c(500, 1000),
    main = "Annual"
)
plot(sim.output.aggregated[[1,2]]$Time,
    sim.output.aggregated[[1,2]]$Pollen,
    type = "l",
    xlim = c(500, 1000),
    main = "2cm"
)
plot(sim.output.aggregated[[1,3]]$Time,
    sim.output.aggregated[[1,3]]$Pollen,
    type = "l",
    xlim = c(500, 1000),
    main = "6cm"
)

#check differences in nrow
nrow(sim.output.aggregated[[1,1]]) #original data
nrow(sim.output.aggregated[[1,2]]) #2cm
nrow(sim.output.aggregated[[1,3]]) #6cm intervals

## End(Not run)