climwin (version 0.1.2)

plotall: Visualise climate window data

Description

Creates a panel of plots to help visualise climate window data.

Usage

plotall(dataset, datasetrand = NULL, bestmodel = NULL,
  bestmodeldata = NULL, cw1 = 0.95, cw2 = 0.5, cw3 = 0.25,
  histq = 0.99, title = NULL)

Arguments

dataset

A dataframe containing information on all fitted climate windows. Output from climatewin.

datasetrand

A dataframe containing information on all fitted climate windows using randomised data. Output from randwin.

bestmodel

A model object. The strongest climate window model. Returned from singlewin or climatewin.

bestmodeldata

A dataframe containing the biological and climate data used to fit the strongest climate window model. Output from singlewin or climatewin.

cw1,cw2,cw3

Cumulative weight levels used to visualise model weight distribution. See plotweights for more detail.

histq

If datasetrand is provided. The quantile of the randomised data to be compared with non-randomised data. Can be used to determine the likelihood of finding a climate window model of a given deltaAICc value by chance.

title

Title of the plot panel.

Value

• Will return a panel of 6-8 plots:
  • DeltaAICc: A colour plot of model deltaAICc values (larger negative values indicate stronger models). DeltaAICc is the difference between AICc of each climate window model and the baseline model containing no climate data.
  • Model weight: A plot showing the distribution of cumulative model weights. Gradient levels determined by parameters cw1, cw2 and cw3. Darker areas have a higher chance of containing the best climate window.
  • Model betas: A colour plot of model beta estimates. Where applicable, 2nd order coefficients (quadratic) and 3rd order coefficients (cubic) will be plotted separately.
  • Histogram(s): If datasetrand is provided, plotall will create two stacked histograms to compare the deltaAICc of non-randomised and randomised data. This can help determine the likelihood of obtaining a deltaAICc value for a fitted climate window model at random. Without datasetrand, plotall will create a single histogram of deltaAICc values for all fitted climate windows.
  • Boxplots: Two boxplots showing the opening and closing day for a subset of best climate windows. Best climate windows make up the cumulative model weight equivalent to the largest value of cw1, cw2 and cw3. Values above boxplots represent the median values.
  • Best Model: If bestmodel and bestmodeldata are provided, plotall will create a scatterplot to show the fit of the best model through the data.

Examples

# Visualise a fixed climate window generated for dataframes Mass and MassClimate

data(MassOutput)
data(Mass)
data(MassClimate)

single <- singlewin(xvar = list(Temp = MassClimate$Temp),
                   cdate = MassClimate$Date, bdate = Mass$Date,
                   baseline = lm(Mass ~ 1, data = Mass),
                   furthest = 72, closest = 15,
                   stat = "mean", func = "lin",
                   type = "fixed", cutoff.day = 20, cutoff.month = 5,
                   cmissing = FALSE, cinterval = "day")

plotall(dataset = MassOutput, bestmodel = single$bestmodel,
       bestmodeldata = single$bestmodeldata,
       cw1 = 0.95, cw2 = 0.5, cw3 = 0.25, histq = 0.99, title = "Mass")