budyko_errbounds: Generate errorbounds from a Budyko fit

In tylerbhampton/budykoR: Computes Budyko statistics, fits, and makes plots

Generate errorbounds from a Budyko fit

Description

Generate errorbounds from a Budyko fit

Usage

budyko_errbounds(
  fit = NULL,
  res = 0.1,
  hshift = FALSE,
  dif = "nls",
  data = NULL,
  alpha = 0.05,
  method = NULL
)

Arguments

fit	list dataframe object with names() method and entries "param" for parameter p, "err" for vertical fit error, and "hshift" for optional hshift value (default to zero)
res	numeric resolution of fit, defaults to 0.01 units
hshift	A boolean. T/F (default F), whether to test for horizontal shift. hs value must be defined in fit
dif	A string. One of the following: "nls" (default) "rsq" "mae". Determines whether to fit automatically with stats::nls, or manually, with best fit determined by either Pearson's R squared or Minimum Mean Absolute Error
data	list dataframe requiring the following two columns: "PET.P" for potential evapotranspiration divided by precipitation (units normalized, dimensionless), and "AET.P" for actual evapotranspiration divided by precipitation or one minus the runoff coefficient (Q/P)
alpha	numeric If using dif="nls", confidence value for bounds. Default to 0.05 (95% confidence)
method	A string. One of the following: Fu, Turc-Pike, Wang-Tang

Value

Given a "fit" dataframe object from budyko::budyko_fit, generate a dataframe of error bounds that can be fed to plot or ggplot2::ggplot. name (as in names()) of fit must be the method

Examples

library(ggplot2)
fitdata = budyko_errbounds(data=testdata,method="Fu")
blankBC+
  geom_polygon(data=subset(fitdata,key!="fit"),aes(lty=key),
               col=1,fill="transparent")+
  geom_line(data=subset(fitdata,key=="fit"),aes(lty="fit"))+
  scale_linetype_manual(values=c("fit"=1,"confidence"=2,"prediction"=3))+
  geom_point(data=testdata)+
  coord_cartesian(xlim=c(0,5),ylim=c(0,1))

tylerbhampton/budykoR documentation built on April 21, 2024, 4:39 a.m.