Build ggplot Graphics for a potData Object — autoplot.potData • potomax

Build ggplot graphics for a potData object.

Usage

# S3 method for potData
autoplot(object,
         type = c("RLplot", "timeplot"),
         aes = FALSE,
         xVar = c("T", "p"),
         group,
         points = c("p", "H"), a = 0.5,
         blockDuration, 
         ...)

Arguments

object: A potData object.
type: Type of plot wanted, either a return-level plot or a time plot with the events shown as vertical segments. Not implemented yet.
aes: Logical. If TRUE the colour, fill and shape of the points are used within the aesthetic function aes wo they are registred to appear in legends.
xVar: Used when type is "RLplot". The variable to map the absissa \(x\) in the return level plot. When xVar is "T" the plotting positions are computed as explained in Renext Computing Details. For xVar = "p" a further step is required to probability of exceedance \(p\) in reference with a given block duration.
group: Character with value in "block" or "source". In the first case the color and the shape of the points depend on the block. In the second case the color and the shape depend on the type or source of the block: "OT", "MAX" and "OTS". By default the value of group is chosen to be "block" when the number of block is small enough and "source" otherwise.
points: Type of plotting positions to be used for a RL plot. See RP.potData.
a: Parameter to used in ppoints when points is "p". See RP.potData.
blockDuration: A block duration used to find the plotting positions when xVar is "p".
...: Other arguments passed to geom_point.

Value

A ggplot layer.

Note

The user might have change the colour and the shape of the points by using scale_colour_manual and scale_shape_manual.

References

Yves Deville (2020). Renext Computing Details. Technical Report.

Examples

pdat <- as.potData(Garonne)
autoplot(pdat) + ggtitle("Garonne potData")

g <- autoplot(pdat, aes = TRUE) + ggtitle("Garonne potData")
g

g <- g +  scale_colour_manual(values = c("firebrick", "SpringGreen")) +
       scale_shape_manual(values = c(21, 22))

## use autolayer
ggplot() + autolayer(pdat) + scale_x_log10() + theme_gray()


## Change the label of the historical block
pdat <- potData(data = Garonne$OTdata$Flow, effDuration = 65,
                MAX.data = list("hist" = Garonne$MAXdata$Flow),
                MAX.effDuration = 143)
autoplot(pdat, aes = TRUE)