pooling {lfstat}

Pooling Procedures of Low Flow Events

Description

Several pooling procedures can be applied to reduce the number of dependent droughts.

Usage

pool_ic(x, tmin = 5, ratio = 0.1)
pool_it(x, tmin = 5)
pool_ma(x, n = 10)
pool_sp(x)
  

Arguments

x

an object of class deficit, e.g. as produced by find_droughts.

tmin

numeric vector of length one interpreted as the number of days between two droughts to be considered independent events. Two droughts are pooled if their inter-event time is less than tmin.

ratio

numeric vector of length. Specifies the minimum ratio of inter-event volume and precedent drought volume. Two droughts are pooled if the critical ratio is deceeded.

n

numeric vector of length one determining the width of the smoothing window

Details

The inter-event criterion (pool_ic) pools subsequent drought events if the inter-event time is less than tmin and the ratio of the drought volume and the inter-event volume is less than a given ratio. The function pool_it is simply a wrapper around pool_ic(..., ratio = Inf).

Pooling by a moving average (pool_ma) simply smooths the time series before finding drought events.

Using the Sequent Peak algorithm (pool_sp), a drought lasts until its cumulative deficit volume is zero again.

Value

an object of class deficit (inherited from xts), with an additional column event.orig.

See Also

find_droughts, summary.deficit

Examples

data(ngaruroro)
ng <- as.xts(ngaruroro)
ng <- ng["1986::1990", ]

drought <- find_droughts(ng)


ic <- pool_ic(drought)
summary(ic)

ma <- pool_ma(drought)
summary(ma)

sp <- pool_sp(drought)
summary(sp)
plot(sp)