which.floodPeaks: Extracting flood peaks
In martindurocher/floodRFA: A R-package for regional frequency of floods

Description Usage Arguments Details References Examples

Returns the indices of the peaks above a threshold according to the declustering method put in place by the Water Resources Council. See Lang et al. (1999) for more details.

which.floodPeaks(x, ...)

## S3 method for class 'matrix'
which.floodPeaks(x, ...)

## S3 method for class 'data.frame'
which.floodPeaks(x, ...)

## S3 method for class 'formula'
which.floodPeaks(form, x, ...)

## S3 method for class 'numeric'
which.floodPeaks(x, dt = NULL, u, r = 1, rlow = 0.75,
  ini = "lmax2")

which.clusters(x, u, ...)

`x`	Sample. Can be a vector, matrix or data.frame. If `form` is not specified the first two rows must be respectively the observations and the time of observation.
`form`	Formula that describes the variables: `obs ~ date`.
`dt`	Date or time of observations.
`u`	Threshold.
`r, rlow, ini`	Declustering parameters. See details.

The declustering method performs two steps, First an initial set of peaks are obtained. By default, if ini = 'lmax' the initial peaks are local maximums. Alternatively, if ini = 'clust' a simple run declustering is realized to identify the maximums of continuous clusters above the threshold u. Afterward two additional conditions are required for peaks to not be rejected. First, two peaks Q1 and Q2 must be separated by a period of at least r days. One recommendation is

5 days + log(A)

where A is the drainage area in miles. The second conditions is

Xmin > rlow * min(Q1,Q2).

By defautlt, rlow = 0.75. When one of the two conditions is not statisfied the lowest of the two peaks is discarded. Modified version of the previous algorithm is also implemented. Using ini = 'lmax2'. Instead of verify jointly both condition a first series of peaks is extrated using only the second condition and next the first condition is verify in the newly extracted peaks. The two version are very similar and differ only on few cases where the modified version is more conservative and reject peaks that are kept in the initial version.

The function which.clusters is returning the indices of the peaks identified by the run declustering method. See clusters.

Lang M, Ouarda TBMJ, Bobée B. (1999) Towards operational guidelines for over-threshold modeling. Journal of Hydrology. Dec 6;225(3):103–17.

area <- 14700
b <- ceiling(5 + log(area/(1.609^2))) ## b = 14
xd <- canadaFlood$daily

# Declustering using recommendation.
cid <- which.floodPeaks(flow~date, xd,
                        u = 1000, r = b, rlow = .75, ini = 'lmax')

plot(xd, type = 'l')
points(canadaFlood$daily[cid,], col = 'red', pch = 16)
abline(h = 1000, col = 3, lwd = 2)

## Using a nonstationary threshold.
fit <- lm(log(flow)~date, canadaFlood$daily)
ut <- exp(fitted(fit))+900
lines(xd$date, ut, col = 4, lwd = 2)

cid <- which.floodPeaks(flow~date, xd,
                        u = ut, r = b, rlow = .75, ini = 'lmax')
points(canadaFlood$daily[cid,], col = 'yellow', pch = 16)