ch_get_peaks {CSHShydRology}R Documentation

Extracts peak flows over a threshold

Description

This function is development code being shared as is. It is expected that the user will be interested in the data frame returned for POT analysis and for plotting (i.e. ch_booth_plot).

This function retrieves peaks greater than or equal to the prescribed threshold. It returns a data frame of peak characteristics suitable for subsequent analysis.

The portion under development is returns a list of the flows during an event with the values of the four preceding days and three subsequent days. If the peak is a single point the fragment is nine points long; if the events is longer the fragment contains all days above the threshold and eight additional days.

Usage

ch_get_peaks(dataframe, threshold)

Arguments

dataframe

a data frame of streamflow data containing columns named Date and Flow

threshold

a value for the threshold. Values above the threshold are tested for peaks.

Value

Returns a list containing:

POTevents

a dataframe contining details of the events

events

a vector with the value 0 when the flow is below the threshold and 1 when above.

event_num

a vector with the value 0 when the flow is below a threshold or the index of the events when the threshold was exceeded. i.e. 1,2,3, etc

st_date

start date of events

case

a list of the daily flows in each individual event (see details for more information)

The POTevents data frame contains five columns:

st_date

starting date of event

max_date

date of maximum in the event

max

maximum discharge during event

volume

flow volume during the event

duration

length of the event in days

The case list contains the flows during an event and also for four preceding and subsequent days. Each event will have a length between nine to n days in length. Note: in rare cases where the event is in progress when data becomes available the event might be shorter than nine days long.

Author(s)

Paul Whitfield

References

Burn, D.H., Whitfield, P.H., Sharif, M., 2016. Identification of changes in floods and flood regimes in Canada using a peaks over threshold approach. Hydrological Processes, 39: 3303-3314. DOI:10.1002/hyp.10861

Whitfield, P.H., and J.W. Pomeroy. 2016. Changes to flood peaks of a mountain river: implications for analysis of the 2013 flood in the Upper Bow River, Canada. Hydrological Processes 30:4657-73. doi: 10.1002/hyp.10957.

See Also

ch_booth_plot

Examples

CAN05AA008 <- CAN05AA008
threshold <- 0.5*max(CAN05AA008$Flow)  # arbitrary threshold
my_peaks <- ch_get_peaks(CAN05AA008, threshold)
str(my_peaks)

[Package CSHShydRology version 1.3.0 Index]