correlation_criteria: Baseflow correlation criterion computation

Description Usage Arguments Details Value Author(s) References See Also Examples

Description

Computation function of correlation criterion used by package baseflow to calibrate parameters of the baseflow separation algorithm. A vectorized version of this function correlation_criteria_vectorized also exists.

Usage

1
corr_crit(BasinData, alpha, tau, updateFunction = "quadr")

Arguments

BasinData

A BasinData object containing hydroclimatic data for the catchment under study.

alpha

Numeric parameter of the filtering reservoir. Its dimension depends on the used update function. Must be positive.

tau

Number of days used to compute cumulative effective rainfall. Must be a positive integer or any object that can be coerced to a positive integer.

updateFunction

Character string giving reservoir function : must be one of "quadr", "lin" or "exp". Default is "quadr".

Details

NA values are not permitted for parameters alpha and tau. If a vector is supplied, the first element is used.

Usage of gradient-based optimization algorithms is not recommended to find the maximum value of this criterion. It is a better option to use vectorized function correlation_criteria_vectorized to perform a grid-search; a non-gradient-based optimization algorithm, like differential evolution, may be used to refine the optimal point afterwards.

Value

A numeric value of the computed criterion, between -1 and 1.

Author(s)

Pelletier and Andreassian (antoine.pelletier@irstea.fr)

References

Pelletier, A. and Andréassian, V.: Hydrograph separation: an impartial parametrisation for an imperfect method, Hydrol. Earth Syst. Sci., 24, 1171–1187, doi: 10.5194/hess-24-1171-2020, 2020

See Also

perform_filtering

BaseflowFilter

corr_crit_vect

bfi

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
library(baseflow)

# Loading example data from airGR package
data(L0123001, package = 'airGR')

# Defining BasinData object

Name <- BasinInfo$BasinName
startDate <- BasinObs$DatesR[1]
endDate <- BasinObs$DatesR[length(BasinObs$DatesR)]
P <- BasinObs$P
PET <- BasinObs$E
Qobs <- BasinObs$Qmm

BasinData_Example <- BasinData(Name, startDate, endDate, P, PET, Qobs, fill = "GR4J")

# Computing correlation criterion
corr_crit(BasinData_Example, 1500, 110, updateFunction = "quadr")

baseflow documentation built on March 19, 2021, 1:06 a.m.