| gr4j {hydromad} | R Documentation |
GR4J model (mode'le du Ge´nie Rural a' 4 parame'tres Journalier).
gr4j.sim(DATA, x1, etmult = 1, S_0 = 0.5,
return_state = FALSE,transformed=FALSE)
gr4jrouting.sim(U, x2, x3, x4, R_0 = 0, split = 0.9,
return_components = FALSE,
epsilon = hydromad.getOption("sim.epsilon"),
transformed=FALSE)
DATA |
time-series-like object with columns P (precipitation, mm) and E (potential evapo-transpiration, mm). |
U |
effective rainfall series. |
x1 |
maximum capacity of the production store (mm). |
x2 |
groundwater exchange coefficient (mm). |
x3 |
one day ahead maximum capacity of the routing store (mm). |
x4 |
time base of unit hydrograph UH1 (time steps). |
etmult |
Multiplier for the |
S_0 |
Initial soil moisture level as fraction of |
R_0 |
Initial groundwater reservoir level as fraction of
|
split |
Fraction to go into quick flow routing, usually fixed at 0.9. |
return_state |
to return the series U, S (storage) and ET (evapotranspiration). |
return_components |
to return the series Xr, Xd and R (reservoir level). |
epsilon |
values smaller than this in the output will be set to zero. |
transformed |
transform parameters before use to improve identifiability. They can be untransformed using |
The default parameter ranges were taken from the "80 intervals" given in Perrin et. al. (2003).
the simulated effective rainfall, a time series of the same length as the input series.
Felix Andrews felix@nfrac.org and Joseph Guillaume josephguillaume@gmail.com
Perrin, C., C. Michel, et al. (2003). "Improvement of a parsimonious model for streamflow simulation." Journal of Hydrology 279(1-4): 275-289
http://www.cemagref.fr/webgr/Modelesgb/gr4j/fonctionnement_gr4jgb.htm
hydromad(sma = "gr4j", routing = "gr4jrouting") to work
with models as objects (recommended).
## view default parameter ranges:
str(c(hydromad.getOption("gr4j"),
hydromad.getOption("gr4jrouting")))
data(HydroTestData)
mod0 <- hydromad(HydroTestData, sma = "gr4j", routing = "gr4jrouting")
mod0
## example from
## http://www.cemagref.fr/webgr/Scilab/CONT_EN/HELP_HYDROGR/c_GR4J.htm
dat <-
cbind(P = c(0,0,0,0,0,0.04,0.59,0.03,0.01,0.16,0.37,8.76,2.65,
0.05,0.02,0.02,0.38,0.00,0.02,0.46,4.46,7.71,5.71,0.79,1.33),
E = c(0,0,0,0,0,0.24,0.24,0.24,0.24,0.24,0.25,0.25,0.26,
0.27,0.28,0.32,0.33,0.34,0.35,0.36,0.36,0.37,0.37,0.38,0.38))
datz <- zoo(dat, as.Date("2000-01-01") + 1:nrow(dat))
modz <- hydromad(datz, sma = "gr4j", routing = "gr4jrouting",
x1 = 665, x2 = 1.18, x3 = 90, x4 = 3.8, S_0 = 0.6, R_0 = 0.7)
xyplot(predict(modz, return_state = TRUE, return_components = TRUE),
strip = FALSE, strip.left = TRUE)
## simulate with some arbitrary parameter values
mod1 <- update(mod0, x1 = 100, x2 = 20, x3 = 1, x4 = 10)
## plot results with state variables
testQ <- predict(mod1, return_state = TRUE)
xyplot(cbind(HydroTestData[,1:2], gr4j = testQ))
## show effect of increase/decrease in each parameter
parRanges <- c(hydromad.getOption("gr4j")[1],
hydromad.getOption("gr4jrouting"))
parsims <- mapply(val = parRanges, nm = names(parRanges),
FUN = function(val, nm) {
lopar <- min(val)
hipar <- max(val)
names(lopar) <- names(hipar) <- nm
fitted(runlist(decrease = update(mod1, newpars = lopar),
increase = update(mod1, newpars = hipar)))
}, SIMPLIFY = FALSE)
xyplot.list(parsims, superpose = TRUE, layout = c(1,NA),
strip = FALSE, strip.left = TRUE,
main = "Simple parameter perturbation example") +
layer(panel.lines(fitted(mod1), col = "grey", lwd = 2))