skillMap: A wrapper of 'spatialPlot' for the creation of verification...

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

View source: R/skillMap.R

Description

A wrapper of spatialPlot for the creation of verification maps for seasonal forecast systems. It provides a convenient interface for map.stippling

Usage

1
2
3
skillMap(easyVeriGrid, stippling = list(threshold = NULL, condition =
  NULL), stippling.point.options = NULL, backdrop.theme = NULL,
  title = NULL, ...)

Arguments

easyVeriGrid

A climatological grid with a verification output. See details

stippling

A key-value list of arguments passed to map.stippling: threshold and condition. Ignored by default, returning a map without stippling.

stippling.point.options

Default to NULL. Further graphical arguments passed to points (e.g. cex, pch etc.)

backdrop.theme

See spatialPlot

title

Title of the plot

...

Further graphical options passed to spatialPlot.

Details

The function applies the spatialPlot function, that in turn uses lattice-methods.

Graphical options

Some examples of specific map graphical options are available in the help of function spplot. In addtion, fine-tuning of the resulting plots can be obtained using the arguments of lattice plots. For an overview, see the help of function xyplot.

Value

A lattice plot of class “trellis”.

Author(s)

J. Bedia

See Also

The bridging function easyVeri2grid from package transformeR allows for the conversion of verification outputs from package easyVerification to the climate4R data structure used by the function.

Many different aspects of the plot can be controlled passing the relevant arguments to spplot.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
## Not run: 

# The package 'easyVerification' will be used to calculate the RPSS:
require(easyVerification)
require(transformeR)
data(tas.cfs)
data(tas.ncep)
# First of all, a data subset is done, considering a spatial domain centered on the North Atlantic:
tas.cfs2 <- subsetGrid(tas.cfs, lonLim = c(-100, 40), latLim = c(-5, 75))
# The same is done with the reanalysis dataset:
tas.ncep2 <- subsetGrid(tas.ncep, lonLim = c(-100, 40), latLim = c(-5, 75))
# In the next step, the reanalysis data are interpolated to the hindcast grid:
tas.ncep2.int <- interpGrid(tas.ncep2, new.coordinates = getGrid(tas.cfs2), method = "nearest")
# We compute the Ranked Probabiloity SKill Score using veriApply, and a cross-validation strategy:
ev <- easyVerification::veriApply(verifun = "EnsRpss",
                                  fcst = tas.cfs2$Data,
                                  obs = tas.ncep2.int$Data,
                                  prob = 1:2/3,
                                  tdim = 2,
                                  ensdim = 1,
                                  parallel = TRUE,
                                  ncpus = 3, 
                                  strategy = "crossval")
# The bridging function 'easyVeri2grid' converts the object returned by 'veriApply' 
#    to a 'climate4R' climatological grid: 

easyVeriGrid <- easyVeri2grid(ev$skillscore,
                              obs.grid = tas.ncep2.int,
                              verifun = "EnsRpss")

# A basic RPSS map, using the spatialPlot defaults:
skillMap(easyVeriGrid = easyVeriGrid, backdrop.theme = "coastline")

# Stippling. Mark significant RPSS values at the 95% ci
thresh <- ev$skillscore.sd*qnorm(0.95)

skillMap(easyVeriGrid = easyVeriGrid,
         stippling = list(threshold = thresh, condition = "GT"), # GT = greater than
         stippling.point.options = list(pch = 19, cex = .2, col = "black"),
         backdrop.theme = "coastline")

# Further customization: For instance a more elaborated title, colorblind-friendly palette etc.:

cb.colors <- colorRampPalette(rev(RColorBrewer::brewer.pal(11, "Spectral")))
skillMap(easyVeriGrid = easyVeriGrid,
         stippling = list(threshold = thresh, condition = "GT"),
         stippling.point.options = list(pch = 19, cex = .2, col = "black"),
         backdrop.theme = "coastline",
         at = seq(-0.7, 0.7, .05),
         set.max = 0.7, set.min = -0.7,
         scales = list(draw = TRUE, alternating = 3),
         colorkey = list(space = "bottom"),
         col.regions = cb.colors(51),
         title = paste("Ranked Probability Skill Score",
                       "Forecasting System: CFSv2 - 24 members", 
                       "November Initializations",
                       "Observing System: NCEP/NCAR Reanalysis 1",
                       "t2m DJF 1983-2010", sep = "\n")
)

## End(Not run)

SantanderMetGroup/visualizeR documentation built on Oct. 21, 2018, 10:15 a.m.