weightmask: Returns cost733 Style Weighting Mask

In retostauffer/Rcost: Cost 733 WLK Classification in R

Description

The cost733 WLK weather type classification algorithm uses a weight mask. This allows to “focus” on specific parts of the area of interest. There are three different zones called “core”, “mid” and “margin”.

By default (cost733) uses a stripsize of 5 (cut's the area in five rows and five columns). This creates a rectangular core of one fifth times one fifth of the full area and two surrounding “bands”. The inner “band” is the “mid” zone, the outer one (reaching the edges of the zone) the “margin”.

Each of these three zones gets a weight, by default 15 for the “core”, 2 for “mid”, and 1 for “margin”. During computation values within these zones are weighted according to the weights set.

Usage

## Basic usage
weightmask( lon, lat, weights=c(15,2,1), stripsize=5 )

Arguments

lon	numeric vector of longitude values.
lat	numeric vector of latitude values.
weights	numeric vector of length three with the weights for the three zones “core”, “mid” and “margin”.
stripsize	numeric value which is used to define the zones.

Value

Returns a matrix of size length(lon) times length(lat) with the corresponding weights.

Author(s)

Reto Stauffer

See Also

weightmaskZAMG, getClassification

Examples

lons <- seq( 5,30,by=0.25)
lats <- seq(40,60,by=0.25)
w1 <- weightmask( lons, lats )
w2 <- weightmask( lons, lats, weights=c(3,2,1) )
w3 <- weightmask( lons, lats, stripsize = 10 )
w4 <- weightmask( lons, lats, weights=c(3,2,1), stripsize = 4.5 )

# Pretty ugly plots
par(mfrow=c(2,2))
   image(w1,zlim=c(1,15),col=1:15,main="weight mask c(15,2,1) stripsize 5")
   image(w2,zlim=c(1,15),col=1:15,main="weight mask c(3,2,1) stripsize 10")
   image(w3,zlim=c(1,15),col=1:15,main="weight mask c(15,2,1) stripsize 5")
   image(w4,zlim=c(1,15),col=1:15,main="weight mask c(3,2,1) stripsize 4.5")

retostauffer/Rcost documentation built on May 7, 2019, 9:43 p.m.