SpatialVx (version 0.1-5)

S1: S1 Score, Anomaly Correlation

Description

Calculate the S1 score and anomaly correlation for a verification set.

Usage

S1(X, Y, gradFUN = "KernelGradFUN", ...)
ACC(X, Y, Xclim = NULL, Yclim = NULL)

Arguments

X,Y

m by n matrices giving the verification and forecast fields, resp.

Xclim,Yclim

m by n matrices giving the climatologies for X and Y, resp. If NULL, the result is simply a usual correlation.

gradFUN

character identifying a function used to calculate the gradient fields for X and Y. The default KernelGradFUN is to use a Laplacian of Gaussian kernel.

...

optional arguments to the gradFUN function. In the case of the default, the kernel can be changed (e.g., if only laplacian is desired), and optional arguments to the kernel2dmeitsjer function (in this case,

Value

• single numeric

Details

The S1 score is given by

S1 = 100*sum(abs(DY_i - DX_i))/sum(max(abs(DY_i),abs(DX_i))),

where DY_i (DX_i)is the gradient at grid point i for the forecast (verification). See Brown et al. (2012) and Thompson and Carter (1972) for more on this score.

The ACC is just the correlation between X - Xclim and Y - Yclim.

References

Brown, B.G., Gilleland, E. and Ebert, E.E. (2012) Chapter 6: Forecasts of spatial fields. pp. 95--117, In Forecast Verification: A Practitioner's Guide in Atmospheric Science, 2nd edition. Edts. Jolliffee, I. T. and Stephenson, D. B., Chichester, West Sussex, U.K.: Wiley, 274 pp.

Thompson, J. C. and Carter, G. M. (1972) On some characteristics of the S1 score. J. Appl. Meteorol., 11, 1384--1385.

See Also

kernel2dmeitsjer

Examples

data(UKobs6)
data(UKfcst6)
S1(UKobs6,UKfcst6)