Description Usage Arguments Value Author(s) References See Also Examples
This function conducts an Empirical Orthogonal Function analysis (EOF) via a covariance matrix (cov4gappy function) using "sinkr" package especially designed to handle gappy raster time series
1 2 3 |
rasterts |
Input raster time series as |
rastermask |
Either a |
nu |
Numeric. Defines the number of EOFs to return. Defaults to return the full set of EOFs. |
gapfill |
Character. Defines the algorithm to be used to interpolate pixels with incomplete temporal profiles. Accepts argument supported as method in function |
centered |
Logical. If TRUE center the input data before EOF computation. Argument is passed through to function |
scaled |
Logical. If TRUE scale the input data before EOF computation. Argument is passed through to function |
cores |
Integer. Defines the number of CPU to be used for multicore processing. Default to "1" core for singlecore processing. Applies only to the masking step. |
... |
Additional arguments to be passed through to function |
Object of class EOFstack
containing the following components:
eof.modes | EOF modes as RasterBrick object |
|
expansion_coefficients | EOF Expansion Coefficients (EC) as xts object |
|
total_variance | Numeric. Total variance of input raster time series | |
explained_variance | Numeric vector. Percentage of variance explained by each EOF mode with respect to the total variance of input raster time series | |
center | Center values from each pixel temporal profile as RasterLayer object (only computed if centered = TRUE ) |
|
scale | Scale values from each pixel temporal profile as RasterLayer object (only computed if scaled = TRUE )
|
Federico Filipponi
Bjoernsson, H. and Venegas, S.A. (1997). "A manual for EOF and SVD analyses of climate data", McGill University, CCGCR Report No. 97-1, Montreal, Quebec, 52pp.
Marc, T.H., Losch, M., Wenzel, M., Schroeter, J. (2013). On the Sensitivity of Field Reconstruction and Prediction Using Empirical Orthogonal Functions Derived from Gappy Data. Journal of Climate, 26, 9194-9205. pdf
eof
, rtsa.scaleEOF
, rtsa.gapfill
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 | ## Not run:
## create raster time series using the 'pacificSST' data from 'remote' package
require(remote)
data(pacificSST)
pacificSST[which(getValues(pacificSST == 0))] <- NA # set NA values
# create rts object
rasterts <- rts(pacificSST, seq(as.Date('1982-01-15'), as.Date('2010-12-15'), 'months'))
## generate raster mask
raster_mask <- pacificSST[[1]] # create raster mask
names(raster_mask) <- "mask"
values(raster_mask) <- 1 # set raster mask values
raster_mask[which(is.na(getValues(pacificSST[[1]])))] <- 0 # set raster mask values
## compute EOF
# compute the first 10 EOFs
eof_result <- rtsa.eof(rasterts=rasterts, nu=10)
# recursively compute the first 10 EOFs using raster mask
eof_result_recursive <- rtsa.eof(rasterts=rasterts, rastermask=raster_mask, nu=10, recursive=T)
# compute the first 10 EOFs applying centering, scaling
# and raster mask computing before eof computation
eof_res_masked <- rtsa.eof(rasterts=rasterts, rastermask="compute", nu=10, centered=T, scaled=T)
## End(Not run)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.