好，我有一些东西.也许您可以用它来修改您的功能:

Ok, I've got a little something. Maybe you an use it to modify your function:

## Generate your layer names (used for indexing later)

nms <- expand.grid(paste0('X',1969:2011),c("01","02","03","04","05","06","07","08","09","10","11","12"),'16')

nms <- apply(nms,1,function(x) paste0(x,collapse = '.'))

nms <- sort(nms)

## Generating fake raster brick

r <- raster()
r[] <- runif(ncell(r))

rst <- lapply(1:length(nms),function(x) r)

rst <- do.call(brick,rst)

names(rst) <- nms

现在，您可以使用层名称对砖进行索引了.循环浏览飓风季节(从Year -1开始):

And now you can index the brick with the layer names. Loop through the Hurricane Seasons (starting with Year1 -1):

for (ix in 1970:2011){


  sel <- rst[[c(grep(paste0(ix-1,'.12'),nms),sapply(paste0(0,1:4),function(x) grep(paste0(ix,'.',x),nms)))]]


  break ## in case you don't want to go through all iterations

  }

对于第一次迭代，我得到以下输出:

For the first iteration, I'm getting this output:

> sel
class       : RasterStack 
dimensions  : 180, 360, 64800, 5  (nrow, ncol, ncell, nlayers)
resolution  : 1, 1  (x, y)
extent      : -180, 180, -90, 90  (xmin, xmax, ymin, ymax)
coord. ref. : +proj=longlat +datum=WGS84 +ellps=WGS84 +towgs84=0,0,0 
names       :  X1969.12.16,  X1970.01.16,  X1970.02.16,  X1970.03.16,  X1970.04.16 
min values  : 5.988637e-06, 5.988637e-06, 5.988637e-06, 5.988637e-06, 5.988637e-06 
max values  :    0.9999771,    0.9999771,    0.9999771,    0.9999771,    0.9999771 

让我知道是否有帮助.

所以也许是一个更适用的示例:

So maybe a more applicable example:

(该函数假定您输入砖的层名称x的格式为Xyyyy.mm.dd)

(the function assumes your layer names of your input brick x have the format Xyyyy.mm.dd)

hadSSTmean <- function(x, years, first.range = 11:12, second.range = 1:4){

  nms <- names(x)

  mts <- c("01","02","03","04","05","06","07","08","09","10","11","12")

  xMeans <- vector(length = length(years)-1,mode='list')

  for (ix in 2:length(years){

    xMeans[[ix-1]] <- mean(x[[c(sapply(first.range,function(x) grep(paste0(years[ix-1],'.',mts[x]),nms)),sapply(1:4,function(x) grep(paste0(years[ix],'.',mts[x]),nms)))]])

  }

  return(do.call(brick,xMeans))
  # you could also return the list instead of a single brick
}