R如何同时将结果导出到多个栅格图层中?

时间:2018-11-28 16:01:10

标签: r loops dplyr lapply raster

首先,我导入一个表,其中包含多个采样站的每日数据(请参见下面的示例)。 随后,我将要为每个日期(日期)创建栅格图层,并在以后使用不同的名称导出这些栅格。 由于我有一个很长的时间序列(40年),其中包含每日数据,因此我希望进行一个循环(请参见下面的程序),但不能为输出栅格指定其他名称。因此,每次他粉碎我上一层。 我是循环编程的新手,您能帮我解决这个问题吗?

预先感谢

data=read.table(file="data.csv", sep=";", header=TRUE)
head(data)
LAMBX   LAMBY   DATE    S
600 24010   20180801    15.3
600 24010   20180802    12
600 24010   20180803    15
600 24010   20180804    14.8
600 24010   20180805    16.8
600 24010   20180806    15.1
601 24011   20180801    11
601 24011   20180802    14
601 24011   20180803    16.8
601 24011   20180804    15.1
601 24011   20180805    13.8
601 24011   20180806    12.7
602 24012   20180801    15.3
602 24012   20180802    14
602 24012   20180803    12
602 24012   20180804    16.8
602 24012   20180805    17.5
602 24012   20180806    15.1


library(dplyr)
library(sp)
library(raster)
# for each date I will do the following manipulations
for (i in data[,"DATE"]) {
# to delimit my study area
  table=filter(data, DATE == i & LAMBX %in% 601:602 & LAMBY %in% 24011:24012) 
# convert geographic coordinates
  table[,c("LAMBX", "LAMBY")]= 100*table[,c("LAMBX", "LAMBY")]

# spatialize the stations
  xy <- table[,c("LAMBX", "LAMBY")]
  sptable <- SpatialPointsDataFrame(coords = xy, data = table,
                                 proj4string = CRS("+proj=lcc +lat_1=46.8 +lat_0=46.8 +lon_0=0 +k_0=0.99987742 +x_0=600000 +y_0=2200000 +a=6378249.2 +b=6356515 +towgs84=-168,-60,320,0,0,0,0 +pm=paris +units=m +no_defs"))

# rasterize my SpatialPointsDataFrame
  rsptable <- rasterFromXYZ(as.data.frame(sptable)[, c("LAMBX", "LAMBY","S")],  crs="+proj=lcc +lat_1=46.8 +lat_0=46.8 +lon_0=0 +k_0=0.99987742 +x_0=600000 +y_0=2200000 +a=6378249.2 +b=6356515 +towgs84=-168,-60,320,0,0,0,0 +pm=paris +units=m +no_defs")
  # export my rasters layers
  writeRaster(rsptable, filename="S_date.tif", format="GTiff", overwrite=TRUE)
}

1 个答案:

答案 0 :(得分:0)

我在您的数据中看不到T_Q变量,因此我将改用S。他们的技巧是创建一个空列表对象,该对象会在每次循环迭代时填充。然后,您可以创建结果的 stack 。堆栈的每一层都是一个单独的日期。

library(dplyr)
library(sp)
library(raster)

data <- read.table(text = "
LAMBX   LAMBY   DATE    S
600 24010   20180801    15.3
600 24010   20180802    12
600 24010   20180803    15
600 24010   20180804    14.8
600 24010   20180805    16.8
600 24010   20180806    15.1
601 24011   20180801    11
601 24011   20180802    14
601 24011   20180803    16.8
601 24011   20180804    15.1
601 24011   20180805    13.8
601 24011   20180806    12.7
602 24012   20180801    15.3
602 24012   20180802    14
602 24012   20180803    12
602 24012   20180804    16.8
602 24012   20180805    17.5
602 24012   20180806    15.1", header = TRUE)

crs_string <- "+proj=lcc +lat_1=46.8 +lat_0=46.8 +lon_0=0 +k_0=0.99987742 +x_0=600000 +y_0=2200000 +a=6378249.2 +b=6356515 +towgs84=-168,-60,320,0,0,0,0 +pm=paris +units=m +no_defs"

res <- list()
for (i in seq_along(unique(data[,"DATE"]))){
  table <- filter(data,
                  DATE == unique(data[,"DATE"])[i] &
                  LAMBX %in% 601:602 &
                  LAMBY %in% 24011:24012)
  table[,c("LAMBX", "LAMBY")] <-  100 * table[,c("LAMBX", "LAMBY")]

  xy       <- table[,c("LAMBX", "LAMBY")]
  sptable  <- SpatialPointsDataFrame(coords = xy, data = table,
                                    proj4string = CRS(crs_string))
  rsptable <- rasterFromXYZ(
    as.data.frame(sptable)[, c("LAMBX", "LAMBY","S")],  crs = crs_string)
  res[[i]] <- rsptable
}

res <- stack(res)
writeRaster(res, filename="S_date.tif", format="GTiff", overwrite=TRUE)