RGISTools (version 1.0.2)

lsCloudMask: Create cloud masks for Landsat images

Description

lsCloudMask creates a cloud mask derived from the band for quality assurance (BQA) from Landsat-7 or Landsat-8 time series. The function is applied to untarred images, such as those resulting from lsDownload or lsDownSearch. The result is a new image band, called cloud (CLD), that is saved as separate GoeTiffs.

Usage

lsCloudMask(
  src,
  AppRoot,
  out.name,
  ls8 = TRUE,
  overwrite = FALSE,
  verbose = FALSE,
  ...
)

Arguments

src

the path to the folder with the untarred images from Landsat-7 or Landsat-8.

AppRoot

the directory where the cloud masks are saved.

out.name

the name of the folder that stores the outputs. If the arguemnt is not defined the folder will be named as "CloudMask".

ls8

logical argument. Defines the lansat satellite number. If TRUE clouds for landsat-8 are computed, otherwise, landsat 4-7 clouds are computed.

overwrite

logical argument. If TRUE, overwrites the existing images with the same name.

verbose

logical argument. If TRUE, the function prints the running steps and warnings.

...

arguments for nested functions.

• dates a vector with the dates being considered for creating cloud masks. This argument is optional.

Value

this function does not return anything. It saves the cloud masks (CLD) as GTiff files in the AppRoot directory.

Details

The valid range for the sensitivity threshold is 0-80000. By defualt, the argument is set to 28000. We recommend 6000 and 28000 for Landsat-7 and Landsat-8 respectively. The NA and 1 values of the mask represent cloudy and clear-sky pixels pixels respectively.

Examples

# NOT RUN {
# load a spatial polygon object of Navarre
data(ex.navarre)
wdir <- file.path(tempdir(),"Path_for_downloading_folder")
print(wdir)

# search and download images from Landsat-8 between
# 01-01-2018 and 20-01-2018 for the region of Navarre
lsDownSearch(satellite = "ls8",
             username = "username",
             password = "password",
             startDate = as.Date("01-01-2018", "%d-%m-%Y"),
             endDate = as.Date("20-01-2018", "%d-%m-%Y"),
             pathrow = list(c(200, 31), c(200, 30)),
             untar = TRUE,
             AppRoot = wdir)
           
# define the path where the GTiff images are located
wdir.ls8 <- file.path(wdir,"Landsat8")
wdir.ls8.untar <- file.path(wdir.ls8,"untar")
            
# mosaic and crop the imagery
lsMosaic(src = wdir.ls8.untar,
         AppRoot = wdir.ls8,
         out.name = "Navarre",
         extent = ex.navarre,
         gutils = TRUE, # using gdalUtils
         overwrite = TRUE) # overwrite
         
# generate the path where mosaicked images are located
wdir.ls8.navarre <- file.path(wdir.ls8, "Navarre")

# calculate the cloud mask from QC layer
lsCloudMask(src=wdir.ls8.navarre,
            overwrite=TRUE) 
            
# load the B1 layer and calculate the CLD layer
files.ls8.navarre.path <- list.files(wdir.ls8.navarre,
                                     full.names = TRUE,
                                     recursive = TRUE,
                                     pattern = "\\.tif$")
tiles.ls8.cld <- files.ls8.navarre.path[grepl("CLD",files.ls8.navarre.path)]
tiles.ls8.b1 <- files.ls8.navarre.path[grepl("B1.tif",files.ls8.navarre.path)]
img.ls8.cld <- lapply(tiles.ls8.cld,raster)
img.ls8.b1 <- lapply(tiles.ls8.b1,raster)

# calculate cloud free b1 layers
img.ls8.b1.cloud.free <- img.ls8.b1[[1]] * img.ls8.cld[[1]]
spplot(img.ls8.b1.cloud.free)
# }