Get Sentinel-2 data and validation

biodivMapR aims at producing spectral diversity maps based on multidimensional raster data. While its application is technically not limited to optical imagery, users experimenting with other types of remote sensing data are expected to make sure that the input information used to produce remotely sensed diversity maps are actually relevant to describe vegetation biodiversity as intended in biodivMapR.

Data requirements

biodivMapR requires raster data as input. Raster data are handled with the R package terra. Data format should then be compliant with the drivers available with this package.

If using multispectral/hyperspectral optical data, the central wavelength should be properly identified for each spectral band provided. Central wavelength is required for the computation of spectral indices or the application of continuum removal.

Download Sentinel-2 image sample

The S2 subset used for this tutorial corresponds to a forested area in Cameroon. We previously identified a S2 acquisition with minimum cloud cover. Visualization can be done with the Copenicus Browser

Once a correct acquisition identified, Sentinel-2 imagery is obtained with the R package preprocS2.

The chunk of code below downloads Sentinel-2 products corresponding to a bounding box from the STAC catalog provided by Microsoft Planetary Computer.

Refl_L2A is the L-2A reflectance for the bounding box
provider_mask is the corresponding SCL (scene classification layer)
Binary_mask is a binary vegetation mask extracted from the SCL
vegetation_mask is an updated binary vegetation mask after application of NDVI and radiometric thresholds to remove non-vegetated, shaded and hazy pixels (very basic filtering)

Additionally, The rasters corresponding to the geometry of acquisition can be downloaded from the CDSE, if valid authentication is provided and geomAcq = T.

# load useful libraries 
library(preprocS2)
library(sf)

################################################################################
# 1- define input & output directories
output_dir_s2 <- './S2_data2'
datetime <- '2018-12-25'
siteName <- 'S2_subset_Cameroon'

bbox <- sf::st_bbox(obj = c('xmin' = 357001, 'ymin' = 341701, 
                            'xmax' = 366730, 'ymax' = 351236))
sf::st_crs(x = bbox) <- 32633

# 2- get tiling grid kml from https://sentiwiki.copernicus.eu/web/s2-products
path_S2tilinggrid <- 'Sentinel-2_tiling_grid.kml'

# 3- get S2 acquisition and geometry of acquisition for study area
# https://shapps.dataspace.copernicus.eu/dashboard/#/account/settings
id <- "sh-XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX"
secret <- "XXXXXXXXXXXXXXXXXXXXXXXXXXXX"

# tip to save credential on your computer: run the following command : 
# > file.edit("~/.Renviron")
# add ID and password for CDSE as well as CURL_SSL_BACKEND using this template
# > CDSE_ID="sh-XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX"
# > CDSE_SECRET="XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX"
# > CURL_SSL_BACKEND="openssl"
# save .Renviron file and restart R session then load your credential this way: 
# > id <- Sys.getenv("CDSE_ID")
# > secret <- Sys.getenv("CDSE_SECRET")
authentication <- list('id' = id, 'pwd' = secret)

# 3- get S2 acquisition for study area
list_files <- preprocS2::get_s2_raster(bbox = bbox, geomAcq = T,
                                       authentication = authentication, 
                                       path_S2tilinggrid = path_S2tilinggrid, 
                                       datetime = datetime, 
                                       output_dir = output_dir_s2, 
                                       siteName = siteName)
# Sentinel-2 L2A reflectance
Refl_L2A <- list_files$Refl_L2A
# Sentinel-2 binary mask identifying vegetation, discarding clouds & shadows
vegetation_mask <- list_files$vegetation_mask
# Sentinel-2 mask from provider ( = SCL from ESA products)
SCL <- list_files$provider_mask
# Sentinel-2 geometry of acquisition (if requested from CDSE)
geometryAcquisition <- list_files$geometryAcquisition

Once the image is downloaded, it can be processed with biodivMapR.

Download vector files used for ‘validation’

A set of vector files is downloaded here, corresponding to circular plots identified in the raster. It will be used in the final step of the tutorial (How to perform validation?.).

The following code allows downloading the zipfile containing the vector data, and unzipping it in the same directory as the one including the raster data.

library(zip)
# name zip file including plots located on the tile
output_dir_val <- file.path(output_dir_s2, 'GroudData')
dir.create(path = output_dir_val, showWarnings = F, recursive = T)
destzip <- file.path(output_dir_val,'S2A_T33NUD_Plots.zip')
# url for the zip file
url <- 'https://gitlab.com/jbferet/myshareddata/-/raw/master/biodivMapR_S2_Sample/VECTOR/S2A_T33NUD_Plots.zip'
download.file(url = url, destfile = destzip)
destunz <- file.path(tmpdir,'S2A_T33NUD_Plots')
unzip(zipfile = destzip, exdir = destunz)
unlink(x = destzip, recursive = T)

Now that the necessary data is downloaded, we can start the tutorial.

Jean-Baptiste Féret, Florian de Boissieu

2025-01-24

Data requirements

Download Sentinel-2 image sample

Download vector files used for ‘validation’