Access and process COPERNICUS Global Land Vegetation products

The aim of copernicus is to access to the COPERNICUS Global Land Vegetation data portal. The package provides functions to easily download, extract, convert, subset and project COPERNICUS data. The user should be registered to access data. The package is at its early development stage. Use with care!
Please report bugs to antoinexstevens@gmail.com

To install copernicus, use:

devtools::install_github("antoinestevens/copernicus")

One of the dependencies of copernicus is the rhdf5 package. It can be installed from bioconductor

source("https://bioconductor.org/biocLite.R")
biocLite("rhdf5")

File extraction and image projection is done via GDAL. GDAL can be otbtained through for instance the OSGeo4W project (windows).

Download

download_copernicus wil download zip files on the data portal based on an extent or given tiles (h,v) pairs and a product name. Only the following products are available: NDVI_V1_V1 (Normalized Difference Vegetation Index - VGT instrument), NDVI_V1_V2 (Normalized Difference Vegetation Index - PROBAV instrument),LAI (Leaf Area Index),FCOVER (Fraction of Vegetation Green Cover), FAPAR (Fraction of Absorbed Photosynthetically Active Radiation),VCI (Vegetation Condition Index),VPI (Vegetation Productivity Index), DMP (Dry Matter Productivity),BA (Burnt Areas). The availability of the data between two dates can be checked with check_time_copernicus and at a given location with check_tile_copernicus. One can get the URL's of images for a given time period, location and product name with get_url_copernicus. Information included in the name of the files (e.g. aglorithm version, instrument, acquisition date) can be retrieved with scan_file_copernicus. The COPERNICUS tiling system can be created with gen_tile_copernicus and fetched with get_tile_copernicus which will retrieve tiles info (h,v,bounding box in LatLon) based on an extent or (h,v) pair.

Extracting images

File conversion can be done with extract_copernicus, which will transfrom raw h5 files to tif, by converting DN to physical values using the associated gain and offset of each product. The function can further subset data on a given extent, resample to a given resolution and project to a given coordinate system. One can also choose which layers of the h5 files can be extracted. Quality Flags can be extracted with convert_qf_copernicus.

Compare raster time series

compare_raster_time and compare_raster_space allow to compare two rasterBrick or rasterStack time series in terms of their spatial and temporal consistency, following the methods explained in Meroni et al. (2012).

Package options

Some of the package default options are set with copernicus_options, such as the download path, user name and password.

TODO:

• wrapper around download_copernicus and extract_copernicus
• check file download integrity
• allow different output format
• change dependency from rhdf5 to h5 (allows partial I/O! --> speed-up!)
• vignette
• add verbose arguments