ardglad

The goal of ardglad is to provide access to the Global Land Analysis & Discovery (GLAD) Landsat Analysis Ready Data (ARD). The GLAD ARD represents a 16-day time series of globally consistent, tiled Landsat normalized surface reflectance from 1997 to the present operationally updated every 16 days.

This R package interacts with the GLAD ARD API to download and process the data.

Installation

You can install the development version of ardglad like so:

# install.packages("pak")
pak::pkg_install("TESS-Laboratory/ardglad")

Example

Here is an example of how to generate the urls for the GLAD ARD tiles for an example location in Rondônia, Brazil.

library(ardglad)

rondonia_eg <- sf::st_point(c(-62.625, -8.989)) |>
  sf::st_sfc(crs = 4326) |>
  sf::st_buffer(20000)

ron_urls <- ard_glad_urls(
  rondonia_eg,
  start_date = "2023-07-01",
  end_date = "2023-07-30"
)

print(ron_urls)
#> 
#> ── < ARD GLAD urls >
#> 
#> ── [[1]] 2023-06-26:2023-07-11 ─────────────────────────────────────────────────
#> https://glad.umd.edu/dataset/glad_ard2/09S/062W_09S/1001.tif
#> https://glad.umd.edu/dataset/glad_ard2/08S/062W_08S/1001.tif
#> 
#> ── [[2]] 2023-07-12:2023-07-27 ─────────────────────────────────────────────────
#> https://glad.umd.edu/dataset/glad_ard2/09S/062W_09S/1002.tif
#> https://glad.umd.edu/dataset/glad_ard2/08S/062W_08S/1002.tif
#> 
#> ── [[3]] 2023-07-28:2023-08-12 ─────────────────────────────────────────────────
#> https://glad.umd.edu/dataset/glad_ard2/09S/062W_09S/1003.tif
#> https://glad.umd.edu/dataset/glad_ard2/08S/062W_08S/1003.tif

And now let’s download the data and plot it. Note that although an object of class ard_glad is returned, in reality this is just a list of lists that contain information about each raster; to convert this to a more general use object for R, use as.list which will coerce the object to a list of SpatRasters.

Note we can use future to parallelise some of the operations such as downloads and masking - by setting the plan here parallisation will be carried out automatically.

future::plan("multisession", workers = 3)

ron_ras <- ard_glad_download(ron_urls, "rondonia_glad")

print(ron_ras)
#> 
#> ── < 3 ARD GLAD mosaics >
#> 
#> ── [[1]] 2023-06-26:2023-07-11 ─────────────────────────────────────────────────
#> source :
#> /home/hugh/TESS-Projects/ardglad/rondonia_glad/rondonia_glad_2023_06_26_2023_07_11/rondonia_glad_2023_06_26_2023_07_11.vrt
#> dimensions : 8004, 4004, 8 (nrow, ncol, nlyr)
#> resolution : 0.00025, 0.00025 (x, y)
#> extent : ext(-63.0005, -61.9995, -10.0005, -7.9995) (xmin, xmax, ymin, ymax)
#> coord. ref. : lon/lat WGS 84 (EPSG:4326)
#> band names : B, G, R, N, S1, S2, T, QA
#> 
#> 
#> ── [[2]] 2023-07-12:2023-07-27 ─────────────────────────────────────────────────
#> 
#> source :
#> /home/hugh/TESS-Projects/ardglad/rondonia_glad/rondonia_glad_2023_07_12_2023_07_27/rondonia_glad_2023_07_12_2023_07_27.vrt
#> dimensions : 8004, 4004, 8 (nrow, ncol, nlyr)
#> resolution : 0.00025, 0.00025 (x, y)
#> extent : ext(-63.0005, -61.9995, -10.0005, -7.9995) (xmin, xmax, ymin, ymax)
#> coord. ref. : lon/lat WGS 84 (EPSG:4326)
#> band names : B, G, R, N, S1, S2, T, QA
#> 
#> 
#> ── [[3]] 2023-07-28:2023-08-12 ─────────────────────────────────────────────────
#> 
#> source :
#> /home/hugh/TESS-Projects/ardglad/rondonia_glad/rondonia_glad_2023_07_28_2023_08_12/rondonia_glad_2023_07_28_2023_08_12.vrt
#> dimensions : 8004, 4004, 8 (nrow, ncol, nlyr)
#> resolution : 0.00025, 0.00025 (x, y)
#> extent : ext(-63.0005, -61.9995, -10.0005, -7.9995) (xmin, xmax, ymin, ymax)
#> coord. ref. : lon/lat WGS 84 (EPSG:4326)
#> band names : B, G, R, N, S1, S2, T, QA

par(mfrow = c(1, 3))
plot(ron_ras)

We can also query specific moments through time, by specifying multiple start_date values, like so:

ron_urls_97_23 <- ard_glad_urls(
  rondonia_eg,
  c("1997-06-30", "2010-06-01", "2023-06-30")
)

Now we can download these data and mask clouds, shadows and haze, ready for further analysis:

ron_ras_97_23_masked <- ard_glad_download(ron_urls_97_23, "rondonia_glad") |>
  ard_glad_mask()

par(mfrow = c(1, 3))
plot(ron_ras_97_23_masked)

As a bonus, here is a false colour composite:

par(mfrow = c(1, 3))
plot(ron_ras_97_23_masked, r = 4, g = 3, b = 2)