NOAA GOES Aerosol Detection Product (ADP)

This article demonstrates a compact workflow for NOAA GOES Aerosol Detection Product (ADP) data.

This vignette runs its live workflow when rendered locally. The heavy download, processing, extraction, and plotting chunks are skipped automatically on CI, CRAN checks, and pkgdown builds; set AMADEUS_RUN_VIGNETTES=true to force live execution in those environments.

Data description

GOES ADP is a geostationary satellite aerosol detection product that reports smoke and dust presence from ABI observations.

Component	Description
Products	`ADP-C` (CONUS), `ADP-F` (Full Disk), `ADP-M` (Mesoscale)
Variables available in `process_covariates(covariate = "goes", ...)`	`Smoke`, `Dust`
Spatial domain	GOES-East (`satellite = "16"`) and GOES-West (`satellite = "18"`) viewing domains; sector depends on product (`C`, `F`, `M`)
Spatial resolution	Native GOES fixed-grid product (approximately 2 km at nadir; effective footprint increases away from nadir)
Temporal domain	Daily archive by UTC day (`YYYY-MM-DD`), with intra-day granules available throughout the day
Temporal resolution	High-frequency geostationary updates (sector-dependent cadence; often every few minutes for CONUS/Full Disk and faster for Mesoscale)
Output format	NetCDF (`.nc`) files from NOAA Open Data S3

Available inputs and data availability

download_data(dataset_name = "goes", ...) wraps download_goes().

satellite accepts "16" (GOES-East) or "18" (GOES-West).
product accepts "ADP-C" (CONUS), "ADP-F" (Full Disk), or "ADP-M" (Mesoscale).
date accepts either a single day or a start/end range in YYYY-MM-DD format.
GOES ADP does not require authentication.
process_covariates(covariate = "goes", ...) supports variable = "Smoke" or "Dust".

Download representative requests

directory_to_save <- file.path(tempdir(), "goes_workflow")
download_goes(
  date = c("2024-01-01","2024-01-01"),
  satellite = "16",
  product = "ADP-C",
  directory_to_save = directory_to_save,
  acknowledgement = TRUE
)

Process one workflow-ready data product

goes_path <- file.path(directory_to_save, "ABI-L2-ADPC", "2024", "001")

processed_mean <- process_covariates(
  covariate = "goes",
  variable = "Smoke",
  date = "2024-01-01",
  path = goes_path,
  daily_agg = TRUE,
  fun = "mean"
)

processed_sum <- process_covariates(
  covariate = "goes",
  variable = "Smoke",
  date = "2024-01-01",
  path = goes_path,
  daily_agg = TRUE,
  fun = "sum"
)

Choosing a daily aggregation

Use fun = "mean" when you want a daily field that behaves like a fractional or probability-like surface. For GOES Smoke/Dust, the mean is usually the better default for exposure-oriented summaries because it represents the typical share of available granules indicating smoke or dust at each pixel during the day.

Use fun = "sum" when you want to integrate granule-level detections into a count-like daily total. This can be useful for identifying places with repeated detections across the day, but it is a coarse approximation because GOES granule cadence varies by sector, satellite operations, and data availability.

For most epidemiologic or environmental exposure workflows, start with fun = "mean" unless your analysis specifically needs a detection-frequency metric.

Computational notes: For memory, runtime, and parallel-processing guidance, see vignette("computational_considerations", package = "amadeus").

Calculate covariates at points

domain_x <- c(terra::xmin(processed_mean), terra::xmax(processed_mean))
domain_y <- c(terra::ymin(processed_mean), terra::ymax(processed_mean))
domain_dx <- diff(domain_x)
domain_dy <- diff(domain_y)

candidate_xy <- expand.grid(
  lon = seq(domain_x[1] + 0.12 * domain_dx, domain_x[2] - 0.12 * domain_dx, length.out = 5),
  lat = seq(domain_y[1] + 0.12 * domain_dy, domain_y[2] - 0.12 * domain_dy, length.out = 5)
)
example_points_sf <- sf::st_as_sf(
  candidate_xy,
  coords = c("lon", "lat"),
  crs = 4326
)
example_points_sf$site_id <- paste0("site_", seq_len(nrow(example_points_sf)))

point_values <- calculate_covariates(
  covariate = "goes",
  from = processed_mean,
  locs = example_points_sf,
  locs_id = "site_id",
  radius = 0,
  fun = "mean",
  geom = "sf"
)

Visualize the point outputs

smoke_col <- grep("^Smoke_", names(point_values), value = TRUE)[1]
if (is.na(smoke_col) || smoke_col == "") {
  excluded <- c("site_id", "h3_id", attr(point_values, "sf_column"))
  fallback_cols <- setdiff(names(point_values), excluded)
  fallback_numeric <- fallback_cols[
    vapply(point_values[fallback_cols], is.numeric, logical(1))
  ]
  smoke_col <- fallback_numeric[1]
}

ggplot(data = point_values) +
  geom_sf(aes(color = .data[[smoke_col]])) +
  ggtitle("Daily GOES Smoke values")

Polygon (areal) extraction example

The same processed daily raster can be summarized over polygon features. This example builds two small synthetic rectangles from the processed raster extent and extracts the mean daily Smoke value for each area.

poly_x <- domain_x[1] + c(0.20, 0.35, 0.50) * domain_dx
poly_y <- domain_y[1] + c(0.20, 0.40, 0.60) * domain_dy
make_rect <- function(xmin, xmax, ymin, ymax) {
  sf::st_polygon(list(rbind(
    c(xmin, ymin),
    c(xmax, ymin),
    c(xmax, ymax),
    c(xmin, ymax),
    c(xmin, ymin)
  )))
}

example_polys_sf <- sf::st_sf(
  area_id = c("area_1", "area_2"),
  geometry = sf::st_sfc(
    make_rect(poly_x[1], poly_x[2], poly_y[1], poly_y[2]),
    make_rect(poly_x[2], poly_x[3], poly_y[2], poly_y[3]),
    crs = 4326
  )
)

polygon_values <- calculate_covariates(
  covariate = "goes",
  from = processed_mean,
  locs = example_polys_sf,
  locs_id = "area_id",
  radius = 0,
  fun = "mean",
  geom = "sf"
)
polygon_values

Kyle Messier, with assistance from GitHub Copilot

2026-05-20