Skip to contents

Fetches the CarbonPlan OffsetsDB daily snapshots of voluntary carbon market projects and credits, stored as Parquet in a public S3 bucket. Covers Verra, ART TREES, Gold Standard, American Carbon Registry, and Climate Action Reserve.

Usage

co2_offsets_db(
  kind = c("projects", "credits"),
  date = NULL,
  max_lookback = 180L,
  refresh = FALSE
)

Arguments

kind

Character. "projects" (default) or "credits".

date

Optional character. ISO date of the snapshot to fetch. The function walks backwards from this date to find a published snapshot. Default Sys.Date().

max_lookback

Integer. How many days to walk back before giving up. Default 180, because CarbonPlan's publication cadence became irregular during 2026 and the last snapshot seen was 1 June 2026.

refresh

Re-download? Default FALSE.

Value

A file path to the downloaded Parquet file. The caller must have arrow or nanoparquet installed to read it.

Details

The CarbonPlan REST API at offsets-db-api.carbonplan.org was deprecated; this function uses the S3 bucket directly. Parquet reading requires an installed Parquet reader; the arrow package is the recommended suggest.

See also

Other voluntary markets: co2_cad_trust(), co2_vrod()

Examples

# \donttest{
op <- options(carbondata.cache_dir = tempdir())
try({
  path <- co2_offsets_db("projects")
  if (requireNamespace("arrow", quietly = TRUE)) {
  df <- arrow::read_parquet(path)
  }
})
#> Warning: ! Newest OffsetsDB snapshot is 99 days old (2026-06-01).
#>  CarbonPlan's publication cadence became irregular in 2026.
#>  Downloading OffsetsDB projects snapshot from 2026-06-01...
options(op)
# }