Land Cover: Grassland, Area from CGLS

Latest observation is 2019; the catalog broadly reports to 2025, so this data is 6 year(s) behind (stale).

Method: Compares the dataset's latest non-projected period with the year most datasets in the catalog end in. Up to 2 years behind is current — annual data is routinely published a year or two late — and more than 5 is stale. Measured from the data, not from when Interva last crawled it.

0 of 36 series skip at least one year inside their own span; the median series reports 100% of its years.

Method: Per series, years present divided by years from its first to its last observation; the median across series is reported. Annual periods are assumed, which holds for every series in the catalog today.

Covers 36 of the 251 countries in the catalog.

Method: Counts distinct geographies with at least one observation. Not scored: a regional dataset covering its region completely is not deficient for omitting the rest of the world.

The median series has 5 observations; the earliest begins in 2015.

Method: Counts non-projected observations per series. Not scored: a long history is useful for trend work and irrelevant for a cross-section.

No observation has been restated in any vintage Interva has recorded.

Method: Counts values the publisher changed between Interva's publishes. Not scored: revising estimates is ordinary and often a sign of care, and the count is here so a reader citing a figure knows whether it moves.

No declared schema or validation rules are attached to this dataset yet.

No field-level validation rules are attached to this dataset yet.

Cross-series consistency rules are not defined yet. The obvious first one — whether two publishers measuring one concept agree within a stated tolerance — is a candidate.