GB Carbon Intensity Interval Revision Evidence API
Pricing
from $1.00 / 1,000 metadata observations
GB Carbon Intensity Interval Revision Evidence API
Observe exact UTC half-hour GB grid intervals with forecast and nullable actual estimates. Match interval boundaries precisely; no regional, certified emissions or energy-price claims.
Pricing
from $1.00 / 1,000 metadata observations
Rating
0.0
(0)
Developer
Salih Can Kurnaz
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
6 days ago
Last modified
Categories
Share
Track up to five exact UTC half-hour starts through the national Carbon Intensity API. Input is YYYY-MM-DDTHH:MMZ with minute00 or30. Build one30-minute source query per requested start, then select the unique exact from/to interval. The provider can also return the interval ending at the requested start; it is excluded rather than silently shifting observations by30 minutes. Validate timestamps, half-hour duration and unique matching intervals. Keep forecast, nullable actual estimate and source intensity index, in gCO2/kWh. Missing actual remains null; zero remains zero. No fill/interpolation, independent grid measurement, aggregation to corporate emissions, forecast accuracy claim, certified carbon accounting, regional/postcode inference, energy tariff or trading advice. The provider’s actual values are estimated operational grid-intensity data and may be revised. Source classifications are declarations and are not an Actor-generated green score. Data supplied by the Carbon Intensity API (National Energy System Operator and partner organisations); original documentation/terms apply.
Input example
{"intervalStarts": ["2026-10-17T00:00Z","2026-10-17T00:30Z"],"maxResults": 2}
Outputs and change comparison
One saved observation is one dataset row with itemKey, sourceUrl, contentHash and observedAt. The SHA256 covers only selected semantic fields; retrieval times, API request URLs, download/popularity counters and people are excluded. Copy SNAPSHOT.previousHashes into the next input to compare the same exact watchlist. Statuses are NO_BASELINE, FIRST_SEEN, UNCHANGED or CHANGED. Unchanged observations are still billable. There is no hidden persistent baseline and no missing-row/deletion inference. OUTPUT records actual requestCount, rowsEmitted, source summaries, errors, status and complete. SNAPSHOT.complete is false after any error or when any requested observation could not be delivered. Reaching the selected result cap is PARTIAL_RESULT_CAP; a monetary budget stop is PARTIAL_BUDGET. Provider restrictions produce PARTIAL_PROVIDER_LIMIT. A saved observation means successful metadata extraction only, not an independent validation, compliance decision or purchase recommendation.
Pricing and bounded work
$0.005 startup plus $0.001 per saved observation, pay per event. Startup quantity follows Apify memory rules with minimum one. Minimum permitted run budget: $0.0055. Default 256 MB, timeout 120 seconds, no restart. Five identifiers and at most five rows; a 55-second fetch deadline and two-MiB decoded-body cap apply. API requests are made only after checking the affordable result capacity. Zero affordable observations causes zero source requests, including no provider request. Before each later request and dataset write the budget is checked again. Responses that do not produce a saved row are not charged as observations; startup can still apply. No browser, paid proxy, paid external API or credentials. No network retries, bypasses, open-ended search or bulk pagination. Shared-IP provider limits still apply across runs: avoid parallel requests that exceed provider policy and use supported bulk data channels for large extraction.
Failure and access behavior
Only constructed official HTTPS API endpoints are contacted. Only fixed official provider endpoints constructed from validated identifiers are used. Private addresses and credential-bearing/cross-provider redirects are rejected. HTTP403/429 or Retry-After stops further provider work; successful current responses can be retained. Unknown/missing identifiers are errors, never guessed missing/safe states. All-source failure exits nonzero after preserving OUTPUT; partial delivery preserves successful rows and explicit errors. Source records and raw declarations can change, and the Actor does not independently verify artifact bytes, publisher identity, signatures or legal rights. Public upstream APIs/data may be free; the fee covers the bounded selection, validation, comparison/output workflow and Apify execution. There is no affiliation, exclusive data access, guaranteed availability, completeness or revenue claim.