遇见数据集

County oil and gas production, United States

收藏
Zenodo2026-08-21 更新2026-10-01 收录
官方服务:

资源简介:

Monthly oil and gas production for 851 counties across 18 US states, 1920-04 to 2026-06 — 281,158 county-month rows, normalized to a single schema (period, oil_bbl, gas_mcf, co2_mcf). A further 1,095 rows are carried under entries that are NOT counties — Alabama's 2 state-waters areas and California's 3 offshore districts, which those regulators publish separately — so the archives hold 282,253 rows in all. The two counts are stated apart because the larger one used to be printed beside the county count. 17 of the 18 archives are compiled from a state regulator's own filings. Illinois is not and should not be cited as such: the Illinois county series is derived from the US Energy Information Administration (federal) for the monthly state total, and the Illinois Petroleum Resources Board (a trade association) for the county shares. No Illinois row carries a method — the derived axis is the county, not the month, and what declares it is the archive envelope (synthetic: true) and the header of every CSV built from it. co2_mcf is empty, not zero, wherever the carbon dioxide cannot be sized. A number means the state regulator classifies gas composition per record, in which case gas_mcf is hydrocarbon only and gas_mcf + co2_mcf is the volume as filed. An empty cell means that source reports no composition, so any carbon dioxide is still inside gas_mcf and cannot be separated. Editions before 2026-08 had no such column and published 5.04 billion Mcf of Jackson Dome CO2 as Mississippi natural gas; see the errata. Read KNOWN-DIVERGENCES.md before using the numbers. It records where this series knowingly differs from the US Energy Information Administration, which states cannot be validated against EIA at all because EIA republishes the same regulator filings, which years EIA publishes as a synthesized flat daily rate rather than a measurement, and which divergences remain open rather than explained. Several states are resampled from periodic filings rather than reported monthly; every such row is flagged in the data itself. Corrections. Figures later found to be wrong are recorded at https://www.pointerminerals.com/datasets/errata, which is maintained outside this record and covers every version. The files in this deposit are fixed at the moment it was deposited and are never edited; anything learned afterwards appears there.

提供机构:
Zenodo
创建时间:
2026-08-21
二维码
社区交流群
二维码
科研交流群
商业服务