AGAGE: Global Atmospheric Gas Measurements from 1978
by Ronald G. Prinn / Massachusetts Institute of Technology
Available on 1 platform
Sign in to view source links and access this dataset
Description
Continuous high-frequency measurements of methane, nitrous oxide, hydrogen, carbon monoxide, and other anthropogenic gases have been collected since 1978 at five globally distributed sites. The program, led by Ronald G. Prinn at MIT, evolved through three instrumentation phases (ALE, GAGE, AGAGE) and includes data through September 2020. AGAGE instruments specifically track gases contributing to stratospheric ozone destruction and the greenhouse effect.
Use Cases
Modeling long-term trends in atmospheric methane and nitrous oxide based on continuous multi-decadal measurements.
Analyzing the global distribution and transport of anthropogenic halocarbons based on data from multiple global stations.
Studying the impact of instrumentation advances (e.g., Medusa GC-MS systems) on the detection of volatile perfluorocarbons.
Validating satellite-derived atmospheric composition data against ground-based chromatographic measurements.
Strengths
Provides a continuous record spanning over 40 years, beginning in 1978.
Data is collected from a network of five globally distributed monitoring stations.
Includes measurements from advanced instrumentation like GC-MS and Medusa systems for improved compound detection.
Limitations
Column-level documentation is absent; field semantics must be inferred after download.
Row count is unknown, which may limit suitability assessment for large-scale ML tasks.
Provenance
Source
Massachusetts Institute of Technology (MIT)
Collection Method
Continuous high-frequency gas chromatographic measurements at fixed global stations.
Time Range
1978 - September 2020
Freshness
Data includes measurements through September 2020, but the last update date for the dataset is unknown.
Geography
Global (sites include Mace Head, Cape Grim, Trinidad Head, Barbados, American Samoa, Zeppelin, Jungfraujoch, and Goan)
License is unknown; users should verify terms before use.