GRAMI · GWTC-v3 · ERA5-based
Global Reanalysis Air Mass Index
GRAMI is the next generation of the Gridded Weather Typing Classification: a global, reanalysis-based classification of daily near-surface air-mass conditions derived from hourly ERA5 data.
Overview
GRAMI is a gridded, globally consistent way to describe the multivariate character of daily weather as an air-mass condition rather than as a single meteorological variable. Each location-day is classified relative to what is typical for that location and time of year.
GRAMI is available in three spatial-smoothing versions: an unsmoothed product that preserves the native grid-cell classification, a mesoscale-smoothed product using a 200-km smoothing scale, and a synoptic-scale-smoothed product using a 500-km smoothing scale.
Local and seasonal context
A warm day in Cleveland in January, a cool day in Miami in August, and a dry day in the tropics are each evaluated relative to their own local climatology.
Multivariate weather
The index captures combinations of near-surface conditions that matter for human health, ecosystems, hazards, and climate impacts.
Global reanalysis backbone
GRAMI moves the weather-typing framework to an ERA5-based foundation for a stable, research-grade, global historical record.
Annual NetCDF archive
GRAMI annual NetCDF files are available for each spatial-smoothing product. Each file contains daily global GRAMI classifications for one calendar year.
Unsmoothed GRAMI
Native grid-cell GRAMI classifications with no spatial smoothing applied.
Mesoscale-smoothed GRAMI
Spatially smoothed classifications designed to reduce local grid-cell noise while retaining regional-scale air-mass structure.
Synoptic-scale-smoothed GRAMI
Spatially smoothed classifications designed to emphasize broad air-mass patterns and large-scale atmospheric structure.
Custom point, bounding-box, and time-subset downloads are planned for a later data-access interface. For now, users can download annual NetCDF files directly from the separate archive browser.
Planned custom data access
GRAMI is being prepared for customized downloads by location, region, and time period. The annual archive above is available first; point and bounding-box subsetting will be added through a data-access service.
Single-point time series
Planned: enter a latitude and longitude to download the full GRAMI history for the nearest grid cell.
Bounding-box subset
Planned: select a latitude/longitude box and optional date range to download a regional GRAMI subset.
Global time subset
Planned: download the whole world for a selected month, year, or custom time range.
Programmatic access
Planned examples will show how to use GRAMI in MATLAB, Python, R, and command-line workflows.
Documentation
Documentation will include class codes, metadata, file structure, citation guidance, and version history.
How the index works
GRAMI classifies daily air-mass conditions using a geographically and seasonally relative framework. Rather than labeling weather only by raw temperature, humidity, or pressure, each day is interpreted relative to what is normal for that grid cell and time of year.
1. Reanalysis inputs
Hourly ERA5 fields provide a consistent global atmospheric record. Near-surface variables are summarized into daily air-mass conditions.
2. Local anomalies
Conditions are standardized relative to local seasonal expectations, so the same raw weather can mean different things in different places and seasons.
3. Air-mass classes
Each location-day is assigned an air-mass category that summarizes the combined thermal, moisture, pressure, cloud, and wind character of the day.
Air-mass categories
The table below shows the GRAMI code, category, abbreviation, and general air-mass description.
| Code | Category | Abbreviation | Description |
|---|---|---|---|
| 0 | Missing | N/A | No valid classification for the location-day. |
| 1 | Humid Cool | HC | Cooler-than-normal and more humid air-mass conditions. |
| 2 | Humid | H | Near-seasonal temperature with higher moisture. |
| 3 | Humid Warm | HW | Warmer-than-normal and more humid air-mass conditions. |
| 4 | Cool | C | Cooler-than-normal conditions without strong humidity or dryness dominance. |
| 5 | Seasonal | S | Conditions close to local seasonal expectations. |
| 6 | Warm | W | Warmer-than-normal conditions without strong humidity or dryness dominance. |
| 7 | Dry Cool | DC | Cooler-than-normal and drier air-mass conditions. |
| 8 | Dry | D | Near-seasonal temperature with lower moisture. |
| 9 | Dry Warm | DW | Warmer-than-normal and drier air-mass conditions. |
| 10 | Cold Front Passage | CFP | Transitional conditions consistent with cold-frontal passage. |
| 11 | Warm Front Passage | WFP | Transitional conditions consistent with warm-frontal passage. |
Potential applications
Climate change indicators
Track changes in the frequency, persistence, and seasonality of air-mass types over time.
Health and biometeorology
Link multivariate air-mass conditions to heat stress, respiratory outcomes, mortality, or other weather-sensitive endpoints.
Ecology and impacts
Study biological, agricultural, hydrological, or hazard responses to changing daily air-mass regimes.
Relationship to GWTC and GWTC-2
GRAMI is the next generation of the Gridded Weather Typing Classification. The original GWTC established an automated, gridded weather-typing framework. GWTC-2 expanded that framework to the global domain using CFS-based data. GRAMI extends the approach again by moving toward an ERA5 reanalysis foundation and a clearer air-mass-index identity.
The rebranding emphasizes the scientific purpose of the dataset: a globally consistent index of relative air-mass character, suitable for climate-scale analysis and interdisciplinary applications.
How to cite
Citation guidance for GRAMI/GWTC-v3 will be updated when the dataset and methods documentation are finalized. Until then, users should cite the source webpage and the foundational GWTC/GWTC-2 publications as appropriate.
- Lee, C. C. 2020. Trends and variability in airmass frequencies: indicators of a changing climate. Journal of Climate, 33(19), 8603–8617.
- Lee, C. C. 2019. The Gridded Weather Typing Classification version 2: a global-scale expansion. International Journal of Climatology, 40(2), 1178–1196.
- Lee, C. C. 2014. The development of a gridded weather typing classification scheme. International Journal of Climatology, 35(5), 641–659.