JRA-55¶
Japanese 55-year Reanalysis
Discontinued
No longer updated
Update of the near real-time JRA-55 data was terminated at the end of January 2024. Data for 1958 to January 2024 remain available, but the record will not be extended. If you need recent years or are starting a new study, consider the successor, JRA-3Q.
Successor: JRA-3Q — Covers September 1947 to present (updated monthly); model resolution TL479 (about 40 km) with 100 levels. Unlike JRA-55, it is licensed under CC BY-NC-SA 4.0 (non-commercial). Reference: Kosaka et al. (2024), J. Meteor. Soc. Japan, 102, 49–109, doi:10.2151/jmsj.2024-004
At a glance¶
| Item | Details |
|---|---|
| Section | Atmosphere › Reanalyses |
| Type | Reanalysis |
| Domains | Atmosphere, Land |
| Provider | Japan Meteorological Agency (JMA) |
| Period | 1958-01 ~ 2024-01 |
| Coverage | Global |
| Horizontal resolution | Model resolution TL319 (about 55 km) / 1.25° × 1.25° latitude–longitude grid / regular Gaussian grid of 640 × 320 (nominally 0.5625°) |
| Vertical resolution | 60 model levels / 37 pressure levels |
| Temporal resolution | 6-hourly analyses, 3-hourly forecasts and diagnostics (monthly means provided separately) |
| Update frequency | Ended (end of January 2024) |
| Licence | CC BY 4.0 (as listed at NCAR GDEX) + JMA terms of use |
| Account required | Yes |
Note on period: Near real-time updates ended, so 31 January 2024 is the last day of data.
How to read the resolution
- Horizontal resolution lists both the native model grid and the grid you download, as given in the official documentation. The same dataset can come on different grids from different providers or products, so check the grid of your downloaded file.
- Grid spacing is not the effective resolution. A provided grid that is finer than the native one carries no extra information.
- Intervals and whether a value is instantaneous, mean or accumulated differ between variables — see 'Available variables' below.
- The 'provided resolution' in lists and the comparison table is a simplified km/degree value for quick comparison. Use this page for exact values.
Overview¶
JRA-55 is the second Japanese global atmospheric reanalysis, conducted by the Japan Meteorological Agency (JMA). It extends back to 1958, coinciding with the establishment of the global radiosonde observing system, and is based on a newer data assimilation and prediction system than JRA-25, with 4D-Var and variational bias correction (VarBC) of satellite radiances. It has been widely used alongside ERA5 as a comparison dataset, especially for East Asia.
What changed from the previous dataset
What changes with the successor¶
JRA-55 → JRA-3Q¶
Transition: 2024-01
JRA-3Q is JMA's third-generation reanalysis, rerun from September 1947 with the operational system upgraded from its 2009 to its 2018 version and with higher resolution, more levels and a higher model top. It greatly reduces JRA-55's energy-budget imbalance, its artificial decline in tropical cyclone detection and its upper-tropospheric warm bias. Boundary conditions, observations, variable names and units, and the licence all changed, so treat it as a separate dataset rather than a drop-in replacement.
| Aspect | JRA-55 | JRA-3Q |
|---|---|---|
| Period | 1958-01 to 2024-01 (near real-time updates ended) | 1947-09 to present (monthly updates at NCAR; separate near real-time stream) |
| Base operational system | JMA global data assimilation system as of December 2009 | JMA global NWP system as of December 2018 (TL479 version) |
| Horizontal resolution | TL319 (about 55 km) | TL479 (about 40 km) |
| Vertical levels and model top | 60 levels, top at 0.1 hPa | 100 levels, top at 0.01 hPa |
| Data assimilation | 4D-Var (inner loop T106), VarBC for satellite radiances, RTTOV 9.3 | Incremental 4D-Var (inner loop TL319L100), VarBC for all satellite radiances, RTTOV 10.2. Static background error covariances (NMC method), with background error variances inflated before July 1998 when observations are sparse (about 155% before 1958) |
| SST and sea-ice boundary conditions | COBE-SST (1°) throughout / sea ice as a single-layer 2 m slab, with a 55% concentration threshold for ice presence | COBE-SST2 (1°, to 1985-05) then MGDSST (0.25°, from 1985-06) / sea ice as a 4-layer 1.5 m slab with separate ice and open-water tiles. A companion run, JRA-3Q-COBE (1985-06 to 1990-12), quantifies the impact of the switch |
| Observations | Observations from 1958 onward; satellite reprocessing available at the time | Adds rescued pre-1958 observations (HadISD, ICOADS R3.0, ISPD v4, IGRA v2, CHUAN and others), reprocessed GMS-5 and MTSAT AMVs, GNSS-RO bending angles, ground-based GNSS zenith delays (1994–2014) and hyperspectral IR sounders. Radiosonde bias correction moves from RAOBCORE to RISE |
| Tropical cyclones | Assimilates tropical cyclone wind retrievals (TCRs) — an artificial downward trend in TC detection and maximum winds from the late 1980s | JMA operational-style TC bogus (sea-level pressure, 850 and 300 hPa winds) — detection rate generally above 90% throughout |
| Global energy budget | Large upward net energy imbalances at both TOA and surface; ocean surface net flux −15.9 W m-2 | TOA net flux −5.5 W m-2 (about half that of JRA-55); ocean surface net flux −6.5 W m-2 |
| Temperature and humidity biases | Upper-tropospheric warm bias near 250 hPa, cold bias near 850 hPa, mid-tropospheric dry bias, excessive tropical precipitation | Upper-tropospheric warm bias much reduced; lower-tropospheric cold bias and mid-tropospheric dry bias alleviated |
| Model physics and external forcing | Longwave radiation with 9 bands every 3 hours, WMO aerosol distribution, MRI-CCM1 ozone, SiB land surface with 1 soil temperature layer and 3 soil moisture layers | Longwave radiation with 11 bands every hour, MASINGAR-based climatology of 5 aerosol species, MRI-CCM2.1 ozone, SiB with 7 soil layers and up to 4 snow layers |
| Output levels and format | 37 pressure levels (1000–1 hPa), 3-hourly surface diagnostics, GRIB1 | 45 pressure levels (1000–0.01 hPa), hourly surface diagnostics and fluxes, GRIB2 (converted to per-variable NetCDF4 at NCAR GDEX) |
| Variable names and units | Precipitation and evaporation in mm day-1 (e.g. tprat); 2 m temperature in file anl_surf125.011_tmp |
Precipitation and evaporation in kg m-2 s-1 (e.g. tprate); 2 m temperature as variable tmp2m-hgt-an-ll125 |
| Licence and distribution | CC BY 4.0 (as listed by NCAR) + JMA terms of use / NCAR d628000 and d628001, DIAS | CC BY-NC-SA 4.0 (non-commercial, share-alike) / NCAR d640000–d640006, DIAS (partly resumed); commercial use via JMBSC |
Things to watch when switching
- The two reanalyses differ in climatology and trends, so do not splice JRA-55 and JRA-3Q into one time series at January 2024. If you need a long record, recomputing everything from JRA-3Q alone is the safer option.
- Precipitation and evaporation units changed from mm/day to kg m-2 s-1. Unchanged scripts will give values 86400 times too small, so update both the variable names and the unit conversion.
- JRA-3Q carries a non-commercial licence (CC BY-NC-SA 4.0), so it cannot be used for commercial projects or services, and derived data cannot be redistributed under other terms. For such uses, check the JMBSC route.
- The SST input switch in June 1985 (COBE-SST2 to MGDSST) and the sea-ice-edge quality problem from June 1985 to April 2013 are sources of discontinuity that JRA-55 did not have. For ocean boundary-layer or polar work, check JRA-3Q-COBE and the quality notices.
- JRA-3Q still underestimates stratospheric warming after volcanic eruptions, and its TC maximum 10 m winds are only 50–60% of best-track values. The improvements over JRA-55 do not extend to these issues.
Sources: Kosaka et al. (2024), The JRA-3Q Reanalysis, J. Meteor. Soc. Japan, 102, 49–109 · Kobayashi et al. (2015), The JRA-55 Reanalysis, J. Meteor. Soc. Japan, 93, 5–48 · Harada et al. (2026), evaluation of JRA-3Q tropical oceanic precipitation, Q. J. R. Meteorol. Soc., 152, e70026 · JMA JRA-3Q official page (products, SST configuration, data providers, quality notices) · JMA TCC training lecture — Reanalysis and JRA-3Q (2024-01, JRA-25/55/3Q comparison tables) · JRA-3Q 1.25° data format document (GRIB2, 45 pressure levels) · JRA-55 Product Users' Handbook (1.25° grid; 37 pressure levels and variable units)
Available variables¶
Main variables from the JRA-55 Product Users' Handbook (1.25° grid) and the NCAR GDEX file names. The short name is the abbreviation after the GRIB1 parameter number in the file name (e.g. anl_p125.011_tmp → tmp). Product categories (file prefixes) are given in parentheses; precipitation and evaporation are mean rates in mm day-1.
Time-stamp convention — instantaneous, mean or accumulated? — click to show
Analyses (anl_*) are instantaneous values at 00, 06, 12 and 18 UTC. Averaged forecast diagnostics (fcst_phy2m125 and others) are averages over forecast hours 0–3 and 3–6 of forecasts started every 6 hours. Dates in file names indicate the beginning of the averaging period (e.g. fcst_phy2m125.1981010100 = average for 00–03 UTC on 1 January). Precipitation, evaporation and fluxes are period-mean rates (mm day-1, W m-2), not accumulations.
- Analysis products (anl_p125, anl_surf125, anl_column, anl_land) are 6-hourly (00, 06, 12, 18 UTC); only the snow depth analysis (anl_snow125) is output once a day at 18 UTC.
- fcst_phy2m125 and fcst_phyland125 are 3-hour averages: 00–03, 06–09, 12–15 and 18–21 UTC are forecast hours 0–3; 03–06, 09–12, 15–18 and 21–24 UTC are forecast hours 3–6.
- fcst_phy3m125 (isobaric average diagnostics) is averaged over forecast hours 0–6.
- fcst_surf125 holds 3-hourly instantaneous values. Cloud cover (tcdc, lcdc, mcdc, hcdc) is marked in the Handbook as switched from average to instantaneous forecast fields in JRA-55.
- The date convention differs from JRA-25, whose file names gave the end of the averaging period. Reading both with the same code shifts averages by 3–6 hours.
- The NCAR GDEX THREDDS aggregations give the averaging interval in time_bounds, so for averaged variables check time_bounds, not just time.
Source: JRA-55 Product Users' Handbook (1.25° latitude-longitude grid), Chapter 4 and Sections 10.2–10.3
instantaneous mean accumulated maximum minimum static — instantaneous: value at the time stamp · mean/accumulated/maximum/minimum: over the indicated interval · static: does not change with time
Isobaric analysis (anl_p125, 6-hourly, 37 levels)¶
| Variable | Short name | Units | Time convention | Notes |
|---|---|---|---|---|
| Geopotential height | hgt |
gpm | instantaneous (6-hourly) | |
| Temperature | tmp |
K | instantaneous (6-hourly) | |
| Dew-point depression (or deficit) | depr |
K | instantaneous (6-hourly) | |
| u-component of wind | ugrd |
m s⁻¹ | instantaneous (6-hourly) | |
| v-component of wind | vgrd |
m s⁻¹ | instantaneous (6-hourly) | |
| Vertical velocity | vvel |
Pa s⁻¹ | instantaneous (6-hourly) | |
| Specific humidity | spfh |
kg kg⁻¹ | instantaneous (6-hourly) | |
| Relative humidity | rh |
% | instantaneous (6-hourly) | |
| Relative vorticity | relv |
s⁻¹ | instantaneous (6-hourly) | |
| Stream function | strm |
m² s⁻¹ | instantaneous (6-hourly) | |
| Velocity potential | vpot |
m² s⁻¹ | instantaneous (6-hourly) |
Surface analysis (anl_surf125, 6-hourly)¶
| Variable | Short name | Units | Time convention | Notes |
|---|---|---|---|---|
| Pressure (surface) | pres |
Pa | instantaneous (6-hourly) | |
| Pressure reduced to MSL | prmsl |
Pa | instantaneous (6-hourly) | |
| 2 m temperature | tmp |
K | instantaneous (6-hourly) | |
| 2 m specific humidity | spfh |
kg kg⁻¹ | instantaneous (6-hourly) | |
| 10 m u- and v-components of wind | ugrd, vgrd |
m s⁻¹ | instantaneous (6-hourly) |
Total column analysis (anl_column, 6-hourly)¶
| Variable | Short name | Units | Time convention | Notes |
|---|---|---|---|---|
| Precipitable water (water vapour only) | pwat |
kg m⁻² | instantaneous (6-hourly) | |
| Zonal water vapour flux | uwv |
kg m⁻¹ s⁻¹ | instantaneous (6-hourly) | |
| Meridional water vapour flux | vwv |
kg m⁻¹ s⁻¹ | instantaneous (6-hourly) |
Fluxes and diagnostics — 3-hour forecast averages (fcst_phy2m125)¶
| Variable | Short name | Units | Time convention | Notes |
|---|---|---|---|---|
| Total precipitation | tprat |
mm day⁻¹ | 3-hour mean | |
| Convective precipitation | cprat |
mm day⁻¹ | 3-hour mean | |
| Large scale precipitation | lprat |
mm day⁻¹ | 3-hour mean | |
| Snowfall rate water equivalent | srweq |
mm day⁻¹ | 3-hour mean | |
| Evaporation | evp |
mm day⁻¹ | 3-hour mean | |
| Latent heat flux | lhtfl |
W m⁻² | 3-hour mean | |
| Sensible heat flux | shtfl |
W m⁻² | 3-hour mean | |
| Momentum flux, u and v components | uflx, vflx |
N m⁻² | 3-hour mean | |
| Downward solar and long wave radiation flux (surface) | dswrf, dlwrf |
W m⁻² | 3-hour mean | |
| Upward solar and long wave radiation flux (surface and nominal top of atmosphere) | uswrf, ulwrf |
W m⁻² | 3-hour mean |
Clouds and ozone (fcst_surf125 3-hourly instantaneous / fcst_p125 6-hourly)¶
| Variable | Short name | Units | Time convention | Notes |
|---|---|---|---|---|
| Total, low, medium and high cloud cover | tcdc, lcdc, mcdc, hcdc |
% | instantaneous (3-hourly) | fcst_surf125. Marked in Handbook Section 10.3 as switched from average to instantaneous |
| Ozone mixing ratio (pressure levels) | ozone |
mg kg⁻¹ | instantaneous (6-hourly) | fcst_p125 (output at 00, 06, 12, 18 UTC). The Handbook does not explicitly call it instantaneous; based on the absence of an averaging interval in the GDEX metadata |
| Cloud water (liquid + ice, pressure levels) | cwat |
kg kg⁻¹ | instantaneous (6-hourly) | fcst_p125. Not explicitly called instantaneous in the Handbook; the GDEX metadata show no averaging interval |
Land surface (anl_land / fcst_phyland125)¶
| Variable | Short name | Units | Time convention | Notes |
|---|---|---|---|---|
| Water equivalent of accumulated snow depth | snwe |
kg m⁻² | instantaneous (6-hourly) | |
| Soil wetness | soilw |
proportion | instantaneous (6-hourly) | |
| Water run-off | rof |
mm day⁻¹ | 3-hour mean | fcst_phyland125 — same 0–3 and 3–6 hour forecast averages as fcst_phy2m125 |
The full variable list is in the official documentation.
How to get the data¶
- NSF NCAR GDEX — 3-hourly and 6-hourly data (d628000) — Web download / subsetting and format conversion / THREDDS
- Requires NCAR registration. WMO GRIB1, about 91 TB in total, so download only the variables and levels you need.
- NSF NCAR GDEX — monthly means and variances (d628001) — Web download
- DIAS (Data Integration and Analysis System, Japan) — Web download
- After long-term maintenance, DIAS announced in July–September 2026 that some services have resumed. The JMA page still says DIAS is inaccessible, so first check whether JRA-55 is available on DIAS, and use NCAR if not.
Practical tips¶
- Under the JMA terms of use, send a copy of any paper or report that uses JRA-55 to JMA (jra@met.kishou.go.jp).
- On the GDEX dataset page, the Publications entry for the JRA-55 paper carries the DOI of the JRA-25 paper (10.2151/jmsj.85.369). The correct DOI for the JRA-55 paper is 10.2151/jmsj.2015-001.
- JRA-55 and JRA-3Q can differ in climatology and trends, so do not splice them together within one analysis.
Notices and change history¶
2026-10-15 · Info · JRA-55 republished on DIAS (15 October 2026)
The DIAS Dataset Republication List shows The Japanese 55-year Reanalysis (JRA-55, dataset ID JRA55) as republished on 15 October 2026, together with DSJRA55 and other datasets. Check the list and the DIAS dataset page before downloading from DIAS.
2025-12-02 · Info · JMA reanalysis website moved to a new URL
The JRA project website moved from jra.kishou.go.jp to www.data.jma.go.jp/jra/html/. Automatic redirection is in place for the time being; please update your bookmarks.
2024-01-31 · Critical · Update of the near real-time JRA-55 data terminated — end of the JRA-55 project
Update of the near real-time JRA-55 data was terminated at the end of January 2024. Data already produced remain available from collaborative organizations such as NCAR, and JRA-3Q is the recommended successor.
2023-12-05 · Warning · JRA-55 will be discontinued at the end of Jan 2024 — JRA-3Q near real-time data at NCAR
NCAR announced that updates of the near real-time JRA-55 data will discontinue at the end of January 2024 and advised users to switch to the JRA-3Q near real-time dataset.
References for your paper¶
Double-check before using in a paper
The citations below were compiled by the maintainer and may contain errors. Always double-check DOIs, authors, years and access-date formats against the provider's official dataset page and your journal's guidelines.
Dataset citation¶
NCAR GDEX 3-hourly and 6-hourly data — DOI 10.5065/D6HH6H41
Japan Meteorological Agency (2013): JRA-55: Japanese 55-year Reanalysis, Daily 3-Hourly and 6-Hourly Data. Research Data Archive at the National Center for Atmospheric Research, Computational and Information Systems Laboratory. https://doi.org/10.5065/D6HH6H41 (Accessed DD MMM YYYY)
Show BibTeX
@misc{jra55_d628000,
author = {{Japan Meteorological Agency}},
title = {{JRA-55}: Japanese 55-year Reanalysis, Daily 3-Hourly and 6-Hourly Data},
year = {2013},
publisher = {Research Data Archive at the National Center for Atmospheric Research, Computational and Information Systems Laboratory},
doi = {10.5065/D6HH6H41},
note = {Accessed DD MMM YYYY}
}
Reference papers¶
JRA-55 reference paper — DOI 10.2151/jmsj.2015-001
Kobayashi, S., Ota, Y., Harada, Y., Ebita, A., Moriya, M., Onoda, H., Onogi, K., Kamahori, H., Kobayashi, C., Endo, H., Miyaoka, K., Takahashi, K. (2015): The JRA-55 Reanalysis: General Specifications and Basic Characteristics. J. Meteor. Soc. Japan, 93(1), 5–48.
Show BibTeX
@article{kobayashi2015jra55,
author = {Kobayashi, Shinya and Ota, Yukinari and Harada, Yayoi and Ebita, Ayataka and Moriya, Masami and Onoda, Hirokatsu and Onogi, Kazutoshi and Kamahori, Hirotaka and Kobayashi, Chiaki and Endo, Hirokazu and Miyaoka, Kengo and Takahashi, Kiyotoshi},
title = {The {JRA-55} Reanalysis: General Specifications and Basic Characteristics},
journal = {Journal of the Meteorological Society of Japan. Ser. II},
volume = {93},
number = {1},
pages = {5--48},
year = {2015},
doi = {10.2151/jmsj.2015-001}
}
Circulation and climate variability paper (optional) — DOI 10.2151/jmsj.2016-015
Harada, Y., et al. (2016): The JRA-55 Reanalysis: Representation of atmospheric circulation and climate variability. J. Meteor. Soc. Japan, 94(3), 269–302.
Acknowledgements¶
The JMA terms of use require publications to acknowledge the data source (the JRA-55 project carried out by JMA). The recommended wording begins "The dataset used for this study is from the Japanese 55-year Reanalysis (JRA-55) project carried out by the Japan Meteorological Agency (JMA)"; copy the full text from the Usage Restrictions on the GDEX dataset page.
If this page helped, please acknowledge the Korean Climate Data Navigator
Please add the sentence below to the Acknowledgements of your paper.