Skip to content

NCEP/NCAR R1

NCEP/NCAR Reanalysis 1 (including its real-time extension, the Climate Data Assimilation System, CDAS)

Discontinued

Get notice alertsCompare with other datasetsReport an error

No longer updated

According to NOAA PSL, the NCEP/NCAR Reanalysis 1 dataset has ceased updates; the last date of data is 17 March 2026. NCEP replaced the Climate Data Assimilation System (CDAS), which had extended R1 in real time, with the new Conventional Observation Reanalysis (CORe) in March 2026 (NWS Service Change Notice 26-12). Historical data remain available from PSL and NCAR but will not be extended, so analyses that need recent years must move to another reanalysis.

Successor: CORe (NOAA Conventional Observation Reanalysis) — not a reprocessing of R1 — NCEP replaced CDAS (the real-time extension of R1) with CORe, a UFS-based (circa-2020 code) Ensemble Kalman Filter reanalysis covering 1950 to the present. It assimilates only conventional observations and atmospheric motion vectors, is distributed in GRIB2, and is updated on the NOAA NODD archive with a 1-day latency. Because the model and observing system differ, it is not the same dataset as R1, and PSL has no immediate plans to serve CORe itself. For general atmospheric research, ERA5 or JRA-3Q are the practical alternatives; the related NCEP–DOE R2 has not been updated since February 2026. To keep using NCEP operational analyses, consider the CFSv2 analysis (2011 onward).

At a glance

Item Details
Section Atmosphere › Reanalyses
Type Reanalysis
Domains Atmosphere, Land
Provider NOAA National Centers for Environmental Prediction (NCEP) and NSF National Center for Atmospheric Research (NCAR); distributed by NOAA PSL and NSF NCAR GDEX
Period 1948-01 ~ 2026-03
Coverage Global
Horizontal resolution Model T62 (about 210 km) / pressure levels on a 2.5° × 2.5° grid (144 × 73) / surface and flux fields on the T62 Gaussian grid (192 × 94, about 1.9°)
Vertical resolution Model: 28 sigma levels / 17 pressure levels (1000–10 hPa). Humidity (shum, rhum) only 1000–300 hPa; vertical velocity (omega) only 1000–100 hPa
Temporal resolution 6-hourly (00, 06, 12, 18 UTC); daily and monthly means provided separately by PSL
Update frequency Ended (last data 17 March 2026)
Licence PSL data are freely available without restriction (PSL asks for an acknowledgement) / NCAR GDEX copy under CC BY 4.0
Account required No

Note on period: 1 January 1948 to 17 March 2026, 18 UTC (last time step in the PSL files). Quality is lower in 1948–1957, when upper-air observations were sparse, and the assimilated data change from 1979 when satellite retrievals enter. The NCAR GDEX copy was updated on a different schedule and may end earlier than PSL.

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

The first long-term global atmospheric reanalysis, produced by NCEP and NCAR in the 1990s with a data assimilation system frozen from 1948 onward and continued in real time as CDAS. Its resolution is low, but it offers the longest period from a single system and is easy to obtain as yearly netCDF files from PSL, so for decades it was widely used for climate diagnostics and index calculations. Updates ended in March 2026 when CDAS was discontinued.

What changes with the successor

NCEP/NCAR R1 → NCEP–DOE R2

Transition: 2002

R2 keeps R1's T62, 28-level system but fixes known errors in R1 processing and updates the model physics, rerun from 1979. Its developers explicitly did not call it a "next-generation" reanalysis: Northern Hemisphere mid-latitude heights and winds are nearly the same, while land-surface variables and surface fluxes are much improved. It was originally recommended as a complement to R1.

Aspect NCEP/NCAR R1 NCEP–DOE R2
Period 1948-01 to 2026-03 (updates stopped when CDAS ended) 1979-01 onward (satellite era only; no updates after 2026-02)
Model resolution T62 (about 210 km), 28 sigma levels Unchanged: T62, 28 levels (no resolution increase)
Southern Hemisphere PAOBS error Southern Hemisphere synthetic surface pressure data (PAOBS) mis-processed for 1979–1992 Fixed
Snow-related errors Snow cover analysis error for 1974–1994, "spectral snow" caused by a humidity diffusion error, and an error in the snowmelt term All fixed; the snow cover mask is interpolated from weekly to daily values and snow handling is revised to use model snow depth
Ocean albedo and cloud tables Ocean albedo error over the whole period; relative humidity–cloud cover tables discontinuous at 0° and 180° longitude Fixed
Soil moisture Nudged towards climatology on a 60-day time scale, giving an excessive annual cycle Nudging removed; observed pentad precipitation (satellite + gauge) corrects model precipitation errors, which feeds into soil moisture
Boundary layer and radiation Original R1 boundary-layer diffusion; shortwave radiation computed every 3 hours on a reduced (128 × 64) grid Hong–Pan non-local boundary-layer diffusion, Chou shortwave radiation, radiation computed hourly on the full Gaussian grid (192 × 94), enhanced cloud-top longwave cooling
Fixed boundary fields SST, sea ice, ozone and desert albedo as used at the time of R1 AMIP-II SST and sea ice, new ozone climatology, CO₂ at 350 ppmv, Briegleb desert albedo, smoothed orography
Observational input Conventional observations + NESDIS satellite temperature retrievals Mostly the same raw observations (some additions possible after 1993); SSM/I not used; still dependent on NESDIS retrievals
Improvements Known problems in land-surface variables, surface fluxes and the water budget Improved surface fluxes, water budget and shortwave fluxes; weaknesses remain in tropical warm-pool OLR and upper-tropospheric tropical moisture (NCAR GDEX description)
PSL file layout Pressure-level humidity shum and rhum up to 300 hPa; sea level pressure file slp; surface fluxes in folder surface_gauss Pressure-level humidity rhum only (up to 10 hPa); sea level pressure file mslp; surface fluxes in folder gaussian_grid

Things to watch when switching

  • R2 starts in 1979, so 1948–1978 exists only in R1. Splicing R1 and R2 at 1979 builds the system differences into your series as a discontinuity — use one dataset per analysis.
  • R2 is the same first-generation system as R1 in resolution and observational input, so switching does not change the upper-air circulation much. Agreement between R1 and R2 in a circulation study is not an independent confirmation.
  • For soil moisture, surface fluxes and the water budget, the developers judge R2 to be better than R1. The R1–R2 difference itself can serve as a measure of analysis uncertainty.
  • File names and folders differ on PSL (slp → mslp, surface_gauss → gaussian_grid), so R1 scripts will not find the R2 files without changes.
  • Neither dataset has been updated since spring 2026; studies that need recent years should move to a current reanalysis such as ERA5 or JRA-3Q.

Sources: Kanamitsu et al. (2002), NCEP–DOE AMIP-II Reanalysis (R-2), BAMS, 83, 1631–1643 · CPC — Overview of NCEP/DOE Reanalysis-2 (Kanamitsu et al.; list of R1 error fixes and physics changes) · CPC NCEP/DOE AMIP-II Reanalysis (Reanalysis-2) home page · NSF NCAR GDEX — NCEP/DOE Reanalysis 2 (d091000) · Kalnay et al. (1996), The NCEP/NCAR 40-Year Reanalysis Project, BAMS, 77, 437–471

Compare the two side by side

NCEP/NCAR R1 CDAS → CORe

Transition: 2026-03

NCEP ended CDAS, the real-time extension of R1, and moved operations to CORe, a reanalysis rebuilt from 1950 with a new model and data assimilation system (operational from 2026-03-19). CORe is not a reprocessed R1 but a separate reanalysis: the model (FV3GFS), data assimilation (80-member EnKF) and observing system are all different.

Aspect NCEP/NCAR R1 CDAS CORe
Period 1948-01 to 2026-03-17 1950-01 to present (about 1-day latency)
Model and vertical levels NCEP global spectral model T62, 28 sigma levels FV3GFS (UFS) C128 cubed-sphere grid, 64 hybrid levels
Data assimilation 3D-Var (SSI), single analysis Ensemble Kalman Filter (EnKF) with 80 members; the distributed fields are the ensemble mean (members for the last 7 days on NOMADS)
Assimilated observations Conventional observations + satellite temperature soundings (from 1979), etc. Conventional observations + satellite atmospheric motion vectors (AMVs) only — no satellite radiances or temperature soundings
Output grids Pressure levels 2.5° × 2.5° (144 × 73); surface and fluxes T62 Gaussian (192 × 94) Both flx and pgb on a 512 × 256 Gaussian grid; 22 pressure levels (1000–1 hPa)
Time step and format 6-hourly, GRIB1 (redistributed as NetCDF by PSL) 3-hourly, GRIB2 (not redistributed by PSL — NODD, NOMADS, NCAR GDEX)
Variables Sea level pressure slp (PRMSL) provided No PRMSL — use mean-sea-level PRES; relative vorticity and divergence must be computed from the winds

Things to watch when switching

  • Do not splice R1 (or CDAS) and CORe into one time series. Model, observations and resolution all differ, so values differ even on the same date.
  • CORe fields are ensemble means, which weakens strong winds such as those in tropical cyclones. For extremes and tropical cyclone work, look at the individual members (last 7 days on NOMADS).
  • Scripts written for R1 (PSL NetCDF on the 2.5° grid) will not work as-is. You will need to read GRIB2 and regrid (e.g. with wgrib2).

Sources: NOAA CPC — CORe home page · NOAA CPC — CORe datasets (comparison with CDAS) · NWS SCN 26-12 (CDAS → CORe, 2026-02) · NOAA PSL — About NOAA CORe Reanalysis

Compare the two side by side

Available variables

Main variables in the NOAA PSL netCDF files (from the OPeNDAP attributes). Names are those inside the files; the PSL folder and file-name pattern are given with each group. Most surface and flux variables are 6-hour forecasts (or forecast averages), not analyses.

Time-stamp convention — instantaneous, mean or accumulated? — click to show

In the PSL 6-hourly files, pressure-level and surface analysis variables (air, hgt, slp, pr_wtr, etc.) are instantaneous values at the time stamp (00, 06, 12, 18 UTC). Precipitation, heat fluxes and radiation in the Gaussian-grid flux files (surface_gauss, other_gauss), by contrast, are 0–6 hour forecast averages started from that analysis, and the time stamp marks the start of the averaging interval (e.g. the 00 UTC value = 00–06 UTC mean).

  • The PSL file attributes tell them apart: instantaneous fields have statistic 'Individual Obs'; 6-hour means have statistic 'Mean' and avg_period '06:00:00' on the time variable.
  • The NCEP documentation defines fluxes as '0–6 hour forecast average/accumulation from the analysis'. The R1 PSL files carry no attribute saying whether the time stamp is the start or end of the interval; we treat it as the start, based on coordinate_defines 'start' in the PSL files of R2, which uses the same system.
  • Precipitation prate is a mean rate (Kg/m^2/s), not an accumulation. Multiply by 21600 for 6-hour totals (mm) or by 86400 for daily totals.
  • 2 m temperature and 10 m winds are 6-hour forecasts, not analyses.
  • Daily means (dailyavgs) average the four values at 00, 06, 12 and 18 UTC (00–24 UTC); monthly means (derived) are averages over the month.

Source: NCEP/NCAR Reanalysis documentation — surface flux code definitions (A = 0–6 h forecast average/accumulation, 6 = 6-h forecast) + PSL netCDF file attributes

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

Pressure levels (pressure/, e.g. air.2025.nc, 2.5°, 17 levels)

Variable Short name Units Time convention Notes
Air temperature air degK instantaneous
Geopotential height hgt m instantaneous
U-wind uwnd m/s instantaneous
V-wind vwnd m/s instantaneous
Omega (vertical velocity in pressure coordinates, 1000–100 hPa) omega Pascal/s instantaneous
Relative humidity (1000–300 hPa) rhum % instantaneous
Specific humidity (1000–300 hPa) shum kg/kg instantaneous

Surface and column-integrated (surface/, 2.5°)

Variable Short name Units Time convention Notes
Sea level pressure slp Pascals instantaneous
Precipitable water for entire atmosphere pr_wtr kg/m^2 instantaneous file pr_wtr.eatm

Surface fluxes and near-surface (surface_gauss/, T62 Gaussian)

Variable Short name Units Time convention Notes
Air temperature at 2 m air degK instantaneous (6-hourly) file air.2m.gauss · A 6-hour forecast according to the NCEP documentation (code 6). The PSL file attributes, however, label it statistic 'Mean' with avg_period 6 hours, which contradicts the documentation — check before use
U-wind at 10 m uwnd m/s instantaneous (6-hourly) file uwnd.10m.gauss · 6-hour forecast (PSL attribute statistic 'Individual Obs')
Precipitation rate prate Kg/m^2/s 6-hourly mean
Latent heat net flux lhtfl W/m^2 6-hourly mean
Sensible heat net flux shtfl W/m^2 6-hourly mean
Downward solar radiation flux at the surface dswrf W/m^2 6-hourly mean file dswrf.sfc.gauss

Top of atmosphere (other_gauss/)

Variable Short name Units Time convention Notes
Upward longwave radiation flux at the nominal top of atmosphere (OLR) ulwrf W/m^2 6-hourly mean file ulwrf.ntat.gauss

The full variable list is in the official documentation.

How to get the data

Example code

The PSL OPeNDAP URLs open directly in xarray. The 2026 files end at 18 UTC on 17 March.

import xarray as xr

url = "https://psl.noaa.gov/thredds/dodsC/Datasets/ncep.reanalysis/pressure/air.2025.nc"
ds = xr.open_dataset(url)
t500 = ds["air"].sel(level=500)   # units degK, 6-hourly
print(t500.mean("time"))

Practical tips

  • Kalnay et al. (1996) classify variables by how strongly observations constrain them — A (upper-air temperature and wind, most reliable), B (humidity, surface temperature), C (clouds, precipitation, surface fluxes; no direct observational influence), and so on. Class C variables largely reflect the model, so use them with caution.
  • Pressure-level humidity stops at 300 hPa and omega at 100 hPa. If you need upper-tropospheric water vapour, use another reanalysis.
  • PSL's list of known problems covers, among others, the Southern Hemisphere PAOBS error (1979–1992), TOVS retrieval problems (1997–2001) and cold Southern Hemisphere land temperatures after a 1998 change in the snow analysis. Check these periods before interpreting long-term trends.
  • R1-based indices and climatologies (e.g. PSL plotting tools, circulation indices) may also have stopped in March 2026. For real-time monitoring, check whether they now use CORe or another reanalysis.

Notices and change history

2026-03-17 · Critical · NCEP/NCAR Reanalysis 1 has ceased updates — last date of data 17 March 2026

PSL announced that the NCEP/NCAR Reanalysis 1 dataset has ceased updates due to the National Weather Service change from CDAS to CORe; historical data remain available on the PSL site. The PSL data notice gives 18 March as NCEP's switchover date, while the CPC CORe page gives 19 March as the start of CORe operations.

Read the original

2026-02-28 · Warning · PSL warns that R1 updates will stop

A PSL data notice announced that the NCEP/NCAR Reanalysis 1 dataset will no longer be updated effective on or around 18 March.

Read the original

2026-02-09 · Warning · NWS Service Change Notice 26-12 — Upgrading the Climate Data Assimilation System (CDAS) to the Conventional Observation Reanalysis (CORe)

NWS announced that CORe would replace CDAS on or about 9 March 2026 and that the cdas and cdas2 directories on NOMADS and FTP would be removed (updated notice reissued on 13 February). CORe files are provided in GRIB2 under /com/core/.

Read the original

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 copy (d090000)

NCEP/NCAR Global Reanalysis Products, 1948-continuing. NSF NCAR Geoscience Data Exchange (GDEX), dataset d090000. https://gdex.ucar.edu/datasets/d090000/ (no dataset DOI is shown on the page; include the date accessed)

Reference papers

R1 reference paper — DOI 10.1175/1520-0477(1996)077<0437:TNYRP>2.0.CO;2

Kalnay, E., Kanamitsu, M., Kistler, R., Collins, W., Deaven, D., Gandin, L., Iredell, M., Saha, S., White, G., Woollen, J., Zhu, Y., Leetmaa, A., Reynolds, R., Chelliah, M., Ebisuzaki, W., Higgins, W., Janowiak, J., Mo, K. C., Ropelewski, C., Wang, J., Jenne, R., Joseph, D. (1996): The NCEP/NCAR 40-Year Reanalysis Project. Bulletin of the American Meteorological Society, 77(3), 437–471.

Show BibTeX 
@article{kalnay1996ncep,
  author  = {Kalnay, E. and Kanamitsu, M. and Kistler, R. and Collins, W. and Deaven, D. and Gandin, L. and Iredell, M. and Saha, S. and White, G. and Woollen, J. and Zhu, Y. and Leetmaa, A. and Reynolds, R. and Chelliah, M. and Ebisuzaki, W. and Higgins, W. and Janowiak, J. and Mo, K. C. and Ropelewski, C. and Wang, J. and Jenne, Roy and Joseph, Dennis},
  title   = {The {NCEP/NCAR} 40-Year Reanalysis Project},
  journal = {Bulletin of the American Meteorological Society},
  volume  = {77},
  number  = {3},
  pages   = {437--471},
  year    = {1996},
  doi     = {10.1175/1520-0477(1996)077<0437:TNYRP>2.0.CO;2}
}

50-year reanalysis monthly means and documentation (optional)

Kistler, R., et al. (2001): The NCEP–NCAR 50-Year Reanalysis: Monthly Means CD-ROM and Documentation. Bulletin of the American Meteorological Society, 82, 247–267.

Acknowledgements

If you obtained the data from PSL, include PSL's requested acknowledgement ("data provided by the NOAA PSL, Boulder, Colorado, USA, from their website at https://psl.noaa.gov") and cite Kalnay et al. (1996). If you used the NCAR copy, also give the GDEX dataset name and the date accessed.

If this page helped, please acknowledge the Korean Climate Data Navigator

Please add the sentence below to the Acknowledgements of your paper.

Acknowledgements

Information on the datasets used in this study was compiled with the help of the Korean Climate Data Navigator (KCDN) (Sim, 2026; https://doi.org/10.5281/zenodo.23300923).