MSU/AMSU (RSS·UAH·STAR)¶

MSU/AMSU(/ATMS) Mean Layer Atmospheric Temperatures — RSS V4.0, UAH v6.1, NOAA STAR v5.0

Available

At a glance¶

Note on period: Start dates differ by product and layer. NOAA STAR v5.0 starts in November 1978 (TMT), December 1978 (TLS) and January 1981 (TUT, TLT); RSS starts in late 1978 for TLT, TMT and TLS and 1987 for TTS; the UAH v6.1 lower-troposphere chart starts in 1979. As checked on 2026-10-13, all three run through September 2026 (STAR via the NCEI copy; files on the STAR server stop at July 2025).

Overview¶

Monthly deep-layer mean temperatures of the atmosphere (lower and middle troposphere, lower stratosphere and others) since late 1978, derived from oxygen thermal emission measured by microwave sounders (MSU → AMSU-A → ATMS) on polar-orbiting NOAA, Metop, Aqua and other satellites. RSS, UAH and NOAA STAR merge the same observations with different intersatellite calibration and orbital drift (diurnal) adjustments, so the size of tropospheric warming since 1979 differs among them. As a four-decade upper-air record with even coverage where radiosondes are sparse, it is a standard reference for climate change detection and for evaluating reanalyses and climate models (including the tropospheric amplification debate).

What changes with the successor¶

UAH v6.0 → UAH v6.1¶

Transition: 2024-11

NOAA-19, whose orbit had drifted too far through the diurnal cycle, is excluded from 2021 onward, leaving Metop-B as the only satellite since then. Recent warm values came down slightly and the 1979–2024 trend fell by 0.01 °C per decade.

Sources: Roy Spencer blog — UAH v6.1 Global Temperature Update for October 2024 (2024-11-02)

NOAA STAR v4.1 → NOAA STAR v5.0¶

Transition: 2023-01

STAR v5.0 merges three generations of sounders (MSU, AMSU-A, ATMS on 16 polar orbiters) by "backward merging" from stable-orbit reference satellites (Aqua, Metop-A, Suomi NPP, NOAA-20), with a new diurnal drift adjustment and recalibration of NOAA-11, 12, 14 and 15; it replaced V3.0, V4.0 and V4.1 from January 2023.

Sources: Readme — STAR V5.0 Mean Layer Temperature CDR · Zou et al. (2023), JGR Atmospheres

Available variables¶

The three products differ in file format and variable names. The variable-name column gives the NOAA STAR v5.0 netCDF names (distributed by NCEI, Mean-Layer-Temperature-NOAA_v05r00_*.nc), checked against the NCEI OPeNDAP metadata (.das, .dds) on 2026-10-14. For RSS V4.0, brightness_temperature_anomaly was confirmed in the NCEI copy's anomaly files (uat4_tb_v04r00_anom_ch.nc); names inside the RSS server files (RSS_Tb_Maps_ch_*_V4_0.nc) are not documented officially and could not be confirmed. UAH provides text tables only, no netCDF. All units are K (anomalies in K).

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

Tropospheric layer temperatures¶

Tropopause and stratospheric layer temperatures¶

Anomalies and auxiliary files¶

The full variable list is in the official documentation.

How to get the data¶

• RSS — Upper Air Temperature (description and downloads) — Web / HTTPS (netCDF maps, text time series)
• RSS netCDF map files (V4.0) — HTTPS direct download
  • One monthly mean file (RSS_Tb_Maps_ch_V4_0.nc) and one anomaly file (RSS_Tb_Anom_Maps_ch_V4_0.nc) per layer.
• UAH v6.1 lower-troposphere time series (text) — HTTP direct download
  • Middle troposphere is tmt/uahncdc_mt_6.1.txt, tropopause ttp/uahncdc_tp_6.1.txt and lower stratosphere tls/uahncdc_ls_6.1.txt.
• NOAA NCEI — Mean Layer Temperature-NOAA CDR (STAR v5.0) — HTTPS, THREDDS, AWS S3, Google Cloud (netCDF4)
  • As checked in October 2026, NCEI THREDDS has data through September 2026. The v5.0 files on STAR's own server stop at July 2025, so get current data from NCEI.
• NOAA NCEI — Mean Layer Temperature-RSS CDR (copy of V4.0) — HTTPS, THREDDS, cloud (netCDF4)
  • NCEI metadata show it ending in February 2025 with "no updates planned". Get current RSS data from the RSS server.

Example code¶

RSS netCDF files download over HTTPS without login. The official page does not give variable names, so open the file to check. For NOAA STAR v5.0, look up the latest file name (…_E_C.nc) in the NCEI THREDDS catalogue and fetch it the same way.

# pip install xarray netCDF4 requests
import requests
import xarray as xr

url = "https://data.remss.com/msu/data/netcdf/RSS_Tb_Anom_Maps_ch_TLT_V4_0.nc"   # RSS V4.0 TLT monthly anomaly maps
with open("rss_tlt_anom.nc", "wb") as f:
    f.write(requests.get(url, timeout=120).content)

ds = xr.open_dataset("rss_tlt_anom.nc")
print(ds)                       # check variable names, dimensions (time, lat, lon) and baseline attributes

Practical tips¶

• TLT and TMT trends since 1979 differ among the three products even though they share the same observations. Check that conclusions do not hinge on one product, and report the range across all three.
• The RSS page mixes statements that TTS and TLS 'remain Version 3.3' with V4.0 labels, while the distributed file names are V4_0 for every layer. In papers, give the version from the file name and the layer-specific references (TLT Mears & Wentz 2017, TMT and TTT 2016, TTS and TLS 2009).
• RSS marks its text files (UAH-compatible format) as deprecated and asks that new code use netCDF.
• RSS states that ATMS is not yet used (AMSU is to be phased out in favour of ATMS). As ageing AMSU satellites drift or fail, the continuity of the RSS record may be affected.
• NOAA STAR v5.0 uses ATMS (Suomi NPP, NOAA-20). NOAA and NASA have announced that Suomi NPP science data delivery ends in early November 2026, so watch for subsequent version or reprocessing notices.
• Stratospheric layers such as TLS and TTS spike after major volcanic eruptions (El Chichón, Pinatubo); state how those periods are handled when fitting linear trends.
• Use cosine-latitude weights for area means of gridded data, and remember that RSS has no values poleward of 82.5° (south of 70°S for TLT).

Notices and change history¶

References for your paper¶

Dataset citation¶

NOAA STAR v5.0 (NOAA CDR of Mean Layer Temperature-NOAA, Version 5) — DOI 10.25921/tn91-wv50

@misc{noaa_star_mlt_v5,
  author    = {Zou, Cheng-Zhi and Hao, Xianjun},
  title     = {{NOAA} Climate Data Record ({CDR}) of Mean Layer Temperature-{NOAA}, Version 5},
  year      = {2023},
  publisher = {NOAA National Centers for Environmental Information},
  doi       = {10.25921/tn91-wv50}
}

RSS V4.0 (NOAA CDR of Atmospheric Layer Temperatures, Version 4 — NCEI copy) — DOI 10.25921/EGRF-Q723

@misc{rss_mlt_v4_cdr,
  author    = {Mears, Carl A. and Wentz, Frank J. and {NOAA CDR Program}},
  title     = {{NOAA} Climate Data Record ({CDR}) of Atmospheric Layer Temperatures, Version 4},
  year      = {2015},
  publisher = {NOAA National Centers for Environmental Information},
  doi       = {10.25921/EGRF-Q723}
}

Reference papers¶

RSS TLT V4.0 — DOI 10.1175/JCLI-D-16-0768.1

@article{mears2017tlt,
  author  = {Mears, Carl A. and Wentz, Frank J.},
  title   = {A Satellite-Derived Lower-Tropospheric Atmospheric Temperature Dataset Using an Optimized Adjustment for Diurnal Effects},
  journal = {Journal of Climate},
  volume  = {30},
  number  = {19},
  pages   = {7695--7718},
  year    = {2017},
  doi     = {10.1175/JCLI-D-16-0768.1}
}

RSS TMT and TTT V4.0 — DOI 10.1175/JCLI-D-15-0744.1

@article{mears2016tmt,
  author  = {Mears, Carl A. and Wentz, Frank J.},
  title   = {Sensitivity of Satellite-Derived Tropospheric Temperature Trends to the Diurnal Cycle Adjustment},
  journal = {Journal of Climate},
  volume  = {29},
  number  = {10},
  pages   = {3629--3646},
  year    = {2016},
  doi     = {10.1175/JCLI-D-15-0744.1}
}

UAH Version 6 methodology — DOI 10.1007/s13143-017-0010-y

@article{spencer2017uah6,
  author  = {Spencer, Roy W. and Christy, John R. and Braswell, William D.},
  title   = {{UAH} Version 6 global satellite temperature products: Methodology and results},
  journal = {Asia-Pacific Journal of Atmospheric Sciences},
  volume  = {53},
  number  = {1},
  pages   = {121--130},
  year    = {2017},
  doi     = {10.1007/s13143-017-0010-y}
}

NOAA STAR v5.0 (backward merging) — DOI 10.1029/2022JD037472

@article{zou2023star,
  author  = {Zou, Cheng-Zhi and Xu, Hui and Hao, Xianjun and Liu, Qian},
  title   = {Mid-Tropospheric Layer Temperature Record Derived From Satellite Microwave Sounder Observations With Backward Merging Approach},
  journal = {Journal of Geophysical Research: Atmospheres},
  volume  = {128},
  number  = {6},
  pages   = {e2022JD037472},
  year    = {2023},
  doi     = {10.1029/2022JD037472}
}

Acknowledgements¶

RSS asks users to cite the paper for each layer (TLT Mears & Wentz 2017, TMT and TTT 2016, TTS and TLS Mears & Wentz 2009 in JTECH). UAH gives no dataset DOI or citation text on its official pages, so cite Spencer et al. (2017) together with the version (v6.1), file name and download date. For NOAA STAR v5.0, cite both the NCEI CDR DOI and Zou et al. (2023).

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).

Links¶

• RSS Upper Air Temperature (MSU/AMSU)
• RSS monthly time series file listing
• UAH latest global temperatures (Roy Spencer)
• NOAA STAR Microwave Sounding Temperature Assessment System (MSCAT)
• NOAA STAR data products (including the v5.0 Readme)
• NCEI THREDDS — Mean Layer Temperature-NOAA