The 5-degree latitude/longitude grids contained in this dataset make up the longest continuous set of daily gridded Northern Hemisphere sea-level pressure data in the DSS archive. These grids have been assembled by DSS from the grids of various meteorological chart digitization projects and operational analyses. Grids begin in 1899 and cover the Northern Hemisphere from 15N to the North Pole. The dataset continues to be updated regularly as new data become available.

In the earlier years, there is one grid per day, but from July 1962 on, there are two grids per day. Prior to 1946, the original grids had a 10-degree resolution and the intermediate points were interpolated and indicated as such. Also in this period, there are times when large numbers of gridpoints are missing, particularly over the Asian continent. For more information, see the documentation about original data points.

This dataset contains the longest continuous time series of monthly gridded Northern Hemisphere sea-level pressure data in the DSS archive. The 5-degree latitude/longitude grids, computed from the daily grids in ds010.0, begin in 1899 and cover the Northern Hemisphere from 15N to the North Pole. The dataset continues to be updated regularly as new data become available.

Each monthly grid is a simple average of all available daily grids for the month. Prior to 1955, there is one grid per day. From July 1962 on, there are two grids each day. In the interim period, the number of daily grids varies between one and two.

The grids for the period 1899-1977 were inspected and many corrections were made by Kevin Trenberth of the Laboratory of Atmospheric Research at the University of Illinois at Champaign-Urbana, and these grids are included in this dataset. For more information about these corrections, see the July 1980 issue of Monthly Weather Review.

This dataset contains once-daily Northern Hemisphere sea-level pressure data on a 10-degree by 5-degree (36x16) latitude/longitude grid for the period 1880 to 1979.

The gridded daily sea-level pressure analyses in this dataset were produced by the operational models of the U.S. Navy Fleet Numerical Oceanography Center (FNOC). The data are arranged in a 63x63 polar-stereographic grid covering the Northern Hemisphere for the period January 1946 to December 1993. DSS has produced a set of monthly grids from the dailies.

This dataset also includes daily 500mb geopotential height grids for the period January 1946 to December 1955 on the same polar-stereographic grid as the sea-level pressure data.

The daily sea-level pressure grids in this dataset were received from Steve Pazan of the U.S. Navy Fleet Numerical Oceanography Center (FNOC). The 63x63 polar-stereographic grids cover the period January 1946 to December 1978.

This dataset contains gridded Northern Hemisphere sea-level pressure and 500mb geopotential height data from the United Kingdom.

The DSS has converted the original daily 10-degree by 5-degree sea-level pressure grids to a 5-degree by 5-degree grid, interpolating data to non-original gridpoints and computing monthly grids. The period covered is December 1944 to December 1946.

The daily 500mb geopotential height grids cover the period from January 1945 to December 1946 on a 10-degree by 5-degree grid.

This dataset contains daily grids of dewpoint depression and cloud coverage for the period September 1963 to April 1972, provided by the U.S. Air Force Global Weather Central (AFGWC). Grids are available twice-daily at 12-hour intervals (00Z and 12Z). Dewpoint depression is available at four levels (850mb, 700mb, 500mb, and 400mb), as are low, middle, high, and total cloud coverage.

Also known as 3-D Nephanalyses

This dataset contains daily gridded Northern Hemisphere analyses from NMC's operational model on a 47x51 polar-stereographic grid centered on the North Pole. Available parameters include geopotential height, temperature, stream functions, thickness, and vertical velocity for various levels and layers through the troposphere. In general, the grids cover the period from January 1959 to December 1972, but periods for individual parameters will vary.

This dataset of daily gridded analyzed geopotential height and temperature for the Northern Hemisphere stratosphere covers the period from January 1946 to December 1980. The 47x51 polar-stereographic grids, centered on the North Pole, were produced by NMC's operational model.

For data beyond 1980, see ds067.0, which provides global coverage of the stratosphere on a pair of 63x63 polar-stereographic grids centered on each Pole.

This dataset contains Fourier wave analyses built from other DSS datasets.

This dataset contains various daily NMC forecast grids on a 47x51 Northern Hemisphere polar-stereographic grid centered on the North Pole. Grids on various tropospheric levels and layers contain parameters which include geopotential height, temperature, wind, surface and sea-level pressure, vertical motion, and stream functions, and they are available for sporadic periods between November 1967 and December 1971.

NMC operationally produced these daily gridded Northern Hemisphere analyses of various tropospheric levels for the period August 1963 to December 1972. Parameters include upper-level winds, surface temperature, sea-level pressure, tropopause pressure and temperature, and 500mb relative humidity on a 47x51 polar-stereographic grid centered on the North Pole.

This dataset contains various gridded Northern Hemisphere tropospheric analyses and 6-hour forecasts produced by NMC's operational model for the period October 1958 to December 1972. Vertical motion analyses at 650mb and 350mb are available for most of the period, and geopotential heights and thicknesses for for various levels and layers are available until January 1965. Six-hour forecasts of vertical motion cover the period from July 1966 to December 1972. The grid is a 47x51 polar-stereographic grid centered on the North Pole.

Originally the 47x51 grid was the one in which the B-3 analyses were done. These are now interpolated from the global analyses. Now NCEP interpolates the 65x65 grid from the global analyses. NCAR extracts the 47x51 subgrid (List A) from the full grid (List B).

More information is available.

DSS presents the Global Stratospheric Analyses on a pair of 65x65 polar grids. These were operationally prepared by NCEP daily, at 12Z.

The analyses are available on 8 mandatory levels from Parameters include pressure, geopotential height, temperature, and total column ozone.

This dataset is in the binary NMC Office Note 84 (ON84) format.

This dataset contains global once-daily and monthly-mean gridded analyses for September 1978 to February 1998, which were produced by NMC's operational models. At mandatory levels from 1000mb up to 0.4mb, grids contain geopotential height and temperature. At the tropopause, grids contain pressure and temperature. Also included are total column ozone and upward flux of longwave radiation.

The grid is a 65x65 polar-stereographic grid, centered on the Pole (North or South). Global coverage is provided by combining two grids, one for the Northern Hemisphere and one for the Southern Hemisphere.

The Flattery analysis method was used through 26 July 1984, so the usefulness of grids in that period is suspect. After that time, the analysis method used was the same as for other NMC global archives.

DSS presents the Global Stratospheric Analyses on a pair of 65x65 polar grids, daily at 12Z. These were operationally prepared by NCEP.

The analyses are available on 8 mandatory levels from 70 to 0.4 mb. Parameters include pressure, geopotential height, and temperature.

This dataset is in GRIB format. It is updated monthly using weekly files prepared for DSS by NCEP. The most current 12 months of data are available online through the DS067.3 data page.

This dataset contains once-daily global gridded geopotential height and temperature at levels from 1000mb up to 1mb for the period from April 1979 to May 1994. Bill Randel of NCAR interpolated bad and missing data in the grids of ds067.1 and stored the new data as zonal Fourier coefficients on an approximate 4.5 degree latitude/longitude grid (consistent with R15 spectral resolution). Further information can be found in NCAR Technical Note 366, which is available from the UCAR publications office at (303) 497-8601.

The products of NMC Eta model are archived in this dataset at spatial resolution of 40km with 38 vertical levels. This is the primary NMC mesoscale data archive over North America including Alaska.

This dataset contains the Limited Fine Mesh (LFM) model output from NCEP. The output contains initial time analyses and forecast fields from 6 hours out to 48 hours. These grids are no longer being produced by NCEP. The December, 1995 data is the last data we will archive from the LFM model.

Limited area NGM grids from NMC. Analyses and 6 hrly forecasts (to 48 hr) made each 12 hours. For North America. Same grid as for LFM.

This dataset contains the FSL MAPS analyses on their native (hybrid-b) coordinate system.

Analyses since 1983Jan were merely interpolated by NMC from the Global Analyses in DS082.0

Flattery's analysis scheme, using second version of Hough coefficients.

See DS081.0 for later analyses.

More information is available.

Flattery's global analysis scheme, using second version of Hough coefficients.

Later analyses, in NMC ON84 format, are in DS082.0. See DS080.0 for earlier analyses.

More information is available.

DSS presents the Global Final (FNL) Analyses on a pair of 2.5x2.5 degree hemispheric grids, every twelve hours. These were operationally prepared by NCEP. The NCAR documents Information about NMC Analysis and Forecast Models and History of Changes in the NMC Analysis Techniques discuss model features and changes.

The analyses are available on the surface, 12 mandatory levels from 1000mb to 50mb, boundary and some sigma layers, tropopause, and a few others. Parameters include surface pressure, sea level pressure, geopotential height, temperature, sea surface temperature, potential temperature, relative humidity, snow depth (weekly, Northern Hemisphere only) precipitable water, u- and v- winds and vertical motion.

This dataset is in the binary NMC Office Note 84 (ON84) format. DSS has prepared a set of GRIB formatted files which have a one-to-one correspondence to the ON84 formatted files. More information is available.

DSS presents a surface subset of the Global Final (FNL) Analyses on a pair of 2.5x2.5 degree hemispheric grids, every twelve hours. These were operationally prepared by NCEP. The NCAR documents Information about NMC Analysis and Forecast Models and History of Changes in the NMC Analysis Techniques discuss model features and changes.

The analyses are available on the surface, at 1000mb and a boundary layer. This subset was prepared by DSS from the analyses in DS082.0. Parameters include surface pressure, sea level pressure, geopotential height, temperature, sea surface temperature, potential temperature, relative humidity, snow depth (weekly, Northern Hemisphere only), and u- and v-winds.

This dataset is in the binary NMC Office Note 84 (ON84) format. More information is available.

NCEP Global Tropospheric Analyses on 145x37 2.5 degree grids, one for each hemisphere, 1976Jul01-1997Apr12. The Flattery analysis method was used through 1978Sep21. It forced a relationship between heights and winds which causes the zonal mean meridional wind at every level to be zero. This also caused other problems with winds in equatorial areas. The optimal analysis (OI) method (12 layer) started 1978Sep22. In 1985Apr the Medium Range Forecast (MRF) model started, with 18 layers. Analyses include geopotential height, temperature, wind (u- and v-components) and potential temperature. In more recent years the temperature analyses are derived from thickness temperatures.

NCEP Global Analyses on 145x37 2.5 degree grids, one for each hemisphere. Has surface, tropospheric, tropopause, and lower stratospheric analyses; at mandatory levels up to 50mb. More... This is a subset of the data in DS082.0.and is identical except that it is in GRIB format on CD-ROM in. Every CDROM contains six months of data.

DSS presents the Global Final (FNL) Analyses on a pair of 2.5x2.5 degree hemispheric grids every twelve hours. These were operationally prepared by NCEP. The NCEP documents About the Global Parallel System and EMC Model Documentation describe the model.

The analyses are available on the surface, 16 mandatory levels from 1000mb to 10mb, boundary layer and some sigma layers, tropopause, and a few others. Parameters include surface pressure, sea level pressure, geopotential height, temperature, sea surface temperature, soil temperature, water content of soil, ice cover, water equivalent of snow depth, relative humidity, specific humidity, u- and v-wind components, vertical motion, vorticity, precipitable water, minimum temperature, maximum temperature and land-sea mask

This dataset is in GRIB format. It is updated monthly using weekly files prepared for DSS by NCEP. The most current 12 months of data are available online through the DS083.0 data page. Information about using the data is here.

DSS presents a surface subset of the Global Final (FNL) Analyses on a pair of 2.5x2.5 degree hemispheric grids, every twelve hours. These were operationally prepared by NCEP. The NCEP documents About the Global Parallel System and EMC Model Documentation describe the model.

The analyses are available on the surface, at 1000mb and a boundary layer. This subset was prepared by DSS from the analyses in DS083.0. Parameters include surface pressure, sea level pressure, geopotential height, temperature, sea surface temperature, soil temperature, water content of soil, ice cover, water equivalent of snow depth, relative humidity, specific humidity, u- and v-wind components, minimum temperature, maximum temperature and land-sea mask

This dataset is in GRIB format. It is updated irregularly. Information about using the data is here.

DSS presents the Global Final (FNL) Analyses on 1.0x1.0 degree grids covering the entire globe every six hours. These were operationally prepared by NCEP. The NCEP documents About the Global Parallel System and EMC Model Documentation describe the model. For questions about the science of the FNL, contact the NCEP Science Liaison.

The analyses are available on the surface, 26 mandatory (and other pressure) levels from 1000mb to 10mb, boundary and some sigma layers, tropopause and a few others. Parameters include surface pressure, sea level pressure, geopotential height, temperature, sea surface temperature, soil values, ice cover, relative humidity, u- and v- winds, vertical motion, vorticity and ozone.

This dataset is in GRIB format. It is updated every 6 hours by DSS using files downloaded from the NCEP server. The most current 12 months of data are available online through the DS083.2 data page. Information about using the data is here.

10 day forecasts at 12 hour intervals from MRF model in spherical harmonics at R30 through 1995Nov beginning 1995Dec, on 2.5 deg grid in GRIB

Analyses from the MRF on 18 or 23 sigma levels in spherical harmonic representation

Fields from MRF analyses, four times per day, plus 12-, 24- and 36-hour forecasts.

This dataset contains a time-series collection of monthly gridded Northern Hemisphere tropospheric data on a 47x51 polar stereographic grid. This collection has been assembled by DSS from the daily grids of various operational models and meteorological projects.

This dataset contains a time-series collection of monthly mean gridded Northern Hemisphere tropospheric data on a 5-degree latitude/longitude grid. This collection was assembled and computed by DSS from the daily grids of various operational models and meteorological projects. The daily grids are located in ds195.1.

This dataset contains a time-series collection of monthly Northern Hemisphere tropospheric data on a 2.5-degree latitude/longitude grid. These grids were created by DSS from the grids of ds085.0, which have a resolution of five degrees. The intermediate points of the grids in this set were interpolated from the existing surrounding gridpoints using a 16-point Bessel interpolation scheme.

This dataset contains monthly means of several meteorological variables, their squares, and their cross-products for nine levels in the atmosphere. These mean grids were prepared by the Climate Analysis Center (CAC) from NMC daily analyses. The grid is a 2.5 degree grid (145x37) for each hemisphere, so that data are available globally.

72x19 hemispheric grids, 8 levels, temperature and height

Products from NCEP/NCAR Reanalysis Project (NNRP or R1) are archived in this dataset. The resolution of the global Reanalysis Model is T62 (209 km) with 28 vertical sigma levels. Results are available at 6 hour intervals. Although the initial plan is to reanalyze the data for a 40-year period (1957-1996), production has gone back to 1948 and going forward continuously. Future plans call for rerunning the entire period as next generation models are ready.

There are over 80 different variables, (including geopotential height, temperature, relative humidity, U and V wind components, etc.) in several different coordinate systems, such as 17 pressure level stack on 2.5x2.5 degree grids, 28 sigma level stack on 192x94 Gaussian grids, and 11 isentropic level stack on 2.5x2.5 degree grid. They are organized as different subgroups in the archive. In addition to analyses, diagnostic terms (for example: radiative heating, convective heating) and accumulative variables (like precipitation rate) are present. The input observations are archived with quality/usage flags in WMO BUFR format. Most of the project outputs are stored in WMO GRIB format. Other files, such as restart files and zonal statistics, are saved in IEEE format.

Some special periods are analyzed more than once to provide data for special research studies. For example, a special NOSAT run of 1979 was made excluding most satellite inputs. This run could be used for evaluating the impact of satellite data on the analysis. During TOGA-COARE experiment period, special runs of reanalysis model without (NOEXP) experimental data are archived under TOGA-COARE directory.

For details and problems, see NCEP/NCAR Reanalysis. Monthly means are on line at ds090.2. The R1 forecasts are in ds090.1 dataset.

This data set contains forecast products from the NCEP/NCAR Reanalysis Project. The resolution of the Reanalysis Forecast Model is T62 (209 km) with 28 vertical sigma levels. Every five days, beginning with 00Z 01 January, an 8 day forecast run is made. The initial time and outputs for every 12 hours are saved from this run. Pressure level and special level data are archived on the 2.5x2.5 latitude-longitude grid and flux fields are archived on the 192x94 Gaussian grid. Details on variables included are available elsewhere. Some special periods will be analyzed more than once to provide data for special research studies. For example, a special NOSAT run of 1979 was made excluding most satellite inputs. This run could be used for evaluating the impact of satellite data on the forecasts. See DS090.0for analysis and first guess fields. All data files of 1948-2005OCT are available as of 2005DEC29.

The monthly means of NCEP/NCAR Reanalysis (R1) products, archived in ds090.0 dataset, are extracted and reorganized into 10 subgroups in this dataset. The 10 subgroups are: dgn, presA, presB, theta, xprod, 00Z, 06Z, 12Z, 18Z and rest subgroups. The monthly means of diagnostic terms, (such as various heating terms, convective moistening rate, etc.) are in the dgn subgroup. The presA subgroup contains most commonly used parameters, like geopotential height, temperature, U and V winds, relative humidity, at all 17 pressure levels. The presB subgroup has the remaining variables on pressure levels. The theta subgroup includes all variables on isentropic levels. The xprod contains cross products of variables at pressure levels. The 00Z, 06Z, 12Z and 18Z subgroups have monthly means computed from those individual hours only. Monthly means of flux terms (like wind at 10m, temperature at 2m, etc) and accumulative variables (like precipitation rate, precipitable water, etc.) are in the rest subgroup.

The data files are in WMO GRIB format. Both the monthly means and their variances are in the same file but in different GRIB records. Examples of separating monthly means from variances are shown in how2use_grads.txt.

All subgroups will be available on line under data. The ones that are not on line yet will be moved over upon request.

This dataset contains products from the NCEP/DOE Reanalysis II (R2) Project. The resolution of the R2 model is the same as NCEP/NCAR Reanalysis (R1) proejct, T62 (209 km) with 28 vertical sigma levels. The output fields are saved 4 times a day as in R1. The main objective of the R2 project is to correct known errors in the R1 project and update the parameterizations of the physical processes. Improvements are seen in surface flux fields, hydrological budget, short wave radiation flux, etc. There are also drawbacks found in the outgoing longwave radiation over tropical warm pool and upper-level tropical moisture. R2 plans to reanalyze the data starting from 1979 and going forward only.

The main R2 products archived at NCAR consitute of 6 subgroups: pressure level analysis (pgb-anl), sigma level analysis (sgb-anl), flux term analysis (flx-ft00), diagnostic analysis (dg3-ft00) and 6-hour forecasts of flux term (flx-ft06) and diagnostic (dg3-ft06) terms. All data files are in GRIB format and archived by month.

This dataset is updated once a month. Most commonly used fields are also available at NCEP's NCEP NOMAD Server. Climate Diagnostic Center (CDC) of NOAA has the same data in netCDF format at CDC Reanalysis 2.

This dataset contains various grids created by DSS to make comparisons of the outputs of Reanalysis vs. Reanalysis and Reanalysis vs. other operational models.

This dataset contains gridded sea-level pressure data for the Southern Hemisphere for January 1951 to December 1958. The grid is a 5-degree latitude longitude grid that covers the hemisphere from 15S to the South Pole.

The data came to DSS originally from South Africa, where it was on a diamond latitude/longitude grid (10-degree longitude resolution by 5-degree latitude resolution) for most of the period. DSS interpolated, where possible, data to the original missing points to provide a 5-degree resolution in both directions.

The grids in this dataset contain Southern Hemisphere sea-level pressure and 500mb height data for the International Geophysical Year (IGY) period of June 1957 to December 1958. The data are on a 5-degree latitude/longitude grid covering the hemisphere from 15S to the South Pole.

The data originally came from South Africa where it was on a 10-degree by 5-degree diamond latitude/longitude grid. DSS interpolated data to the missing gridpoints to provide a resolution of five degrees in both directions.

This dataset contains a time-series of sea-level pressure data on a 63x63 polar-stereographic grid over the Southern Hemisphere. These grids were subsetted from the Navy grids in ds240.6 and cover the period from August 1974 to June 1983.

This dataset contains sea-level pressure data on a 5-degree latitude/longitude grid over the global tropics from 25S to 25N. The time period covered by the data is for the International Geophysical Year (IGY) of June 1957 to December 1958.

From the Republic of South Africa Weather Bureau, twice daily Southern Hemisphere grids on a polar stereographic projection.

From the Australian National Meteorological Research Centre, twice daily Southern Hemisphere grids on a polar stereographic projection. DSS derived the monthly mean grids. Later and ongoing data may be available directly from ABM.

This dataset contains 500mb heights for the Southern Hemisphere for the period June 1968 to August 1977. The grid has a resolution of 10 degrees in the longitudinal direction and 5 degrees in the latitudinal direction and covers the hemisphere from 15S to the South Pole.

The data were received from the Australian Numerical Meteorology Research Centre.

This dataset contains gridded sea-level pressure data that covers the area around New Zealand. The grid is a 16x16 grid over the area from 5S to 80S and 60E to 150W, with a resolution of five degrees in latitude and ten degrees in longitude. For the period from June 1957 to February 1958, the sea-level pressures were extracted from the 00Z meteorological charts.

From European Centre for Medium Range Weather Forecasts (ECMWF), twice daily grids. Monthly mean grids derived in U.S.

The ECMWF analyses for 1980-1989 were used to derive a global year-month wind stress climatology. Several climate monthly mean grids were also formed. Wind component grids are also available as are standard deviation grids for each variable computed.

The ECMWF analyses for 1980-1989 were used to derive global long term means.

The ds111.n collection of datasets archives output from ECMWF's daily operational global numerical model. Dataset ds111.0 contains geopotential, temperature, wind, and moisture fields on constant pressure levels, starting in 1985. The horizontal resolution of the DSS archive is T106 spectral resolution in spherical harmonics, however, in later years, the model is actually run at a much higher resolution (T511 as of 2001). All fields are output 4x a day at 00Z, 06Z, 12Z and 18Z. This dataset is updated twice a year. ECMWF maintains an online description of its operational model and its evolution.

The surface data can be found in ds111.1 and the supplemental fields inlcuding radiation,stress and flux terms are in ds111.3. The model precipitation is archived in ds111.6. The ECMWF operational model output on the 2.5 degree latitude-longitude grid can be found in ds111.2.

This dataset contains the high resolution output from the ECMWF operational model. This set contains only surface data on a gaussian (n80) grid with a resolution of about 1.125 degrees.

ECMWF now has a description of its operational model and its evolution online.

This dataset contains 2.5 degree latitude-longitude output grids from the ECMWF operational model. Both upper air and surface grids are archived in this dataset.

ECMWF now has a description of its operational model and its evolution.

This dataset contains supplementary output grids from the ECMWF operatoinal model. These grids are on a gaussian grid (n80) with approximtely a 1.125 degree resolution.

ECMWF now has a description of its operational model and its evolution online.

ECMWF now has a description of its operational model and its evolution online.

These are year-month means on 2.5 degree grids derived from DS111.2, by DSS. Both surface and upper air. Bad grids found by Kevin Trenberth in TN-373 are excluded.

ECMWF now has a description of its operational model and its evolution online.

This dataset contains various output grids from the ECMWF operational model. All of the grids are on a gaussian (n80) grid with a resolution of approximately 1.125 degrees daily accumulations from forecast hours 12 - 36 of 12Z run.

ECMWF now has a description of its operational model and its evolution online.

As part of a data exchange involving climate data, DSS received various global and hemispheric gridded analyses from the U.S.S.R. The analyses include a long time series of Northern Hemisphere surface pressure spanning the period 1880-1979, global total cloud amount for 1966-1985, and surface temperature, sea-level pressure, and upper-level geopotential heights for the overarching period of 1948-1984 (actual periods vary by parameter and atmospheric level).

This dataset contains surface grids from the ECMWF 15 year Reanalysis Project. All grids are on a 2.5 degree latitude- longitude grid and are archived 4x/day.

This dataset contains upper air grids from the ECMWF 15 year Reanalysis Project. All grids are on a 2.5 degree latitude- longitude grid and are archived 4x/day.

This dataset contains the supplementary fields from the ECMWF 15 year reanalysis project. All fields are on a 2.5 degree latitude-longitude grid and are accumlations derived from forecast grids.

This dataset contains the high resolution surface grids from the ECMWF 15 year reanalysis project. All of the grids are on a gaussian (N80) grid with an approximate resolution of 1.125 degrees and are archived 4x/day.

This dataset contains the high resolution uppper air grids from the ECMWF 15 year reanalysis project. All of the grids are in spherical harmonic representation (t106) which can be gridded to about of 1.125 degrees.

This dataset contains high resolution supplementary grids from the ECMWF reanalysis project. The grids are all on a reduced gaussian (n80) grid with an approximate resolution of 1.125 degrees. This fields are all derived from forecasts.

This dataset contains high resolution moisture grids from the ECMWF reanalysis project. The grids are all on a reduced gaussian (n80) grid with an approximate resolution of 1.125 degrees.

This dataset contains monthly mean data from the ECMWF reanalysis project. The grid are all 2.5 degree latitude-longitude grids. There are separate monthly means for each hour (00,06,12 and 18) as well as a daily mean.

The ECMWF Re-Analysis (ERA40) is a global atmospheric analysis of many conventional observations and satellite data streams for the period Sept,1957- Aug,2002. There are numerous data products that are separated into dataset series based on resolution, vertical coordinate reference, and likely research applications. Descriptions of the series organization and direct links to information about all ERA40 products are available.

This dataset has been segmented into 4 archives:

• Surface - 46 mostly surface and near surface fields, including 10m wind, 2m air temp, soil temp and moisture, cloud cover info
• Vertical Integrals - 35 fields including column water vapor, column ozone, kinetic energy
• Other - 27 fields, including precip, fluxes, radiation, stresses
• LNSP - Log of the surface pressure

This dataset contains ECMWF ERA40 model output consisting of geopotential, temperature, wind and moisture variables on 23 constant pressure surfaces for the period September, 1957 to August, 2002. All variables, except moisture, are archived at global T159 spectral resolution in spherical harmonics, while the moisture is archived on a reduced N80 gaussian grid with an approximate horizontal resolution of 150 kilometers. All data are archived 4x daily at 00z, 06z, 12z and 18z. All data is in GRIB format.

The ECMWF ERA40 project is a global atmospheric analysis of many conventional observations and satellite data streams for the period September,1957- August,2002. There are numerous data products that are separated into dataset series based on resolution, vertical coordinate reference, and possible research applications. Descriptions of the series organization and direct links to information about all ERA40 products are available

The ECMWF Re-Analysis (ERA40) is a global atmospheric analysis of many conventional observations and satellite data streams for the period Sept,1957- Aug,2002. There are numerous data products that are separated into dataset series' based on resolution, vertical coordinate reference, and likely research applications. Descriptions of the series organization and direct links to information about all ERA40 products are available.

This dataset contains 9 variables on the original 60 model levels, as well as surface geopotential and log of surface pressure. Variables are archived at T159 spectral resolution in spherical harmonics or the reduced n80 gaussian grid with a resolution of approximately 125 kilometers. All variables are output 4x a day for the entire period.

The ECMWF Re-Analysis (ERA40) is a global atmospheric analysis of many conventional observations and satellite data streams for the period Sept,1957- Aug,2002. There are numerous data products that are separated into dataset series' based on resolution, vertical coordinate reference, and likely research applications. Descriptions of the series organization and direct links to information about all ERA40 products are available.

This dataset contains 7 variables on 15 isentropic surfaces. Some of the variables are on a T159 spherical harmonics grid and some are on the N80 reduced gaussian grid. All variables are output 4x a day for the entire period.

The ECMWF Re-Analysis (ERA40) is a global atmospheric analysis of many conventional observations and satellite data streams for the period Sept,1957- Aug,2002. There are numerous data products that are separated into dataset series' based on resolution, vertical coordinate reference, and likely research applications. Descriptions of the series organization and direct links to information about all ERA40 products are available.

This dataset contains 7 variables on the PV+-2 surface. All of the variables are on the N80 reduced gaussian grid. All variables are output 4x a day for the entire period.

The ECMWF Re-Analysis (ERA40) is a global atmospheric analysis of many conventional observations and satellite data streams for the period Sept,1957- Aug,2002. There are numerous data products that are separated into dataset series based on resolution, vertical coordinate reference, and likely research applications. Descriptions of the series organization and direct links to information about all ERA40 products are available.

This dataset contains 4 variables on the original 60 hybrid model levels on the reduced N80 Gaussian grid. The data are accumulated over the 3 and 6 hour forecasts for the 00, 06, 12 and 18 UTC analyses.

The ECMWF Re-Analysis (ERA40) is a global atmospheric analysis of many conventional observations and satellite data streams for the period Sept,1957- Aug,2002. There are numerous data products that are separated into dataset series based on resolution, vertical coordinate reference, and likely research applications. Descriptions of the series organization and direct links to information about all ERA40 products are available.

This dataset contains the Net tendencies from parameterized processeses. There are 4 variables in this dataset, each of which are on the original 60 hybrid model levels on the reduced N80 Gaussian grid. The data are accumulated over the 3 and 6 hour forecasts for the 00, 06, 12 and 18 UTC analyses.

This dataset contains 56 surface and single level variables on an equally spaced global 2.5 degree latitude-longitude grid. All variables are reported 4x a day.

The ECMWF Re-Analysis (ERA40) is a global atmospheric analysis of many conventional observations and satellite data streams for the period Sept,1957- Aug,2002. There are numerous data products that are separated into dataset series based on resolution, vertical coordinate reference, and likely research applications. Descriptions of the series organization and direct links to information about all ERA40 products are available.

The ECMWF Re-Analysis (ERA40) is a global atmospheric analysis of many conventional observations and satellite data streams for the period Sept,1957- Aug,2002. There are numerous data products that are separated into dataset series' based on resolution, vertical coordinate reference, and likely research applications. Descriptions of the series organization and direct links to information about all ERA40 products are available.

This dataset contains 11 variables on 23 pressure surfaces on an equally spaced global 2.5 degree latitude-longitude grid. All variables are reported 4x a day for the entire period.

The ECMWF Re-Analysis (ERA40) is a global atmospheric analysis of many conventional observations and satellite data streams for the period Sept,1957- Aug,2002. There are numerous data products that are separated into dataset series based on resolution, vertical coordinate reference, and likely research applications. Descriptions of the series organization and direct links to information about all ERA40 products are available.

This dataset has been segmented into 3 archives:

• Surface - 47 mostly surface and near surface fields, including 10m wind, 2m air temp, soil temp and moisture, cloud cover info
• Vertical Integrals - 35 fields including column water vapor, column ozone, kinetic energy
• Accumulated (other) Fields - 27 fields including precip, fluxes, radiation, stresses

Monthly means have been computed for each analysis hour as well as a daily mean.

The ECMWF Re-Analysis (ERA40) is a global atmospheric analysis of many conventional observations and satellite data streams for the period Sept,1957- Aug,2002. There are numerous data products that are separated into dataset series' based on resolution, vertical coordinate reference, and likely research applications. Descriptions of the series organization and direct links to information about all ERA40 products are available.

This dataset contains monthly means for 9 variables on 23 pressure surfaces at T159 spectral resolution in spherical harmonics. Monthly means have been computed for each analysis hour as well as a daily mean.

The ECMWF Re-Analysis (ERA40) is a global atmospheric analysis of many conventional observations and satellite data streams for the period Sept,1957- Aug,2002. There are numerous data products that are separated into dataset series' based on resolution, vertical coordinate reference, and likely research applications. Descriptions of the series organization and direct links to information about all ERA40 products are available.

This dataset contains monthly means for 9 variables on the original 60 model levels, as well as surface geopotential and log of surface pressure. Variables are archived at T159 spectral resolution in spherical harmonics or the reduced n80 gaussian grid with a resolution of approximately 125 kilometers. All variables are output 4x a day for the entire period. Monthly means have been computed for each analysis hour as well as a daily mean.

The ECMWF Re-Analysis (ERA40) is a global atmospheric analysis of many conventional observations and satellite data streams for the period Sept,1957- Aug,2002. There are numerous data products that are separated into dataset series' based on resolution, vertical coordinate reference, and likely research applications. Descriptions of the series organization and direct links to information about all ERA40 products are available.

This dataset contains monthly means for 7 variables on 15 isentropic surfaces. Some of the variables are on a T159 spherical harmonics grid and some are on the N80 reduced gaussian grid. Monthly means have been computed for each analysis hour as well as a daily mean.

The ECMWF Re-Analysis (ERA40) is a global atmospheric analysis of many conventional observations and satellite data streams for the period Sept,1957- Aug,2002. There are numerous data products that are separated into dataset series' based on resolution, vertical coordinate reference, and likely research applications. Descriptions of the series organization and direct links to information about all ERA40 products are available.

This dataset contains monthly means for 7 variables on the PV+-2 surface. All of the variables are on the N80 reduced gaussian grid. All variables are output 4x a day for the entire period.

The ECMWF Re-Analysis (ERA40) is a global atmospheric analysis of many conventional observations and satellite data streams for the period Sept,1957- Aug,2002. There are numerous data products that are separated into dataset series' based on resolution, vertical coordinate reference, and likely research applications. Descriptions of the series organization and direct links to information about all ERA40 products are available.

This dataset contains variances and covariances on the original 60 model levels, as well as the variance of the surface pressure. All Variables are archived on the full n80 gaussian grid with a resolution of approximately 125 kilometers. The variances and covariances are produced with respect to the monthly means for each hour four synoptic hours (00, 06, 12 and 18 UTC) and for the diurnal mean.

The ECMWF Re-Analysis (ERA40) is a global atmospheric analysis of many conventional observations and satellite data streams for the period Sept,1957- Aug,2002. There are numerous data products that are separated into dataset series' based on resolution, vertical coordinate reference, and likely research applications. Descriptions of the series organization and direct links to information about all ERA40 products are available.

This dataset contains monthly means of the ERA40 2.5 dergree latitude-longitude gridded surface and single level analysis. Monthly means have been computed for each analysis hour as well as a daily mean.

The ECMWF Re-Analysis (ERA40) is a global atmospheric analysis of many conventional observations and satellite data streams for the period Sept,1957- Aug,2002. There are numerous data products that are separated into dataset series' based on resolution, vertical coordinate reference, and likely research applications. Descriptions of the series organization and direct links to information about all ERA40 products are available.

This dataset contains 11 variables on 23 pressure surfaces on an equally spaced global 2.5 degree latitude-longitude grid. Monthly means have been computed for each analysis hour as well as a daily mean.

The ECMWF Re-Analysis (ERA40) is a global atmospheric analysis of many conventional observations and satellite data streams for the period Sept,1957- Aug,2002. There are numerous data products that are separated into dataset series based on resolution, vertical coordinate reference, and likely research applications. Descriptions of the series organization and direct links to information about all ERA40 products are available.

This dataset contains 75 variables from the from the forecast cycle of the ERA-40 model. The forecasts extend out to 36 hours beginning from the 00 and 12 UTC analyses and out to 6 hours beginning with the 06 and 18 UTC analyses. The analysis fields have been archived in ds117.0. All data are archived on an N80 reduced gaussian grid 4x a day.

This dataset contains ECMWF ERA40 model forecasts of geopotential, temperature, wind and moisture variables on 23 constant pressure surfaces for the period September, 1957 to August, 2002. The forecasts extend out to 36 hours beginning from the 00 and 12 UTC analyses and out to 6 hours beginning with the 06 and 18 UTC analyses. All variables, except moisture, are archived at global T159 spectral resolution in spherical harmonics, while the moisture is archived on a reduced N80 gaussian grid with an approximate horizontal resolution of 125 kilometers. The anlaysis fields have been archived in ds117.1.

The ECMWF ERA40 project is a global atmospheric analysis of many conventional observations and satellite data streams for the period September,1957- August,2002. There are numerous data products that are separated into dataset series based on resolution, vertical coordinate reference, and possible research applications. Descriptions of the series organization and direct links to information about all ERA40 products are available

This dataset contains 9 variables on the original 60 model levels, as well as surface geopotential and log of surface pressure. All Variables are from the forecast cycle of the ERA-40 model and are archived at T159 spectral resolution in spherical harmonics or the reduced n80 gaussian grid with a resolution of approximately 125 kilometers. The forecasts extend out to 36 hours beginning from the 00 and 12 UTC analyses and out to 6 hours beginning with the 06 and 18 UTC analyses. The anlaysis fields have been archived in ds117.2.

The ECMWF Re-Analysis (ERA40) is a global atmospheric analysis of many conventional observations and satellite data streams for the period Sept,1957- Aug,2002. There are numerous data products that are separated into dataset series' based on resolution, vertical coordinate reference, and likely research applications. Descriptions of the series organization and direct links to information about all ERA40 products are available.

The ECMWF Re-Analysis (ERA40) is a global atmospheric analysis of many conventional observations and satellite data streams for the period Sept,1957- Aug,2002. There are numerous data products that are separated into dataset series based on resolution, vertical coordinate reference, and likely research applications. Descriptions of the series organization and direct links to information about all ERA40 products are available.

This dataset contains monthly means for 73 surface and single level variables from the forecast cycle of the ERA40 model. Means are available for the 3 and 6 hour forecasts for each synoptic hour (00, 06, 12 and 18 UTC) as well as a diurnal mean for the 6 hour forecast.

Monthly means have been computed for each analysis hour as well as a diurnal mean.

This dataset contains ECMWF ERA40 monthly mean model forecasts of geopotential, temperature, wind and moisture variables on 23 constant pressure surfaces for the period September, 1957 to August, 2002. The means have been computed for the 6 hour forecast only. All variables are archived at global T159 spectral resolution in spherical harmonics, except the moisture, ozone and potentical vorticity, which are archived on a reduced N80 gaussian grid with an approximate horizontal resolution of 125 kilometers. The monthly mean analysis fields have been archived in ds119.1.

The ECMWF ERA40 project is a global atmospheric analysis of many conventional observations and satellite data streams for the period September,1957- August,2002. There are numerous data products that are separated into dataset series based on resolution, vertical coordinate reference, and possible research applications. Descriptions of the series organization and direct links to information about all ERA40 products are available

The ECMWF Re-Analysis (ERA40) is a global atmospheric analysis of many conventional observations and satellite data streams for the period Sept,1957- Aug,2002. There are numerous data products that are separated into dataset series' based on resolution, vertical coordinate reference, and likely research applications. Descriptions of the series organization and direct links to information about all ERA40 products are available.

This dataset contains monthly means for 9 variables on the original 60 model levels, as well as surface geopotential and log of surface pressure. Variables are archived at T159 spectral resolution in spherical harmonics or the reduced n80 gaussian grid with a resolution of approximately 125 kilometers. Synoptic monthly means have been computed for each analysis hour and a diurnal mean has also been computed. Both means are only computed for the 6 hour forecast.

The ECMWF Re-Analysis (ERA40) is a global atmospheric analysis of many conventional observations and satellite data streams for the period Sept,1957- Aug,2002. There are numerous data products that are separated into dataset series based on resolution, vertical coordinate reference, and likely research applications. Descriptions of the series organization and direct links to information about all ERA40 products are available.

This dataset contains 23 variables from the ERA40 WAVE model on an equally spaced global 1.5 degree latitude-longitude grid . All variables are reported 4x a day.

There has been a problem with the wave analysis since the introduction of ERS data at the start of 1991. The size of waves of low magnitude (less than 1 metre) may be too high and the corresponding mean periods too high. The problem is largely confined to enclosed or semi-enclosed seas such as the Mediterranean Sea, Black Sea, Baltic Sea, the Gulf of Mexico and the East China Sea.

The ECMWF Re-Analysis (ERA40) is a global atmospheric analysis of many conventional observations and satellite data streams for the period Sept,1957- Aug,2002. There are numerous data products that are separated into dataset series based on resolution, vertical coordinate reference, and likely research applications. Descriptions of the series organization and direct links to information about all ERA40 products are available.

This dataset contains 23 variables from the ERA40 WAVE model on an equally spaced global 1.5 degree latitude-longitude grid . All variables are reported 4x a day.

DS124.0 represents a dataset implemented and computed by NCAR's Data Support Section, and forms an essential part of efforts undertaken in late 2004, early 2005, to produce an archive of selected segments of ERA-40 on a standard transformation grid.

In this case, forty six ERA-40 6-hourly surface and single level analysis variables were transformed from a reduced N80 Gaussian grid to a 256x128 regular Gaussian grid. (These are designated as "SURFACE", or "SURFACE/SINGLE LEVEL", or "SFC/sfc", variables below.) All fields were transformed using routines from the ECMWF EMOS library, including 10 meter winds which were treated as scalars because of a lack of 10 meter spectral vorticity and divergence. A missing value of 9.999e+20 occurs in the sea surface temperature and sea ice (SSTK and CI) fields to mask grid points occurring over land. Fields formerly archived as whole integers, such as vegetation indices and cloud cover, occur as integers plus a fractional part in the T85 version due to interpolation.

In addition, twenty seven ERA-40 6-hourly surface and single level 6-hour forecast variables were transformed from a reduced N80 Gaussian grid to a 256x128 regular Gaussian grid. (These are designated as "OTHER", or "othr", variables below.) All variables are valid 6 hours after the forecast was initiated. Thus, 00Z 6-hour forecast evaporation is valid at 06Z. Four of the variables are "instantaneous" variables, and the remaining twenty three variables are "accumulated" over the 6-hour forecast time. Divide the accumulated variables by 21600 seconds to obtain instantaneous values. (Multiplication by minus one may also be necessary to match the sign convention one is accustomed to.) All fields were transformed using routines from the ECMWF EMOS library, including three pairs of stresses which were treated as scalars because of a lack of spectral precursors.

The choice of a T85 Gaussian grid was based on considerations of limiting the volume of new data generated to a moderate level, and to match the horizontal resolution of the Community Atmosphere Model (CAM) component of NCAR's Community Climate System Model (CCSM).

The ECMWF Re-Analysis (ERA40) is a global atmospheric analysis of many conventional observations and satellite data streams for the period Sept,1957- Aug,2002. There are numerous data products that are separated into dataset series based on resolution, vertical coordinate reference, and likely research applications. Descriptions of the series organization and direct links to information about all ERA40 products are available.

DS124.1 represents a new data set implemented and computed by NCAR's Data Support Section, and forms a substantial part of ongoing efforts to produce an archive of selected segments of ERA-40 on a standard transformation grid. In this case, ERA-40 upper air variables on 23 pressure levels have been transformed from either spherical harmonics (geopotential, temperature, vertical pressure velocity, vorticity, divergence, relative humidity), or a reduced N80 Gaussian grid (potential vorticity, specific humidity, ozone mass mixing ratio), to a 256x128 regular Gaussian grid at T85 spectral truncation. In addition, horizontal wind components have been derived from spectral vorticity and divergence and also archived on a T85 Gaussian grid. All scalar fields have been transformed using routines from the ECMWF EMOS library, whereas the horizontal winds have been obtained using NCAR's Spherepack library. The choice of a T85 Gaussian grid was based on considerations of limiting the volume of new data generated to a moderate level, and to match the horizontal resolution of the Community Atmosphere Model (CAM) component of NCAR's Community Climate System Model (CCSM).

Sub-setting of DS124.1 is available to registered users of DSS ERA-40 data. Users may select a set of variables, and designate pressure levels, and start and end dates, for each variable. Output, in GRIB or netCDF, can be retrieved via restricted ftp.

DS124.2 represents a dataset implemented and computed by NCAR's Data Support Section, and forms an essential part of efforts undertaken in late 2004, early 2005, to produce an archive of selected segments of ERA-40 on a standard transformation grid. In this case, ERA-40 6-hourly upper air variables on 60 model levels were transformed from either spherical harmonics (surface geopotential, temperature, vertical velocity, vorticity, logarithm of surface pressure, divergence), or a reduced N80 Gaussian grid (specific humidity, ozone mass mixing ratio, cloud liquid water content, cloud ice water content, cloud cover), to a 256x128 regular Gaussian grid at T85 spectral truncation. In addition, horizontal wind components were derived from spectral vorticity and divergence and also archived on a T85 Gaussian grid. All scalar fields were transformed using routines from the ECMWF EMOS library, whereas the horizontal winds were obtained using NCAR's Spherepack library. The choice of a T85 Gaussian grid was based on considerations of limiting the volume of new data generated to a moderate level, and to match the horizontal resolution of the Community Atmosphere Model (CAM) component of NCAR's Community Climate System Model (CCSM).

DS126.0 represents a dataset implemented and computed by NCAR's Data Support Section, and forms an essential part of efforts undertaken in late 2004, early 2005, to produce an archive of selected segments of ERA-40 on a standard transformation grid.

In this case, forty seven ERA-40 monthly mean (designated as "MODA" or "moda" below) surface and single level analysis variables were transformed from a reduced N80 Gaussian grid to a 256x128 regular Gaussian grid. (These are further designated as "SURFACE", or "SURFACE/SINGLE LEVEL", or "SFC/sfc", variables below.) All fields were transformed using routines from the ECMWF EMOS library, including 10 meter winds which were treated as scalars because of a lack of 10 meter spectral vorticity and divergence. A missing value of 9.999e+20 occurs in the sea surface temperature and sea ice (SSTK and CI) fields to mask grid points occurring over land. Fields formerly archived as whole integers, such as vegetation indices and cloud cover, occur as integers plus a fractional part in the T85 version due to interpolation.

Twenty seven ERA-40 monthly mean (designated as "MODA" or "moda" below) surface and single level 6-hour forecast variables were transformed from a reduced N80 Gaussian grid to a 256x128 regular Gaussian grid. (These are further designated as "OTHER", or "othr", variables below.) Four of the variables are "instantaneous" variables, and the remaining twenty three variables are "accumulated" over the 6-hour forecast time. Divide the accumulated variables by 21600 seconds to obtain instantaneous values. (Multiplication by minus one may also be necessary to match the sign convention one is accustomed to.) All fields were transformed using routines from the ECMWF EMOS library, including three pairs of stresses which were treated as scalars because of a lack of spectral precursors.

In addition, all corresponding 00Z, 06Z, 12Z, and 18Z monthly mean surface and single level analysis variables and 6-hour forecast variables were also transformed to a T85 Gaussian grid (designated as "MNLY", "mnly", or "mnth" below)

All forecast variables are valid 6 hours after the forecast was initiated. Thus, 00Z 6-hour forecast evaporation is valid at 06Z. Divide the accumulated variables by 21600 seconds to obtain instantaneous values. (Multiplication by minus one may also be necessary to match the sign convention one is to.)

The choice of a T85 Gaussian grid was based on considerations of limiting the volume of new data generated to a moderate level, and to match the horizontal resolution of the Community Atmosphere Model (CAM) component of NCAR's Community Climate System Model (CCSM).

DS126.1 represents a dataset implemented and computed by NCAR's Data Support Section, and forms an essential part of efforts undertaken in late 2004, early 2005, to produce an archive of selected segments of ERA-40 on a standard transformation grid. In this case, ERA-40 monthly mean upper air variables (designated as “MODA” or “moda” below) on 23 pressure levels were transformed from either spherical harmonics (geopotential, temperature, vertical pressure velocity, vorticity, divergence, relative humidity), or a reduced N80 Gaussian grid (potential vorticity, specific humidity, ozone mass mixing ratio), to a 256x128 regular Gaussian grid at T85 spectral truncation. In addition, horizontal wind components were derived from spectral vorticity and divergence and also archived on a T85 Gaussian grid. All scalar fields were transformed using routines from the ECMWF EMOS library, whereas the horizontal winds were obtained using NCAR's Spherepack library. All corresponding 00Z, 06Z, 12Z, and 18Z monthly mean upper air variables on 23 pressure levels were also transformed to a T85 Gaussian grid (designated as “MNLY”, “mnly”, or “mnth” below). The choice of a T85 Gaussian grid was based on considerations of limiting the volume of new data generated to a moderate level, and to match the horizontal resolution of the Community Atmosphere Model (CAM) component of NCAR's Community Climate System Model (CCSM).

DS126.2 represents a dataset implemented and computed by NCAR's Data Support Section, and forms an essential part of efforts undertaken in late 2004, early 2005, to produce an archive of selected segments of ERA-40 on a standard transformation grid. In this case, ERA-40 monthly mean upper air variables (designated as “MODA” or “moda” below) on 60 model levels were transformed from either spherical harmonics (surface geopotential, temperature, vertical pressure velocity, vorticity, logarithm of surface pressure, divergence), or a reduced N80 Gaussian grid (specific humidity, ozone mass mixing ratio, cloud liquid water content, cloud ice water content, cloud cover), to a 256x128 regular Gaussian grid at T85 spectral truncation. In addition, horizontal wind components were derived from spectral vorticity and divergence and also archived on a T85 Gaussian grid. All scalar fields were transformed using routines from the ECMWF EMOS library, whereas the horizontal winds were obtained using NCAR's Spherepack library. All corresponding 00Z, 06Z, 12Z, and 18Z monthly mean upper air variables on 60 model levels were also transformed to a T85 Gaussian grid (designated as “MNLY”, “mnly”, or “mnth” below). The choice of a T85 Gaussian grid was based on considerations of limiting the volume of new data generated to a moderate level, and to match the horizontal resolution of the Community Atmosphere Model (CAM) component of NCAR's Community Climate System Model (CCSM).

DS127.0 represents a dataset implemented and computed by NCAR's Data Support Section, and forms an essential part of efforts undertaken in late 2004, early 2005, to produce an archive of selected segments of ERA-40 on standard transformation grids.

In this case, forty six ERA-40 6-hourly surface and single level analysis variables were transformed from a reduced N80 Gaussian grid to a 320x160 regular Gaussian grid. (These are designated as “SURFACE”, or “SURFACE/SINGLE LEVEL”, or “SFC/sfc”, variables below.) All fields were transformed using routines from the ECMWF EMOS library, including 10 meter winds which were treated as scalars because of a lack of 10 meter spectral vorticity and divergence. A missing value of 9.999e+20 occurs in the sea surface temperature and sea ice (SSTK and CI) fields to mask grid points occurring over land. Fields formerly archived as whole integers, such as vegetation indices and cloud cover, occur as integers plus a fractional part in the T106 version due to interpolation.

In addition, twenty seven ERA-40 6-hourly surface and single level 6-hour forecast variables were transformed from a reduced N80 Gaussian grid to a 320x160 regular Gaussian grid. (These are designated as “OTHER”, or “othr”, variables below.) All variables are valid 6 hours after the forecast was initiated. Thus, 00Z 6-hour forecast evaporation is valid at 06Z. Four of the variables are “instantaneous” variables, and the remaining twenty three variables are “accumulated” over the 6-hour forecast time. Divide the accumulated variables by 21600 seconds to obtain instantaneous values. (Multiplication by minus one may also be necessary to match the sign convention one is accustomed to.) All fields were transformed using routines from the ECMWF EMOS library, including three pairs of stresses which were treated as scalars because of a lack of spectral precursors.

DS127.1 represents a higher resolution version of ds124.1. Both ds124.1 and 127.1 were implemented and computed by NCAR's Data Support Section, and form an essential part of efforts undertaken in late 2004, early 2005, to produce an archive of selected segments of ERA-40 on a standard transformation grid. In the case of ds127.1, ERA-40 6-hourly upper air variables on 23 pressure levels were transformed from either spherical harmonics (geopotential, temperature, vertical pressure velocity, vorticity, divergence, relative humidity), or a reduced N80 Gaussian grid (potential vorticity, specific humidity, ozone mass mixing ratio), to a 320x160 regular Gaussian grid at T106 spectral truncation. In addition, horizontal wind components were derived from spectral vorticity and divergence and also archived on a T106 Gaussian grid. All scalar fields were transformed using routines from the ECMWF EMOS library, whereas the horizontal winds were obtained using NCAR's Spherepack library.

DS127.2 represents a dataset implemented and computed by NCAR's Data Support Section, and forms an essential part of efforts undertaken in late 2004, early 2005, to produce an archive of selected segments of ERA-40 on standard transformation grids. In this case, ERA-40 6-hourly upper air variables on 60 model levels were transformed from either spherical harmonics (surface geopotential, temperature, vertical velocity, vorticity, logarithm of surface pressure, divergence), or a reduced N80 Gaussian grid (specific humidity, ozone mass mixing ratio, cloud liquid water content, cloud ice water content, cloud cover), to a 320x160 regular Gaussian grid at T106 spectral truncation. In addition, horizontal wind components were derived from spectral vorticity and divergence and also archived on a T106 Gaussian grid. All scalar fields were transformed using routines from the ECMWF EMOS library, whereas the horizontal winds were obtained using NCAR's Spherepack library.

DS129.1 represents a higher resolution version of ds126.1. Both ds126.1 and 129.1 were implemented and computed by NCAR's Data Support Section, and form an essential part of efforts undertaken in late 2004, early 2005, to produce an archive of selected segments of ERA-40 on a standard transformation grid. In the case of ds127.1, ERA-40 monthly mean upper air variables (designated as “MODA” or “moda” below) on 23 pressure levels were transformed from either spherical harmonics (geopotential, temperature, vertical pressure velocity, vorticity, divergence, relative humidity), or a reduced N80 Gaussian grid (potential vorticity, specific humidity, ozone mass mixing ratio), to a 320x160 regular Gaussian grid at T106 spectral truncation. In addition, horizontal wind components were derived from spectral vorticity and divergence and also archived on a T106 Gaussian grid. All scalar fields were transformed using routines from the ECMWF EMOS library, whereas the horizontal winds were obtained using NCAR's Spherepack library.

This and later data is now completely contained in The Berlin Stratospheric Data Series cdrom, created by Karin Labitzke and Collaborators.

If you acquire data from this CD, please acknowledge: K. Labitzke and Collaborators, 2002: The Berlin Stratospheric Data Series, CD from Meteorological Institute, Free University Berlin.

The data described below is retained for historical purposes.

The weekly stratospheric analyses of geopotential height and temperature in this dataset were produced from available radiosonde and rocketsonde data for the 5mb, 2mb, and 0.4mb levels. Grids are available for the months of January, April, July, and October from 1964 through 1966 and cover the western Northern Hemisphere from 140E longitude to 360E.

Satellite radiance information was used to produce a "first guess" of the geopotential height and temperature fields at the 5mb, 2mb, and 0.4mb stratospheric levels, and available rocketsonde data was used to adjust the fields through an objective analysis procedure. The 65x65 polar-stereographic grids cover the Northern Hemisphere and are available weekly from July 1976 through April 1980 for the months of January, April, July, and October.

This dataset contains analyses of geopotential height and temperature at the 100mb, 10mb, 5mb, and 2mb stratospheric levels over the Northern Hemisphere. The data are on a latitude/longitude grid with a 10-degree resolution. The analyses were produced by the National Meterological Center (NMC) from available radiosonde and rocketsonde data, and they are available at weekly intervals from January 1972 through June 1973.

This dataset contains a time-series collection of daily gridded Northern Hemisphere tropospheric data on a 47x51 polar-stereographic grid. These grids have been assembled by DSS from the analyses of various operational models and meteorological projects, and they cover the overarching period from January 1946 forward, although periods for individual parameters and levels will vary. The grids continue to be updated with new data.

Available parameters include sea-level pressure, surface and sea-surface temperatures, and upper-level geopotential height, temperature, wind, and vertical motion at levels from 1000mb up to 100mb.

The polar-stereographic grids in ds195.0 were interpolated to create this dataset of 5-degree Northern Hemisphere latitude/longitude grids until 1976, when the National Meteorological Center (NMC) began producing its operational analyses on a latitude/longitude grid.

This dataset contains various tropospheric analysis time series' created from the grids of other DSS datasets and converted to GRIB (for grids that were not already in this format).

Currently, the only available time series' are series' of 500mb geopotential height.

Version III CD-ROMs include twice daily analyses from 1946 through December 1994. Selected daily grids (not on CD-ROMs ) of 1995-2003 are now available as of 2004APR06.

A selected climatology of the Southern Hemisphere using the Navy Marine Atlas Series and monthly mean rawinsonde data from "climat temp" reports as published in NCC's "Monthly Climatic Data for the World".

The data is free and has been combined with the Northern Hemisphere data in ds205.0.

A climatology of the Southern Hemisphere from Australia's Bureau of Meteorology Research Centre.

A selected 15-year climatology of the Northern Hemisphere using the Navy Marine Atlas Series and 5-year mean grids summarized from daily grids from the National Climatic Center (NCC). The input data is from about January 1950 through December 1964.

A selected 17-year climatology of the Southern Hemisphere using the Navy Marine Atlas Series and monthly mean rawinsonde data from "climat temp" reports as published in NCC's "Monthly Climatic Data for the World". The input data is from about 1950 through about 1966.

Climatologies for January and July were created for several levels and variables for the tropics and mid-latitudes from analyzed maps. Data from 1961 to 1980 were used to create the 2.5-degree resolution gridded climatologies.

Surface and Upper air climatology. Upper air climatology interpolated from data supplied by NCAR from DS200.0and DS205.0, and data supplied by NOAA, USAF(ETAC), and others.

Long term monthly mean winds which Sadler derived from his collection of aircraft data (which we have in DS365.0.) and average rawindsonde data. The data are free from the ds208.0 data directory. The aircraft data were obtained from two sources: operational GTS reports collected in Honolulu and the FHWF; and aircraft logs from many routes which often were not reported over the GTS. The aircraft winds were summarized for each month in 5-degree latitude-longitude squares. The average monthly rawindsonde data were then combined with these in manual analyses of streamlines and isotachs. The 2.5- degree grids of wind speed and direction were manually read from these.

Aircraft and rawinsonde (300, 250, and 200mb) observations over the tropical Pacific were analyzed to produce monthly grids of wind at approximately 250mb for the periods January 1966 to September 1968 and August 1970 to December 1973. The grid has a resolution of 2.5 degrees and runs from 30S to 45N latitude and 75E to 70W longitude.

Based on 1850-1974 data.

Includes annual and 20 year averages.

Contains monthly grids of surface parameters.

Heat Budget climatology components for the Topical Eastern Pacific and Atlantic Oceans.

This and later data is now completely comtained in The Berlin Stratospheric Data Series cdrom, created by Karin Labitzke and Collaborators.

If you acquire data from this CD, please acknowledge: K. Labitzke and Collaborators, 2002: The Berlin Stratospheric Data Series, CD from Meteorological Institute, Free University Berlin.

The data described below is retained for historical purposes. It should be noted that analyses in the stratosphere are particularly sensitive to changes in instrumentation and analysis methods. The German methods have been quite stable.

Northern Hemisphere monthly surface temperature anomalies for 1891-1972 from the U.S.S.R. as part of an agreement on the exchange of climate data.

Analyses of Northern Hemisphere surface temperature received from the U.S.S.R. as part of an agreement on the exchange of climate data.

Some selected data, managed by a variety of people and projects within the Climatic Research Unit (CRU), University of East Anglia.

This data is periodically downloaded as updates become available. Please appropriately acknowledge any use of this data; referencing suggestions are available on the free data page.

The "classic" derived by Bram Oort from MIT, GTS, ETAC, NCDC and NCAR data. See also DS431.0.

Twenty-two monthly statistical quantities make up this dataset of grids which describes the atmospheric general circulation at eleven vertical levels across the troposphere and stratosphere over a 32-year period. These gridded analyses, created from upper air soundings across the globe, span the period from May 1958 to December 1989. The statistical parameters included, which are useful for comparing climate model simulations to actual observations, include monthly means and variances of horizontal and vertical wind, temperature, geopotential height, moisture, and horizontal wind divergence. There are also covariance terms describing the relationship between wind, temperature, moisture, and geopotential height.

In general, the analyses near the surface of the earth (900mb and below) tend to be less reliable than those of the higher levels because the individual rawinsonde reports are often incomplete due to missing 1000mb reports.

The data were provided to DSS by Bram Oort of the Geophysical Fluid Dynamics Laboratory (GFDL). The original grids are in an IEEE floating-point format, and DSS has created a netCDF version of the full dataset.

This dataset contains monthly longterm climatological means, variances, and covariances of geopotential height, temperature, specific humidity, wind, and vertical motion for the atmosphere from 1000mb up to 50mb. Longterm means are computed over the period from October 1978 to September 1985. The data are on a global latitude/longitude grid with a resolution of 2.5 degrees.

This dataset contains longterm means of geopotential height, temperature, and wind for the troposphere from 1000mb up to 100mb. Means are computed for the period from 1979 to 1988 and are stored on a global latitude/longitude grid with a resolution of 2.5 degrees.

This dataset contains yearly, six-year, and ten-year seasonal means of geopotential height, temperature, wind, specific humidity, and vertical motion for atmospheric levels from 1000mb up to 30mb. These means are based on the ECMWF daily operational analyses and are on a global 5-degree latitude/longitude grid.

The ten-year period is March 1979 to February 1989. The six-year climatology is included because the tropical fields were improved significantly when diabatic initialization was introduced into the analysis scheme in September 1982.

More information is available.

U.S.A.F. ETAC Global Tropospheric and Stratospheric Analyses, monthly 1956-1960

Monthly analysis of Arctic surface temperature.

A historic monthly gridded precipitation dataset for global land.

Ship and buoy observed winds are used at FSU/COAPS to produce objectively analyzed grids of psuedo-wind stress for the tropical Pacific and Indian Ocean.

Note: this version of ds230.0 data has been superseded.

The Center for Ocean-Atmospheric Prediction Studies at the Florida State University has a {new in-situ based air-sea flux product}. The product, known as the FSU3, is an objectively derived monthly surface flux product based only on individual ship, buoy, and drifter observations. Fields of sensible and latent heat flux, wind stress, pseudo-wind stress, and the parameters needed to derive these fluxes (wind speed, air and sea temperature, specific humidity) are available.

Currently, the FSU3 is available for the Atlantic Ocean (north of 34S) from 1978-2004 and for the Indian Ocean (north of 30S) from 1978-1997. An extension of the Indian Ocean through 2004 will be release shortly and the 1978-2004 Pacific product will be available by the end of 2006. All products are based on the ICOADS versions 2.1 (1978-1997) and version 2.2 (1998-2004).

Tropical Pacific Ocean Wind And Wind Stress

This is a 12 month wind stress climatology derived from ocean surface wind observations, 1870-1976.

A monthly mean climatological wind stress was derived from the National Climate Center TDF-11 ocean observations. The dataset was derived in 1989 using the Large and Pond algorithm for ocean wind stress.

Using various sources for sea ice concentration John Walsh and Bill Chapman have derived a monthly time series of Arctic and Antarctic sea ice concentration.

Seasonal longterm climatology of sea ice thickness in the arctic digitized by Benjamin Felzer from Bourke and Garrett maps from submarine underice sonar profiles.

Monthly sea-ice area data is available for each 10 degrees of longitude for Antarctica starting January 1973. These data are based on weekly charts prepared by ohe Navy-NOAA Joint Ice Center.

This data set contains Northern Hemisphere surface pressure and precipitation data prepared by the U.S.S.R. The surface pressure grids include average monthly pressures for 1873-1981 and climatological normals by month for 1873-1973 and 1931-1960. Precipitation grids include percentages of normal for all Januarys and all Julys for 1891-1979.

DSS has created a GRIB version of this set for users who prefer the GRIB format.

Cort J. Willmott, Kenji Matsuura and David R. Legates (Center for Climate Research, University of Deleware), created a new version of this dataset (version 2.01) in November 1998. You may find more recent data at the Willmott site.

Willmott and Rowe have created a terrestrial water budget data archive on a 1-degree grid. They identified 13332 global stations with complete records and created 30-year (1950-1979) climatological means for each month of the year for air temperature, precipitation, evaporation, soil moisture, and snow cover. These monthly climatological means were then interpolated to the 1-degree gridpoints. This data set contains the 1-degree gridpoint data and the individual station data.

DSS has created a GRIB version of the gridpoint data for users who prefer the GRIB format.

This is a dataset of world monthly precipitation data from stations, and the associated monthly gridded analyses, 1851-1989. This data has been contributed to by a number of PIs. It has the NCAR monthly surface tape data plus data added by Jones, Bradley, Diaz, Eischeid, NCDC, etc. It is a later version of the "DOE" tape.

This is a monthly merged precipitaion data set. The data has been merged from rain gauge measurements, IR, SSMI and the ECMWF model results. Contact Data Support at NCAR to obtain more information on the blending scheme. Note that these are only preliminary results.

Earlier grids DSS received "stand alone," later grids were extracted by DSS from the "full stack" of grids in DS240.1

Fleet Numerical Oceanographic Center in Monterey California prepares many analyses, from which we receive this surface subset on a 63x63 grid. It includes sea surface temperature, air surface temperature, surface vapor pressure, 19.5m surface winds, and sea level pressure.

Extracted by DSS from the "full stack" of grids in DS240.6

Prepared by Fleet Numerical Oceanographic Center in Monterey California.

These grids, prepared by the Fleet Numerical Oceanographic Center in Monterey, California are in addition to what you will find in DS240.1 and DS240.6. This set contains 125x125 hemispheric analyses, spherical (global) analyses and band (tropical) analyses. We do not have a complete time series of any one of these grids.

Analyzed at the TOGA Sea Level Data Center.

Collected for EPOCS by CAC from various sources.

NOTE: THIS DATASET HAS BEEN SUPERSEDED BY OTHER DATASETS. See the links section below.

Results from water analysis using shipboard automated gas chromatography. CDIAC dataset NDP044, Weiss, et. al.

The phosphate, nitrate and silicate data used to produce the monthly climatologies were taken from the World Ocean Atlas 1998 produced by the Ocean Climate Laboratory at the National Oceanographic Data Center (NODC). The monthly grids are analyzed onto the top 14 NODC standard levels 0, 10, 20, 30, 50, 75, 100, 125, 150, 200, 250, 300, 400 and 500 m. The horizontal resolution is 2 degrees x 2 degrees.

SST and buoy position every six hours

Quality controled drifting buoy data - input for COADS

These data are collected from equatorial and near equatorial moored buoys and some low lying island stations.

University of Washington Polar Science Center ARGOS buoy data for January-June and October-December 1987. This data is supplied by the National Snow and Ice Data Center. Dataset includes gridded surface temperature for 1979-1986 and pressure for 1979-1990.

Data from Canadian buoys through Nov. 1998, data from two NOAA buoys now operated by Canada, and data from three Canadian waverider buoys.

The Upper Ocean Processes (UOP) Group at WHOI, through research projects, has deployed Air-Sea Interaction Meteorology (ASIMET) systems on moored buoys and volunteer observing ships. The ASIMET system is a suite of meteorological and sea surface sensors that are capable of measuring air temperature, specific humidity, sea surface temperature and conductivity, wind speed and direction, barometric pressure, shortwave radiation, longwave radiation, and precipitation. Many more details about the projects and online data access are provided at the UOP website.

Time series of water temperature and salinity collected at various points on Canadian coasts.

Twenty-nine time series data files with temperature and salinity recorded at approx. 14 locations along the west coast of the U.S.

Nantucket, Diamond Shoals, and other Lightship data from a digitization project at NCDC

Western Arctic ice drifting stations, AIDJEX, ARLIS I, ARLIS II, Ice Station Alpha, Ice Station Charlie, T-3 (called Ice Station Bravo during IGY), ship Maud and Fram provide surface meteorological data for the Arctic Ocean. These data were collected and organized by NCDC and NSIDC leading to the production of a CD-ROM.

Several different versions of the data are available, extending from the original key entered data to variable subsets and QC versions developed in uniform format for all stations.

Baltic Sea Ship database, 1961-1990, from Miroslaw Mietus (Poland), May 2000.

There are 360k records covering the Baltic Sea. These QC'ed data are a result from the RA VI Project, Climate of the Baltic Sea Basin. The project report is available from WMO (1998, WMO/TD-No.933, Marine Meteorology and Related Oceanographic Activities, Report No.41).

Marine surface and upper air observations recovered from microfilm copies of cards at NCDC. In the 1960's-1970's NCDC stored images of selected 80-character punched card decks on microfilm (16mm) using a device known as FOSDIC (Film Optical Sensing Device for Input to Computer). FOSDIC was also used to recall the images back to digital records.

Field measurements of nutrients, chlorophyll, primary production, bacteria, HPLC pigments, and other biogeochemical parameters are compiled for nine JGOFS time-series and process study sites. Data are provided both as depth profile measurements and as mixed layer averages, and were designed to facilitate regional and global ecosystem model development by providing a consistent set of observations for key JGOFS sites.

Data are presented for four Time-Series locations: BATS, HOT, KERFIX, and Station P, and four U.S. JGOFS Process Study sites: Arabian Sea, Equatorial Pacific (EqPac), North Atlantic Bloom Experiment (NABE), and the U.S. Southern Ocean Survey (which includes two distinct regions: the Ross Sea and Antarctic Polar Front Zone).

More information about JGOFS can be found at the following websites.

• U.S. JGOFS program
• International JGOFS program

Grids of temperature from XBT's taken during 1950-1993 were used to contruct grids on 11 levels from 0 to 400 meters.

The current version of OAFlux includes daily and monthly-mean, 1x1 degree gridded data, global surface latent and sensible heat fluxes for the period from January 1, 1981 to December 31, 2002; and net longwave and shortwave radiations for the period from January 1, 1983 to December 31, 2002.

The global surface latent and sensible heat fluxes from the OAFlux project (Yu and Weller, 2006; Yu et al. 2004) have been developed and validated against in situ flux measurements. Data from 2003 onward will be made public after validation. The net longwave and shortwave radiations are derived from ISCCP (Zhang et al. 2004) are in three-hour resolution and on 2.5x2.5 degree grid and is kindly provided by Dr. William B. Rossow for distribution along with the OAFlux products. The data are daily/monthly averaged and re-gridded to match OAFlux's temporal and spatial resolution.

Based on digitized charts available prior to 1970.

Monthly Average SST and Ice-Pack Limits, based on 1970's and pre-1970's data, derived from DS270.0 and Navy data.

Based on 2 degree COADS and monthly SST climatology for 1950-1979.

Monthly Climatology based on 1900-1979 COADS 2x2 degree monthly summary data, and in the case of surface pressure the NCAR 5x5 long term climatology.

Derived from FCDS Navy data.

Tropical Pacific Ocean monthly grids and climatology of SST and wind.

Monthly Pacific sea-surface temperature analyses for 1949 to 1962.

Northern Hemisphere analyses from the U.K. Meteorological Office.

Weekly and Monthly 1x1 global analyses are available for November 1981 through a current date. These are determined by blending marine surface observations and satellite AVHRR data using an OI method.

Monthly 2x2 global analysis for 1854-2002 is available in a collection termed "Extended Reconstructed Historical Monthly Analyses". EOF methods are used to create these grids using marine surface data (ICOADS).

Various climate indicies (e.g. SOI) are available from the CPC

Recent weekly SST plots and other products are available at the NCEP Climate Monitoring Bulletin for SST Analysis

Monthly average fields derived from weekly analysis using the NCEP Ocean-Data Assimilation System (ODAS). Please reference ds277.6 for data of the current version of NCEP Global Ocean Data Assimilation System (GODAS).

NOTE: The GISST time series will be discontinued on February 2003. HadISST , also from the Hadley Centre for Climate Prediction and Research, is a superior product. We recommend the usage of HadISST over GISST.

Monthly longterm global SST and sea ice analysis. In situ sea surface observations and satellite derived estimates at the sea surface are included in the analysis. SST bucket corrections have been applied to gridded fields from 1871 through 1941. And a blend of satellite AVHRR (for SST), SSMI (for ice), and observations are used in the modern periods.

Access to GISST 2.3b has proprietary restrictions. Users must register with the authors before receiving the data. This is easy to do. Simply print a copy of the registration form restr_form) and follow the FAX instructions. When this is completed contact Steve Worley (NCAR) worley@ucar.edu and the data will be made available.

HadISST is a unique combination of monthly globally-complete fields of SST and sea ice concentration on a 1 degree latitude-longitude grid from 1870 to date. In situ SST observations and satellite derived estimates at the sea surface are included in the analyses. SST measurements taken using the water bucket method have been corrected and used to determine the gridded fields from 1870 through 1941. When satellite data become available, in the early 1980's, they (SSMI for sea ice and AVHRR for SST) are used in conjunction with the observations to form a blended analysis. This high quality data covers over 130 years and is regularly extended to a current date.

The data may be obtained from a data server at the Hadley Centre in the UK. Here the registration may be filed online and data may be downloaded directly. The data at NCAR will be updated infrequently for archive purpose. Access from the Hadley Centre is recommended.

The Met Office Hadley Centre's mean sea level pressure (MSLP) data set, HadSLP1, replaces the Global Mean Sea Level Pressure (GMSLP) data sets, and is a unique combination of monthly globally-complete fields of land and sea pressure observations a 5 degree latitude-longitude grid from 1871 to 1998.

NOTE: The Hadley Centre now makes this data freely available from their observations datasets webpage. The data at NCAR will be updated infrequently for archive purposes. Access from the Hadley Centre is recommended.

HadSLP2, which runs from 1850-2003, has now been developed. The product is the final stages of evaluation and is planned for release in February 2006.

The extended reconstructed sea level pressure (ERSLP) analysis was constructed using the most recently available Comprehensive Ocean-Atmosphere Data Set (COADS) SLP data and improved statistical methods that allow stable reconstruction using sparse data. This monthly analysis begins January 1854, however the analyzed signal is heavily damped before 1910 because of sparse data.

All the data are from the original NOAA ERSLP website.

The National Centers for Environmental Prediction (NCEP) uses a Global Ocean Data Assimilation System (GODAS) to routinely produce monthly and pentad (5-day) mean global ocean gridded data. The analysis system, that was originally applied only to the tropical pacific, has evolved to use GFDL MOM.v3 model and give near global coverage. Briefly, the model domain specifications are 65N to 75S, 1 degree resolution with 1/3 degree resolution within 10 degree of the equator, 40 levels and 10 meter resolution in the upper 200 meters. These data are a valuable community asset for monitoring different aspects of ocean climate variability. Please click introduction_godas.pdf for more information.

Reference Time series of monthly averages by variable derived from GODAS operational datasets for monthly averaged data in NetCDF format.

Historical data archives from NUC and NODC were combined to form monthly and annual grid of subsurface ocean temperature and salinity.

Samuel and Cox (GFDL) took 15 commonly used data sets for ocean models and put them on a uniform 1x1 degree grid. Original data sources include Levitus, Hellerman, Esbensen and Kushnir to name just a few.

In situ sea surface salinity measurements from the tropical Pacific Ocean were used to derive monthly grids (1969-1995), and annual and monthly climatologies based on the 1979-1992 period. The source "bucket" data (1969-1998) and thermosalinograph data (1990-1999) from ships are also included in the dataset. These data along with moored buoy, hydrocast, and CTD-derived data were used to compute the gridded fields.

Derived from pilot charts from Noo, of the US Naval Oceanographic Office. Based on data from 1901-1978.

This dataset contains monthly global marine data by Marsden (10-degree) square. The period covered is 1850 to 1970, and available parameters include sea-surface temperature, air pressure and temperature, wet-bulb temperature, wind speed, and wave data.

This dataset contains monthly surface temperature analyses for 1880-1987 for eighty equal-area zones over the globe.

The World Ocean Database (WOD) has observed data from the ocean surface and subsurface profiles. The World Ocean Atlas (WOA) has gridded analyses for annual, seasonal, and monthly means of temperature, salinity, oxygen, nutrients, and other parameters on standard levels (typically at 33 depths). Both products were derived at the Ocean Climate Laboratory (OCL) of the National Oceanographic Data Center (NODC). Development of the World Ocean products began in 1982 and they have been updated in 1994, 1998, and 2001. The featured products are now based on the 2001 versions. Some products from older version are still unique.

The WOD is the world's largest profile collection covering the global ocean and contains measurements from 1874 through 2000. The WOA are climatological grids at fixed depths derived from quality checked WOD profiles. This is a recognized world-wide standard and is often referenced in oceanographic research.

WOD05 was published June 2006. These data are NOT available here, please go to the Ocean Climate Laboratory web site as shown in "Related Sites" below.

More information is available

Gordon (Columbia University), Molinelli (Science Applications Inc.), and Baker (Lamont-Doherty Geological Observatory) assembled the high-quality station data from the Southern Ocean found in the NODC archive. These individual stations (over 6000) are available (11 MBytes). The data were also interpolated to a uniform grid at 47 levels (14 MBytes). The variables include; temperature, salinity, sigma-t, silicate, oxygen, phosphate, and nitrate. Further information and contour maps are given in "Southern Ocean Atlas, published by the Columbia University Press".

The climatology provides temperature and salinity data at 1X1 degree intervals for all the earth's oceans, down to a depth of 5500m, at incremental depths identical to those provided in the National Oceanographic Data Center's (NODC) World Ocean Atlas (WOA). This global climatology is the combination of NODC's 1998 world climatology (WOA), the EWG Arctic Ocean Atlas (AOA), and select Canadian data provided by the Bedford Institute of Oceanography (BIO). While the NODC data includes the Arctic ocean, the AOA data provides a better description of this region. Neither of these data fields provides a good representation of the Canadian Archipelago region and nearby bays in the winter months. The data from the Bedford Institute of Oceanography allowed us to bridge this data hole. These three data sets were merged using an optimal interpolation routine such that our PHC retains the high quality world description provided by the WOA while improving the Arctic with the AOA fields and Canadian data.

In summary, PHC = WOA (Levitus) '98 everywhere except in our arctic domain, where we have blended in the AOA field (from EWG), and the BIO data to produce a more realistic arctic region.

These Version 3.0 data were received from the Polar Science Center, Applied Physics Lab., University of Washington, during April 2005.

Monthly objectively analyzed subsurface temperature (Version 6.1) at 16 levels in the upper 700m for 1945-2003 prepared by M. Ishii et al.

The analysis is based on the World Ocean Database 2001 (WOD01) (Boyer et al. 2001, ds285.0), the Global Temperature-Salintiy in the tropical pacific from IRD (L'institut de recherche pour le developement, France, ds279.1), and the Centennial in situ Observation Based Estimates (COBE) sea surface temperature (Ishii et al. 2005). ARGO profiling buoy data have also been used in the final several years.

The WOCE Global Hydrographic Climatology has been derived using World Ocean Data 1998 (WOD98), described in dataset ds285.0, and profiles from the WOCE Hydrographic Programme, with the exception of a few cruises, collected between 1990 and 1998. Several other small data collections from Germany, Russia, and France were also included. Data quality checks, cruise bias corrections, averaging on isopycnal surfaces, and finally optimal interpolation on 45 standard levels was used to determine the climatologies and error estimates of salinity, temperature, oxygen, and a suite of nutrients. Many more details and figures are available in Gouretski et al. 2004, WOCE Global Hydrographic Climatology.

Hydrography analysis for the North Atlantic ocean including temperature, salinity, oxygen, phosphate, nitrate, and silicate. Based on data from 1981-1985.

Formatted and Unformatted data archive used to create initial- ization grids for the GFDL MOM model.

This archive has three distinct Miami Isopycnic Coordinate Ocean Model (MICOM) outputs. The output grids are for isopycnal layers, at one-twelveth degree spatial and three-day temporal resolution. Water temperature, salinity, and velocity components are the primary output variables. The model domain is the Atlantic Ocean from 65-70N to 28S latitude and in two cases includes the Mediterranean Sea. The three cases are briefly outlined below and differ primarily in the atmospheric forcing that is used.

• MICOM.ICOADS
• International Comprehensive Ocean-Atmosphere Data Set (ICOADS) mean wind forcing
• Domain: 28S to 65N, 98W to 17E
• 16 isopycnal layers
• Six years of data
• MICOM.ECMWF
• ECMWF mean wind forcing
• Domain 28S to 70N, 98W to 36E, includes the Mediterranean Sea
• 20 isopycnal layers
• Three years of data
• MICOM.DAILY
• ECMWF 6-hourly forcing (wind, surface radiation, air temperature and humidity)
• Domain 28S to 70N, 98W to 36E, includes the Mediterranean Sea
• 20 isopycnal layers
• 1979-1986

Details about the model forcing, mixing parameterization, bottom topography, computational requirements, and browse graphics are available at the MICOM website

Shea, Trenberth, and Reynolds (CAC) made improvements to the Reynolds (CAC) SST data set in the regions of the Gulf Stream and Kiroshio Current.

GOSTAplus (Atlas 8) and the Blended Land Surface and SST Atlas were taked from the GOSTAplus CD-ROM received at NCAR February 1998. The period of record for these grids is 1851-1995.

The sea and land surface temperature portions of this data collection, which span 1856 to 1995, have in general been superceded by continued developments of HadISST (DS277.3) , and Jones' Monthly Global Surface Temperature (DS215.0) .

Dimethylsulfide (DMS) measurements (15748 samples) made in the surface global oceans during 1972-1998 are given as raw data samples, along with the associated environmental data data at the time of data collection, and as 1x1 global objectively analyzed grids. The grids are for both a monthly climatology (12 monthly grids), and an overall annual mean. ALL THE DATA AND DOCUMENTATION ARE ONLINE AND AVAILABLE AT NO COST.

A monthly climatology of sea-level pressure, sea-surface temperature, surface temperature, and precipitation on a 2.5-degree global grid for the 30-year period from 1950 to 1979.

Total cloud cover and cloud type amounts over both land and ocean, based on data from FNOC and COADS.

An update is coming. See RELATED SITES below for a preliminary and partial updated maps.

TOTAL CLOUD Edition

EECRA contains surface synoptic weather reports for the entire globe, gathered from various available data sets. The reports were processed, edited, and rewritten to provide a single dataset of individual observations of clouds, spanning the 46 years 1952-1997 for ship data and the 26 years 1971-1996 for land station data. In addition to the cloud portion of the synoptic report, each edited report also includes the associated pressure, present weather, wind, air temperature, and dew point (and sea surface temperature over oceans).

Land data are collected from U.S. Navy SPOT Global Surface and Upper Air Observations for 1971-1976 and from dataset NCEP ADP Global Surface Observations for 1977-1996; ship observations are from dataset ICOADS.

The data may be obtained online from the RELATED SITES.

Krishnamurti sent his analyses to DSS.

This dataset contains 100 days of tropical wind analyses for the 300mb, 250mb, and 200mb levels. These analyses were provided by T.N. Krishnamurti at Florida State University. The u- and v-components of the wind are contained in a 2.5-degree grid covering the tropical globe.

Produced by Krishnamurti for MONEX.

This dataset contains monthly tropical wind analyses at 850mb and 200mb for January 1965 to December 1974. The data are on a 2.5-degree grid covering the global tropics from 25S latitude to 45N latitude.

Produced from satellite and FGGE data for MONEX.

Calculated from ECMWF grids.

Analyses of data collected during the GARP DST III period.

Collected from the European Centre for Medium Range Weather Forecasts for FGGE.

ECMWF reanalyses of the FGGE Special Observation Periods (SOP: 1979Jan10 - 1979Mar05 and 1979May05 - 1979Jul31) based on FGGE IIIb, 1985 version. No plans to do the rest of the FGGE year (as of April 1989). These are the 1986Dec versions of the analyses, and are the initialized grids.

Extracted from the 192x49 1.875-degr grids in DS307.0 by DSS.

ECMWF global analyses collected for ALPEX for the Spring of 1982.

ECMWF tropospheric surface and tendency analyses on a global grid with T63 resolution.

This dataset contains global analyses of tropospheric geopotential height, temperature, winds, relative humidity, and vertical motion from the Geophysical Fluid Dynamics Laboratory (GFDL) for FGGE Special Observation Periods 1 and 2.

This dataset contains global IIIb tropospheric analyses from Goddard Space Flight Center for FGGE Special Observation Periods 1 and 2.

These volumes contain the Ooyama, Esbensen and Chu analysis of upper air soundings taken during Phase III of GATE. The analyzed data has been interpolated to a regular grid using a statistical interpolation technique developed by Ooyama (1987: Mon. Wea. Rev., v.115, pp.2479-2506). The method was applied to the wind data by KV Ooyama and J-H Chu, and to the temperature and relative humidity data by SK Esbensen. A technical report on the temperature and relative humidity analysis is available from SK Esbensen .

Includes initialized and non-initialized analyses and diagnostics.

Derived from NMC and Navy maps.

This dataset contains climate model outputs of various meteorological parameters for use in assessing the impact of increased levels of atmospheric CO2 on the global climate.

This dataset contains model outputs from climate models at the Geophysical Fluid Dynamics Laboratory (GFDL) for use in assessing the impact of increased atmospheric CO2 concentrations on the global climate.

Various climate model outputs from the UK.

This dataset contains outputs from climate models at the Goddard Institute for Space Studies (GISS) for use in assessing the impact of increased atmospheric CO2 concentrations on the global climate.

Three different 100 year runs from German climate models.

The Canadian Centre for Climate Modelling and Analysis (CCCMA) has developed several climate simulation models for use in projecting natural climate change and assessing the impact of human activities on climate change. CCCMA's first-generation global coupled model (CGCM1) was run with three scenarios: one with constant atmospheric forcing (control run), one with carbon dioxide (CO2) increasing at a compounded one percent per year rate, and one with a combination of CO2 and sulfate aerosol concentrations.

The model was run at T32 resolution, resulting in a 97x48 gaussian latitude/longitude grid with approximately 3.75 degree by 3.75 degree resolution. Daily (once- or twice-daily depending on the parameter) and monthly output grids were produced. Upper level (850mb and 500mb) parameters include geopotential heights, temperature, winds, and humidity. At the surface, the model outputs include temperature, wind, humidity, pressure, sea-level pressure, precipitation, and radiation fluxes.

DSS has used the monthly-mean grids to compute decadal means for years zero through nine (e.g. 2000-2009) for each decade. DSS also computed smoothed decadal means which are 30-year running means of the raw decadal means. Because of the large variability in the decadal means of parameters such as precipitation, it is recommended that the smoothed decadal-mean grids be used and applyed to the middle decade of the period (e.g. apply the 2000-2029 mean to the year 2015).

This dataset contains climate model output products from the United Kingdom's Hadley Centre for Climate Prediction and Research. The model output consists of monthly grids from three runs of HCCPR's HadCM2 model. The five runs are a control (constant atmospheric forcing) run, two greenhouse gas runs which assume a constant compounded percentage per year (either 0.5% or 1%) increase in CO₂, and two aerosol runs which combine the CO₂ and sulfate aerosol concentrations. The model has a spatial resolution of 3.75-degree by 2.5-degree (96x73).

DSS has computed decadal mean grids for each decade-month from the monthly mean grids. The decade means are computed for years 0 through 9 of each decade. We also computed smoothed decadal means by computing 30-year running means of the decadal mean grids. Because of the large variability in the decadal means of variables such as precipitation, we recommend using the 30-year (smoothed decadal) means and applying them to the year at the middle of the 30-year period (i.e. - the 30-year mean for 2070-2099 should be applied at year 2085).

This dataset contains climate model output from Deutches Klimarechenzentrum (DKRZ) in Germany. The model output consists of monthly mean grids for several variables from three runs of the ECHAM4 model. The three runs are a control (constant atmospheric forcing) run, a greenhouse gas run which assumes a compounded 1%/year increase in CO₂, and an aerosol run which combines CO₂ and sulphate aerosol concentrations. The model is run at T42 resolution, resulting in a 128x64 latitude/longitude gaussian output grid (approximately 2.8-degree by 2.8-degree resolution). The grids are in GRIB format (Edition 1).

DSS has computed decadal means for each decade-month from the monthly mean grids. The decade means are computed from years 0 through 9 of each decade. We also computed smoothed decadal means by computing 30-year running means of the decadal mean grids. Because of the large variability in the decadal means of variables such as precipitation, we recommend using the 30-year (smoothed decadal) means and applying them to the year at the middle of the 30-year period (i.e. - the 30-year mean for 2070-2099 should be applied at year 2085).

A special set of fields for the TOGA COARE region during 15 Nov 1992-15 Feb 1993. Fields consist of initialized analyses and short range forecasts from reruns of the T106L31 ECMWF model. Use of this data is restricted to COARE scientists only. A permission form must be signed and returned to DSS for use of this data.

This free dataset, obtained via the Unidata Internet Data Destribution System (IDD), contains gridded analyses and forecasts from US National Centers for Environmetal Prediction (NCEP) and European Centre for Medium-Range Weather Forecasts (ECMWF) models. Models include NCEP ETA/NAM and RUC covering the Continental US, NCEP Ensemble and GFS covering North America, and GFS Extended and ECMWF covering the globe, at various spatial and temporal resolutions. Potential variables found in the model output include pressure, relative humidity, temperature, geopotential height, U-component wind speed, and V-component wind speed. Model specific information can be found in the ds335.0 documentation page.

The most current three months of model output are available online through the ds335.0 data page. Online archives are populated daily with model output generated two days prior to the current date. Older historical files, beginning in December 2002, are available upon request. All model output files are stored in their native GRIB1 format.

More information is available.

This free dataset, derived from real-time Global Telecommunications System (GTS) reports distributed via the Unidata Internet Data Distribution System (IDD), includes surface station (SYNOP) reports at 3-hour intervals, upper air (RAOB) reports at 3-hour intervals, surface station (METAR) reports at 1-hour intervals, and marine surface (BUOY) reports at 1-hour intervals. Select variables found in all report types include pressure, temperature, wind speed, and wind direction. Report specific descriptions and variable types can be found in the ds336.0 documentation page.

The most current three months of data are available online through the ds336.0 data page. Online archives are populated daily with reports generated two days prior to the current date. Older historical files, begining in May 2003, are available upon request.

All reports are encoded from the raw GTS datastream format into NetCDF format without quality control performed. More complete, quality controlled archives of surface observations can be found in ds464.0, and ds461.0. More complete, quality contolled archives of upper air observations can be found in ds353.4, and ds351.0.

More information is available.

DSS presents a global synoptic set of 6 hourly upper air reports operationally collected by the National Centers for Environmental Prediction (NCEP). These include radiosondes, pibals and aircraft reports from the Global Telecommunications System (GTS) and satellite data from the National Environmental Satellite, Data, and Information Service (NESDIS). The reports can include pressure, geopotential height, temperature, dewpoint depression, wind direction and speed. This data is the primary input to the Global Data Assimilation System (GDAS), which is used to make forecasts and the Final Analyses (FNL). DS353.4 has been derived from this dataset since April 2000.

The ADPUPA data includes land and ship launched radiosondes and pibals. This involves, at 00Z and 12Z, about 650 - 1000 stations, and at 06Z and 18Z (which are mostly pibals), about 150 - 400 stations. The counts have declined in recent years. Data may be available at up to 20 mandatory levels from 1000mb to 1mb, plus a few significant levels. The AIRCFT data includes commercial, some military and reconnaissance reports. They are flight level reports, numbering about 400 at 12 hourly intervals to, more recently, about 20000 at 6 hourly intervals. The SATWND data includes winds derived from cloud drift analysis. The AIRCAR data includes data from aircraft takeoff and landings, with between 5000 and 10000 reports every 6 hours.

This dataset is maintained in NCEP BUFR format, and updated monthly by DSS using weekly files prepared by NCEP. Software which extracts data from the BUFR file, and places it into a basic text file can be found in BUFRdecode_ADPuprair.tar. Data archived within this dataset spans the period from April 2000 to near current. DS353.4 provides the same data from 1973 to near current in NMC Office Note 29 (ON29) format. BUFR format reports contain additional information that is not carried over to ON29 format, including the the original raw GTS string reports.

NCEP does some operational Quality Control (QC) on the data which sets some QC flags in the reports to qualify, but not quantify any corrections. After the data is used to prepare the FNL analyses, no additional corrections are made. Older data is not reprocessed with newer procedures.

More information is available.

DSS presents a regional synoptic set of 6 hourly upper air reports centered over North America, operationally collected by the National Centers for Environmental Prediction (NCEP). These include radiosondes, pibals and aircraft reports from the Global Telecommunications System (GTS) and satellite data from the National Environmental Satellite, Data, and Information Service (NESDIS). The reports can include pressure, geopotential height, temperature, dewpoint depression, wind direction and speed. This data is the primary input to the ETA Data Assimilation System (EDAS)/NAM Data Assimilation System (NDAS starting Jan. 23, 2005), which is used to make forecasts and the Final Analyses (FNL). DS351.0 provides global data coverage over the same time period.

The ADPUPA data includes land and ship launched radiosondes and pibals launched at 6 hourly intervals. Data may be available at up to 20 mandatory levels from 1000mb to 1mb, plus a few significant levels. The AIRCFT data includes commercial, some military and reconnaissance reports. They are flight level reports at 6 hourly intervals. The SATWND data includes winds derived from cloud drift analysis. The AIRCAR data includes data from aircraft takeoffs and landings at 6 hourly intervals

This dataset is maintained in NCEP BUFR format, and updated monthly by DSS using weekly files prepared by NCEP. Software which extracts data from the BUFR file, and places it into a basic text file can be found in BUFRdecode_ADPuprair.tar. Data archived within this dataset spans the period from October 2000 to near current. DS353.4 provides global data from 1973 to near current in NMC Office Note 29 (ON29) format. BUFR format reports contain additional information that is not carried over to ON29 format, including the the original raw GTS string reports.

NCEP does some operational Quality Control (QC) on the data which sets some QC flags in the reports to qualify, but not quantify any corrections. After the data is used to prepare the FNL analyses, no additional corrections are made. Older data is not reprocessed with newer procedures.

More information is available.

DSS presents a global synoptic set of 12 hourly upper air data reports. These were operationally collected by NMC. They include radiosondes, pibals and aircraft reports received via the Global Telecommunications System (GTS). NMC used Automated Data Processing (ADP) to decode and reformat the GTS reports. This data was the primary input (known as "B-3s") for NMC's modeling and forecasting. It was also a major input for the NCEP/NCAR and ECMWF Reanalysis Projects. The NCAR document TN404 An Introduction to Atmospheric and Oceanographic Datasets, (chapter 3) discusses the collection of this data in greater detail.

The ADPUPA data includes land and ship launched radiosondes and pibals. Data may be available at up to 20 mandatory levels from 1000mb to 10mb, plus a few significant levels. The AIRCFT data includes flight level reports from commercial, military and reconnaissance sources. The data may include pressure, d-value (deviation from standard atmosphere geopotential height), temperature, dewpoint, wind direction and speed.

This dataset is in an NCAR blocked version of the NMC Office Note 20 (ON20) format, which involves using the DSS RPTOUT software. For data after 1972 see DS353.4. More information is available.

DSS presents a global synoptic set of 6 hourly upper air data reports. These were operationally collected by NCEP. They include radiosondes, pibals and aircraft reports received via the Global Telecommunications System (GTS) and satellite data from the National Environmental Satellite, Data, and Information Service (NESDIS). This data is the primary input to the Global Data Assimilation System (GDAS), which is used to make forecasts and the Global Final Analyses (FNL). It was also a major input for the NCEP/NCAR and ECMWF Reanalysis Projects. NCEP uses Automated Data Processing (ADP), to decode and reformat the GTS reports. The NCEP document Observational Data Processing at NCEP and the NCAR document TN404 An Introduction to Atmospheric and Oceanographic Datasets, (chapter 3) discusses the collection of this data in greater detail.

The ADPUPA data includes land and ship launched radiosondes and pibals. This involves, at 00Z and 12Z, about 900 - 1000 stations, and at 06Z and 18Z (which are mostly pibals), about 300 - 400 stations. Data may be available at up to 20 mandatory levels from 1000mb to 1mb, plus a few significant levels. The AIRCFT data includes flight level reports from commercial, military and reconnaissance sources. There are about 1200 every 6 hours. The SIRSOB data includes infrared sounding data (discontinued in early 1978) The SATWND data includes winds derived from cloud drift analysis. The AIRCAR data includes data from aircraft takeoff and landings (beginning mid 1991) The reports may include pressure, geopotential height, temperature, dewpoint depression, wind direction and speed.

This dataset is in an ASCII character format, known as the NMC Office Note 29 (ON29) format. DSS has collected for users a variety of important usage details.

Extracted from other data sources, in support of GATE.

DSS obtained this data and/or extracted it from other datasets for the Global Atmospheric Research Project.

DSS presents a global synoptic set of 6 hourly upper air data reports. These were operationally collected by NCEP. They include radiosondes, pibals and aircraft reports received via the Global Telecommunications System (GTS) and satellite data from the National Environmental Satellite, Data, and Information Service (NESDIS). This data is the primary input to the Global Data Assimilation System (GDAS), which is used to make forecasts and the Global Final Analyses (FNL). It was also a major input for the NCEP/NCAR and ECMWF Reanalysis Projects. NCEP uses Automated Data Processing (ADP), to decode and reformat the GTS reports. The NCEP document Observational Data Processing at NCEP and the NCAR document TN404 An Introduction to Atmospheric and Oceanographic Datasets, (chapter 3) discuss the collection of this data in greater detail.

The ADPUPA data includes upper air station data from land and ship launched radiosondes and pibals. This involves, at 00Z and 12Z, about 650 - 1000 stations, and at 06Z and 18Z (which are mostly pibals), about 150 - 400 stations. The counts have declined in recent years. Data may be available at up to 20 mandatory levels from 1000mb to 1mb, plus a few significant levels. The AIRCFT data includes aircraft flight level reports from commercial, military and reconnaissance sources. There are about 400 every 12 hours in the oldest files, to about 20000 every 6 hours in more recent files. The SIRSOB data includes satellite infrared sounding observation data (discontinued in early 1979) The SATWND data includes satellite winds derived from cloud drift analysis. The AIRCAR data includes data from aircraft takeoff and landings (beginning mid 1991), with between 5000 and 10000 reports every 6 hours. The reports may include pressure, geopotential height, temperature, dewpoint depression, wind direction and speed.

This dataset is in an ASCII character format, known as the NMC Office Note 29 (ON29) format. It is updated monthly by converting the weekly BUFR files prepared for DSS by NCEP. The BUFR files are kept in DS351.0. The most current 12 months of data are available online through the DS353.4 data page. For data prior to 1973 see DS353.0. DSS has collected for users a variety of important usage details.

DSS obtained this data, after errors were corrected in the original data.

Global upper air observations from the U.S.A.F. for the period from July 1966 to August 1969.

Dropsonde observations were made by the U.S. Air Force along the "Ptarmigan" track of its reconaissance program. This track was mainly over the Beaufort Sea and the western Arctic Ocean. Vertical profiles of geopotential height, temperature, and dewpoint temperature at various pressure levels are available for eleven years between 1950 and 1961 (excluding 1953) for the lower troposphere mainly below 500mb.

In comparing the locations on the dropsondes to the locations on the aircraft reconaissance reports in ds360.0, DSS found that the dropsonde longitudes are off by 10 degrees. For dropsondes over the Western Hemisphere, the positions are 10 degrees east of the aircraft positions, and for the Eastern Hemisphere, they are 10 degrees west of the aircraft position.

The NOAA Profiler Network (NPN) has 35 unmanned Doppler Radar sites in 18 central US states and Alaska. They provide hourly vertical profiles of wind data. The hourly netCDF files disseminated from Forecast Systems Laboratory (FSL) of NOAA starting from May 1992 are archived in this dataset. It continues to be updated monthly.

U and V wind components from 500m above ground level to 16.25km are recorded at every 250m increment. Several stations may have been moved or added. Problems of bird contamination are also reported. Flags are used to indicate the quality of data.

Detailed description of NPN and recent data (of last 14 days) are available at FSL's NPN site.

This dataset contains vertical wind profiles for several tropical Pacific islands and TOGA-COARE ships. All data were provided by NOAA's Aeronomy Lab, which operates several of the wind profiling radars. Other contributors include the Australian Bureau of Meteorology, which owns the Darwin radar, and TOGA-COARE which provided data from ships and some land stations.

Data are available from as early as August 1985, and the Galapagos radar is still currently in operation. Please consult the inventory for full period-of-record information for each individual station.

Two periods (1947Jan-1959Jan and 1976Jan-1985Dec) of global aircraft observations from U.S.A.F. ETAC decode and obtained by DSS.

Aircraft reports from Australia's real-time archive

This dataset contains aircraft reports from New Zealand. Each report contains date, aircraft ID, location, temperature, and wind information.

Global aircraft observations containing temperature and wind measurements for the period from January 1960 to December 1973. Obtained from James Sadler at the University of Hawaii.

DSS presents a comprehensive set of global, 6 hourly upper air radiosonde observation (RAOB) feedback record reports from RAOBs used in the ECMWF 45-year reanalysis (ERA-40) covering the period from September 1957 to August 2002. RAOBs, from multiple data sources, were assimilated into an ECMWF model and produced ERA-40. The metadata generated during the ERA-40 quality control (QC)/data assimilation process were appended to the input observations. The final combination of input observations and feedback metadata are called feedback record reports.

Feedback record reports include a set of input observation (obs) variables, analyzed obs variables, and associated metadata information. Input obs variables include pressure, vertical sounding significance, geopotential, temperature, dewpoint temperature, wind speed, and wind direction. Analyzed obs variables are derived from input obs variables, and were used in the model assimilation. Analyzed obs variables include temperature, pressure, height, relative humidity, specific humidity, U-wind, and V-wind speed components. The feedback metadata available includes QC flags for entire reports and individual obs variables, and differences between individual obs variables and interpolated six-hour forecasts and final model analyses. Further, in the case of QC report or obs variable rejection, some details why rejection occurred are given in the codes provided.

This global set of RAOB feedback record reports includes land lauched radiosondes that fall within the ERA-40 "temp" data category. Reports are derived from individual input datasets. Data may be available at mandatory or significant levels from 1000mb to 1mb. Higher vertical resolution is also possible.

These data are available only for non-commercial research in the U.S. and some locations in Canada. User requirements and a registration form can be accessed here. This dataset is maintained in a DSS ASCII format. Weekly ASCII files are available online to authorized users through the ds366.0 data page. Authorized users can also access the archive via the NCAR Mass Store System. Software which extracts data from the ASCII file and places it into memory arrays for further processing can be found in the software directory. Primary ECMWF data products including forecast and analysis fields can be found under the ERA-40 dataset home page.

DSS presents a comprehensive set of global, upper air satellite wind observation (SATWND) feedback record reports from SATWNDs used in the ECMWF 45-year reanalysis (ERA-40) covering the period from September 1957 to August 2002. SATWNDs, from multiple data sources, were assimilated into an ECMWF model and produced ERA-40. The metadata generated during the ERA-40 quality control (QC)/data assimilation process were appended to the input observations. The final combination of input observations and feedback metadata are called feedback record reports.

Feedback record reports include a set of input observation (obs) variables, analyzed obs variables, and associated metadata information. Input obs variables include pressure, wind speed, and wind direction. Analyzed obs variables are derived from input obs variables, and were used in the model assimilation. Analyzed obs variables include U-wind, and V-wind speed components at the observed pressure level. The feedback metadata available includes QC flags for entire reports and individual obs variables, and differences between individual obs variables and interpolated six-hour forecasts and final model analyses. Further, in the case of QC report or obs variable rejection, some details why rejection occurred are given in the codes provided.

This global set of SATWND feedback record reports includes geostationary satellite wind observations. Reports are derived from individual input datasets. Data are available on single pressure levels, with one pressure level per report.

These data are available only for non-commercial research in the U.S. and some locations in Canada. User requirements and a registration form can be accessed here. This dataset is maintained in a DSS ASCII format. Weekly ASCII files are available online to authorized users through the ds367.0 data page. Authorized users can also access the archive via the NCAR Mass Store System. Software which extracts data from the ASCII file and places it into memory arrays for further processing can be found in the software directory. Primary ECMWF data products including forecast and analysis fields can be found under the ERA-40 dataset home page.

This dataset contains flight-level aircraft measurements from five GASP aircraft during the period of March 1975 to July 1979. Parameters measured by the aircraft include various meteorological variables, ozone, carbon monoxide and particle densities, and flight information. The data records are in ASCII format, and are a constant length of 512 bytes.

Aircraft observations from the ALPine EXperiment.

Radiosonde observations, by four ships, from the Barbados Oceanogaphic and Meteorological EXperment, collected by Florida State University.

Data collected during the GARP Atlantic Tropical Experiment.

Data collected during the GARP Atlantic Tropical Experiment.

This dataset contains data from Portable Automated Mesonets from various meteorological experiments and for various periods between April 1977 and November 1982.

Sounding data collected from various areas of the world including Argentina, Australia, Brazil, Canada, France, India, Japan, New Zealand, Russia, Singapore, South Africa, and United Kingdom.

U.S. controlled station time series of upper air observations (soundings) comprise the data in this set. From the National Climatic Data Center (NCDC), DSS obtains annual updates of the DSI-6301 UPPER AIR DIGITAL FILES, which are then quality checked, reformatted, hydrostatically checked, and merged into existing station time series.

Primarily, stations are located in the contiguous U.S., Alaska, the Caribbean, Hawaii and other Pacific Islands, plus Keflavik, Ascension Island, Diego Garcia, and the South Pole. There are currently about 110 stations received.

Mandatory, standard, and significant levels may include pressure, height, temperature, moisture, wind direction and speed. Earliest data is for 1945, but periods of record vary widely among the stations.

Combined summarized coverage of time series rawinsondes in ds390.0 and ds390.1.

Daily radiosonde observations from the U.S.A.F. for the years from 1948 to 1965.

Collection of time series raobs for 1943-1974, for selected areas of the world, produced and maintained by USAFETAC, Asheville, NC. Entire set was sent to GFDL for intermediate processing, then returned to NCAR for final reformatting.

Data from USAF at NCDC in DATSAV format.

A collection of monthly time series upper air soundings, derived by DSS from the collections of daily soundings in ds390.0 and ds390.1.

This dataset contains monthly profiles of temperature and zonal winds at stratospheric levels between 70mb and 10mb for the period from January 1953 to December 1988.

Argentina has provided upper air soundings for twenty stations for the general period 1958-1991, but the actual period varies by station.

NCDC has provided rawinsonde data for eleven Brazilian stations and pibal data for thirty stations for the general period January 1950 to November 1986. The actual period, however, varies widely by individual station. The data were recorded by the Brazilian Air Force.

Daily world radiosonde observations collected by the Massachusetts Institute of Technology.

This dataset contains radiosonde observations from the Ascension Islands for the period from January 1968 to March 1977.

This data set contains daily rawinsonde data for San Cristobal Island (Galapagos) for the general period of December 1990 to December 1998 with some missing months. The data were keyed from the original forms as part of the Climate Modernization Database Program at NCDC and sent to DSS for post-processing and quality control.

Generally, soundings are available once-daily, but there are occasional days with two soundings. The regular soundings (mandatory and significant levels) will be converted to the DSS Time-Series Raob format and run through a hydrostatic check program, and they should be available by the end of 2006.

Also available by special request are high-resolution soundings in the original comma-delimited keying project format. The high-resolution soundings contain data at 10-second intervals up to five minutes from release time, and thereafter, data are recorded every thirty seconds. In the most recent years, high-resolution data are available at 10-second intervals through the duration of the sounding.

This dataset contains upper air reports from six stations in the U.S.S.R.: Olyonok, Kustanay, Petropavlosk-Kamchatskiy, Omsk, Alma-ata, and Chita, for the period from January 1961 to December 1970.

This version of the Russian Ice Island raobs was provided by the Russian delegation to meetings at NCDC in November 1999.

The Japanese Meteorological Agency has provided upper-air and surface observations for Japanese stations. Daily upper-air data are available at 6-hourly and 12-hourly intervals for the years from 1981 to 1997. Monthly upper-air data are available back to 1961. Daily surface observations at 3-hourly intervals span the period from 1961 to 1989, and the period from 1990 to 1997 is covered by hourly surface data.

Upper air data from the Monthly Climatic Data for the World, obtained from the National Climatic Data Center (NCDC).

Monthly mean sounding data at 00Z and 12Z for various stations in the global radiosonde network for May 1958 to December 1989.

Monthly mean rawinsonde data (and metadata) created by Jack Woollen from the daily soundings input to the NCEP/NCAR Global Reanalysis.

Global surface observations from the U.S.A.F. for February 1967 to April 1967 and for August 1976.

DSS presents a global synoptic set of hourly surface data reports. These were operationally collected by the National Centers for Environmental Prediction (NCEP). The data includes land and marine reports received via the Global Telecommunications System (GTS). Variables recorded in the reports include pressure, temperature, dewpoint depression, wind direction and speed. Precipitation data has been decoded for the U.S. and Canada. A similar dataset, DS464.0, has been derived from this dataset since April 2000.

The ADPSFC data includes SYNOP, METAR, AWOS, and ASOS land report types. The SFCSHP data includes moving ship, fixed ship, MARS (moving and fixed) and buoy (moored and drifting) marine report types. Reports number in the thousands.

This dataset is maintained in NCEP BUFR format, and updated monthly by DSS using weekly files prepared by NCEP. Software which extracts data from the BUFR file, and places it into a basic text file, can be found in BUFRdecode_ADPsfc.tar . Data archived within this dataset spans the period from April 2000 to near current. DS464.0 provides the same data from 1973 to near current in NMC Office Note 29 (ON29) format. BUFR format reports contain additional information that is not carried over to ON29 format, including the the original raw GTS string reports. Similar data produced at the National Climatic Data Center (NCDC) from a USAF binary set (DATSAV2) is available in ds463.2, covering the period from 1901 to near current.

More information is available.

DSS presents a regional synoptic set of hourly surface data reports centered over North America. These were operationally collected by the National Centers for Environmental Prediction (NCEP). The data includes land and marine reports received via the Global Telecommunications System (GTS). Variables recorded in the reports include pressure, temperature, dewpoint depression, wind direction and speed. Precipitation data has been decoded for the U.S. and Canada. DS461.0 provides global surface data coverage.

The ADPSFC data includes SYNOP, METAR, AWOS, and ASOS land report types. The SFCSHP data includes moving ship, fixed ship, MARS (moving and fixed) and buoy (moored and drifting) marine report types. Reports number in the thousands.

This dataset is maintained in NCEP BUFR format, and updated monthly by DSS using weekly files prepared by NCEP. Software which extracts data from the BUFR file, and places it into a basic text file, can be found in BUFRdecode_ADPsfc.tar . Data archived within this dataset spans the period from October 2000 to near current. DS464.0 provides global data from 1973 to near current in NMC Office Note 29 (ON29) format. BUFR format reports contain additional information that is not carried over to ON29 format, including the the original raw GTS string reports. Similar data produced at the National Climatic Data Center (NCDC) from a USAF binary set (DATSAV2) is available in ds463.2, covering the period from 1901 to near current.

More information is available.

Sample of data collected by the U.S. Air Force Global Weather Central (AFGWC)

Samples of data collected by the U.S. Air Force Global Weather Central (AFGWC).

DSS acquired this data for Kreitzberg's work in the Severe Storms in American Midwest Experiment. Data exist only for a few days across the period from April to September 1977.

Global surface observations for many stations in modified SYNOP format, primarily derived from USAF(ETAC) DATSAV format by Dick Davis at NCDC.

Global surface observations from USAF(ETAC) at NCDC in DATSAV3 format. This set is superceded by ds463.2.

This ASCII data set (DATSAV3) is produced at the National Climatic Data Center (NCDC) from a USAF binary set (DATSAV2), a collection of worldwide surface weather observations from sources such as the the Global Telecommunications System (GTS) and Automated Weather Network (AWN). The Air Force Weather Agency (AFWA) decoded most of the collected observations for the Air Force Combat Climatology Center (AFCCC) where final decoding, validation, and quality control created DATSAV2.

Manadatory data includes pressure, temperature, visibility, sky condition, and wind observation, but there may also be additional data, data remarks, or data element quality sections.

NCDC provides quarterly updates to DSS, usually with a two-month delay. The earliest data is for 1901. The earliest years have the least volume of data, a few hundred megabytes; the volume in the 1950s-early 1960s increased by a factor of 10-20; 1965-1972 shows a decrease because of a loss of data; beginning in 1973 with about 5 gigabytes of data, volume has gradually increased to about 20 gigabytes per year. There are now about 10,000 active stations.

Similar data sets prepared by the National Centers for Environmental Prediction (NCEP) are ds464.0 (beginning February 1975), and ds461.0 (beginning April 2000).

This worldwide collection of surface weather observations is produced at the National Climatic Data Center (NCDC) from NCDC Surface Airway Hourly Observations (TD3280), U.S.A.F. DATSAV3 Surface Hourly Observations (TD9956), and NCDC Cooperative Hourly Precipitation (TD3240). Observations are all in the same ASCII format.

Manadatory data includes pressure, temperature, dew-point, visibility, sky condition, and wind observation, but there may also be additional data, data remarks, or data element quality sections.

The earliest data is for 1901. The earliest years have the least volume of data, a few hundred megabytes; the volume in the 1950s-early 1960s increased by a factor of 10-20; 1965-1972 shows a decrease because of a loss of data; beginning in 1973 with about 5 gigabytes of data, volume has gradually increased to about 20 gigabytes per year. There are now about 10,000 active stations. For frequent updates to recent data and occasional updates to the historical data, access the data directly from NCDC.

Similar data sets prepared by the National Centers for Environmental Prediction (NCEP) are ds464.0 (beginning February 1975), and ds461.0 (beginning April 2000).

Requests for this data from non-NCAR users will be forwarded to NCDC.

DSS presents a global synoptic set of 3 hourly surface data reports. These were operationally collected by NCEP. They include land and marine reports received via the Global Telecommunications System (GTS). This dataset is DSS' primary surface observation set. NCEP uses Automated Data Processing (ADP) to decode and reformat the GTS reports. The NCEP document Observational Data Processing at NCEP and the NCAR document TN404 An Introduction to Atmospheric and Oceanographic Datasets, (chapter 3) discuss the collection of this data in greater detail.

The ADPSFC data includes these land surface station report types: SYNOP, METAR, and beginning in the mid 1990s, AWOS and ASOS. The SFCSHP data includes these marine surface report types: moving ship, fixed ship, MARS (moving and fixed) and buoy (moored and drifting). There are thousands of reports every six hours. An enumeration of this, available in our inventory summaries, shows how the volume has increased over the years. Similar information since 1997 is not available, but the counts have increased considerably. This is because NCEP began collecting hourly reports and in limited regions, 20-minute reports, in the late 1990s. The reports may include pressure, temperature, dewpoint depression, wind direction and speed. Precipitation data has been decoded for the U.S. and Canada. Prior to 2000, cloud type data was available for the world, except for the U.S., which discontinued much of the cloud data in 1995.

This dataset is in an ASCII character format, known as the NMC Office Note 29 (ON29) format. It is updated monthly by converting the weekly BUFR files prepared for DSS by NCEP. The BUFR files are kept in DS461.0. The most current 12 months of data are available online through the DS464.0 data page. DSS has collected for users a variety of important usage details.

Intended for area subsets extracted from ds464.0 by DSS. Presently just for Greenland, 1978-1999May.

The Navy Fleet Numerical Weather Center (FNWC) collects, on an operational basis, surface and upper air data from the Global Telecommunications System (GTS). DSS created the subsets. Coverage and quality inferior to NCEP/NMC ADP - we recommend using DS464.0 and DS353.4.

This dataset contains monthly precipitation values on a 2.5-degree grid covering the tropical Pacific Ocean from 30S latitude to 30N and from 105E longitude to 290E. Grids are available from January 1965 to December 1972.

This dataset contains monthly and daily precipitation data from the Pacific Rainfall Data Base (PACRAIN). Monthly data begin in January 1874 and continue through June 2004. Daily station data are available beginning in 1971.

Global synoptic surface observations from the Air Weather Service for 1901-1973, called "Tape Deck 13" because it is the combination of multiple card decks (datasets), conversion to a standard format and transfer to tape. There are usually 1 to 4 records per station each day.

Wind direction (dd), wind speed (ff), and station elevation (stnelev) are "standard" variables usually available in each deck; the rest of the variables are "supplemental fields" which are not in some decks (groups of stations). Sea level pressure (slp) "standard" format excludes the high order digits, retaining only tens, units and tenths mb; however, some decks provide the hundreds/thousands digit in the "supplemental fields".

This dataset contains daily surface observations from ninety-five Canadian stations from January 1963 to December 1976.

This data set, updated annually from the National Climatic Data Center (NCDC), contains hourly or 3-hourly surface weather observations, primarily measured at major airports and military bases. The stations are operated by the National Weather Service (NWS), the Federal Aviation Administration (FAA), the U.S. Air Force (Air Weather Service), and the U.S. Navy (Navy Weather Detachment). The NWS and FAA sites are located in the contiguous U.S., Alaska, Puerto Rico, Hawaii and other Pacific Islands. The Air Force and Navy operate stations worldwide; however, only a few are included in this set.

The observations are generally recorded midnight to midnight for the 24-hour period. The major data parameters are clouds, visibility, wind, temperature, sky cover, relative humidity, pressure, and descriptive weather. The earliest data is for 1938 but most stations begin in 1948. There are about 300 stations.

TD14 refers to the NCDC format of the original data up through 1983, which has been converted into a packed binary format; TD3280 refers to the ASCII format used at NCDC beginning with the 1984 data.

A homogeneous time series of wind speed at the 10-meter height above the ground/surface for the entire period of digital record (usually, since 1948 for the First Order Stations, since early 1970s for other sites with complete metadata, and since the ASOS implementation at the sites without metadata). Includes elevation homogenization of the near-surface wind time series based on anemometer elevations changes for all stations.

Hourly airways data over US, Canada for about 1,000 stations. There are also some observations for Mexico and most Central American countries.

This dataset contains daily surface observations from polar automatic weather stations on Antarctica from January 1980.

This dataset contains daily surface observations from polar automatic weather stations on Greenland for January 1987 to October 1991.

This dataset consists of 31 Russian north polar drifting stations which took observations for the periods 1937-1938 and 1950-1991. We received the latest version of this data from the Arctic and Antarctic Research Institute (AARI) via the National Snow and Ice Data Center (NSIDC).

This dataset contains daily 6-hourly and 3-hourly surface observations for many U.S.S.R. stations for January 1936 to December 1983.

DSS presents a comprehensive set of global, 3 hourly surface observation feedback record reports from synop land observations used in the ECMWF 45-year reanalysis (ERA-40) covering the period from September 1957 to August 2002. Synop land observations, from multiple data sources, were assimilated into an ECMWF model and produced ERA-40. The metadata generated during the ERA-40 quality control (QC)/data assimilation process were appended to the input observations. The final combination of input observations and feedback metadata are called feedback record reports.

Feedback record reports include a set of input observation (obs) variables, analyzed obs variables, and associated metadata information. Input obs variables include pressure, mean sea level pressure, three hour pressure change, characteristic of pressure tendency, temperature, dewpoint temperature, relative humidity, wind speed, and wind direction. Analyzed obs variables are derived from input obs variables, and were used in the model assimilation. Analyzed obs variables include pressure, height, relative humidity, U-wind, and V-wind speed components. The feedback metadata available includes QC flags for entire reports and individual obs variables, and differences between individual obs variables and interpolated six-hour forecasts and final model analyses. Further, in the case of QC report or obs variable rejection, some details why rejection occurred are given in the codes provided.

This global set of surface obs feedback record reports includes land surface observations that fall within the ERA-40 "synop land" data category. Reports are derived from individual input datasets.

These data are available only for non-commercial research in the U.S. and some locations in Canada. User requirements and a registration form can be accessed here. This dataset is maintained in a DSS ASCII format. Weekly ASCII files are available online to authorized users through the ds476.0 data page. Authorized users can also access the archive via the NCAR Mass Store System. Software which extracts data from the ASCII file and places it into memory arrays for further processing can be found in the software directory. Primary ECMWF data products including forecast and analysis fields can be found under the ERA-40 dataset home page.

Daily station rainfall data for India. There were 1632 stations in 1901, 2536 in 1970. Note that 1473 stations have 67 or more years of data. Total daily observations 76.4 million.

This dataset contains daily precipitation data for many Indian stations from January 1975 to December 1984.

India weekly precipitation, 1979 through 1985, for 34 mainland divisions and an island, have been digitized by Eric Smith, Florida State University, from India's "Weekly Weather Report", provided by D. Sikka. This is based on the stations that report operationally (probably 500-1000 stations). More stations are available in delayed time. The division precipitation for a week is the average of stations that did report during that week. The weeks are continuous. The division normals for the period 1901-1970 are included in the data.

Data is in station time series sort.

This dataset consists of daily and monthly precipitation data for 191 stations in Australia for January 1840 to December 1990. These stations are a subset of the stations in ds482.0 and were chosen because they are believed to be the most reliable for studying rainfall trends in Australia.

Each complete data record contains at least a monthly precipitation total, and most records contain daily values as well. Some records do have missing data.

Malaysia, Thailand and Indonesia surface observations for 1951 to 1985.

The Pacific Rainfall Database (PACRAIN) dataset contains daily and monthly precipitation from stations in the tropical Pacific Ocean. The most up to date versions of this data can be downloaded direclty from the PACRAIN web site

Daily station precipitation and monthly soil temperature from China, 180 stations. Prepared by China under exchange agreements.

This set contains hourly surface data from several Brazilian stations. Reported variables include temperature and moisture information, visibility, present weather, cloud and wind information, pressures, and precipitation.

This dataset, provided by the Meteorological Service of Canada, contains a subset of their hourly station data archive.

"This data set contains precipitation data originally recorded in log books at 65 coastal and island meteorological stations, and later digitized at the Arctic and Antarctic Research Institute (AARI), St. Petersburg, Russia, under the direction of Vladimir Radionov. Records from most stations begin in 1940. Instrumentation was generally a rain gauge with Nipher shield until the early 1950s (for most stations), when the Tretyakov precipitation gauge replaced earlier instrumentation. Data have not been adjusted for gauge type or wind bias. Data from 1967 and later are corrected for wetting loss (this correction was made by observers as they recorded the station data).", quoted from NSIDC site.

The data files and associated documentation were downloaded to this DSS site in January 2007. Please properly cite this data by referring to the 'DS DOCUMENTATION' page.

"This data set contains monthly means of meteorological observation data from Russian stations from 1961-2000 (for most stations). The Russian station observations were provided by Vladimir Radionov, Arctic and Antarctic Research Institute (AARI), St. Petersburg, and include 2-meter air temperature, sea level pressure, total and low cloud amount, precipitation, and relative humidity. Data are available in ASCII text format via FTP.

NSIDC strongly encourages you to register as a user of this data product. As a registered user, you will be notified of updates and corrections.", quoted from NSIDC site.

The data files and associated documentation were downloaded to this DSS site in January 2007. Please properly cite this data by referring to the 'DS DOCUMENTATION' page.

Hourly barometric measurements at the Dutch Royal Magnetic and Meteorological Observatory at Batavia (Djakarta), keypunched in a project by Kevin Hamilton (McGill University) and Rolando Garcia (NCAR) in 1986 from the Observatory Yearbooks published during the period of record.

This dataset contains hourly surface observations from the Denver, Colorado and Buckley Air Force Base, Colorado observing stations. These data were provided to DSS by the U.S.A.F.

The Forecast Systems Laboratory (FSL) (formerly PROFS) had operated a 22-station experimental surface mesoscale meteorological network (Mesonet) along the front range of Colorado from May 1985 to Jun 1994.

Daily temperature and precipitation data have been assembled from Boulder newspaper weather pages and other available sources to provide a history of Boulder, CO climate from January 1897 to December 1977.

This dataset contains wind speed, direction, and gust data for Boulder, Colorado during strong downsloping wind events in 1969.

Users may want to check out more recent similar sets such as - http://rredc.nrel.gov/solar/old_data/nsrdb/tmy2/

This dataset contains daily surface data which represent the "typical meteorological year" (TMY) for stations in nine regions across the United States. Also in this dataset are solar radiation data which represent the TMY for 26 SOLMET stations across the U.S.

The hourly precipitation observations contained in this set are obtained as months become available from the National Climatic Data Center (NCDC). The reports are taken by observers at principle (primary) stations, secondary stations, and cooperative observer stations operated by the National Weather Service (NWS) and the Federal Aviation Agency (FAA). The sites are located in the contiguous U.S., Alaska, Puerto Rico, Virgin Islands, Hawaii and other Pacific Islands.

While most stations begin observations in 1948, several have observations from the early 1940s. The earliest data is for 1936. There are currently about 2800 stations reporting. Periods of record vary widely among the stations.

The 15-minute precipitation observations contained in this set are obtained as months become available from the National Climatic Data Center (NCDC). The sites are located in the contiguous U.S., Alaska, Puerto Rico, Virgin Islands, Hawaii and other Pacific Islands.

The digital file began in May 1971, but there is a very small amount of data in 1970. There are currently about 2300 stations reporting. Periods of record vary widely among the stations.

This data set contains U.S. station surface observations that were digitized from the original forms by the Climate Data Modernization Program (CDMP).

Data are available for more than 200 stations (mainly at airports, but also some city weather bureau offices) that made observations at hourly intervals at least during the daytime hours and often over the full 24-hour day. The general period of record for these stations is 1928-1948, though this varies by individual station. To find out what is available, see this inventory. A significant effort was made by DSS to correct errors in the digitized data, especially dates, times, and pressures. For more information about this work, see this document.

We have also received data for more than 130 U.S. stations that were digitized from Form 1001. These stations, which were usually city weather bureau offices, generally took observations once, twice, or four times daily. Some stations have data back as far as late 1892. An inventory can be viewed which shows stations and their date range coverage. These data will be made available to the community when errors in the digitized data have been corrected.

Hourly and daily precipitation analyses for 1963 to 1994 on a 2.5-degree by 2-degree grid over North America.

Daily Tmin, Tmax, Precip data for 1881-1985, interpolated to fill missing intervals. Prepared by the Oklahoma Climatological Survey under the direction of Amos Eddy. Long term averages of daily values are also available as they are prerequisite to the interpolation scheme.

This data set, updated annually from the National Climatic Data Center (NCDC), contains summary of the day first order station surface weather observations. The stations are operated by the National Weather Service (NWS), the Federal Aviation Administration (FAA), the U.S. Air Force (Air Weather Service), and the U.S. Navy (Navy Weather Detachment). The NWS and FAA sites are located in the contiguous U.S., Alaska, Puerto Rico, Hawaii and other Pacific Islands. The Air Force and Navy operate stations worldwide.

The observations are generally recorded midnight to midnight for the 24-hour period. The major data parameters are wind, maximum temperature, minimum temperature, wet-bulb temperature, dew-point temperature, maximum relative humidity, minimum relative humidity, precipitation, station pressure, sea level pressure, snowfall, snow depth, and present weather.

The earliest data is for 1881 but most stations begin in 1948. There are currently about 630 first order stations reporting. Periods of record vary widely among the stations.

The summaries of daily land surface observations in this data set are primarily from stations in the National Weather Service (NWS) cooperative station network, but also include NWS principal (first order) climatological stations, and stations supported by the Federal Aviation Administration (FAA), National Park Service (NPS), Bureau of Land Management (BLM), U.S. Forest Service (USFC), U.S. Geological Survey (USGS), and Tennessee Valley Authority (TVA), along with a handful of stations at Air Force and Navy bases. The sites are located in the contiguous U.S., Alaska, Puerto Rico, Virgin Islands, Hawaii and other Pacific Islands.

The major parameters are daily weather, precipitation, snowfall, snow depth, maximum temperature, minimum temperature, and observation temperature. The earliest data is for 1890, but most stations begin in 1948. There are currently about 8500 reporting stations. Periods of record vary widely among the stations.

This set is free, only contains pan evaporation and associated parameters, and does not extend beyond 1978. This data, obtained from the National Climatic Data Center (NCDC), is now included in ds510.0: NCDC TD3200 U.S. Control Summary of Day, 1890(1948)-cont.

Daily and monthly surface observations for Texas stations for January 1900 to December 1973.

This dataset contains summary of day data for various U.S. stations over the period 1890 to 1970.

This dataset contains daily observations of maximum and minimum temperature, precipitation, snowfall, and snow depth for 1062 observing stations in the lower 48 United States. These stations are part of the U.S. Historical Climatology Network, which is a National Climatic Data Center (NCDC) database.

We have obtained these data from the Carbon Dioxide Information Analysis Center's (CDIAC) ftp server.

This global summary of the day and month data set is obtained on a delayed monthly basis from the Climate Prediction Center (CPC) of the National Centers for Environmental Prediction (NCEP). CPC extracts surface synoptic weather observations from the Global Telecommunications System (GTS) and performs limited automated validation of the parameters. The data is then summarized for all reporting stations on a daily basis to current operational requirements related to the assessment of crop and energy production.

Data coverage begins in 1979. In 1987 there is a format change and additional parameters were added. Major parameters include maximum temperature, mimimum temperature, precipitation, vapor pressure, sea level pressure, maximum relative humidity, and minimum relative humidity. If the maximum or minimum temperatures are not reported, they are estimated from reported air temperatures in the regular synoptic reports when sufficient data exist. Starting in 1994, total sky cover, 3-hourly wind direction and speed, and total snow depth are included. There are approximately 8900 actively reporting stations. Periods of record vary widely among the stations.

CAUTIONARY NOTE: NCEP incorrectly decoded the wind units indicator from 01Feb2001 until 15Z,11Jun2002, which caused a knots versus m/sec problem. Not all stations were affected. Users may, with caution, apply the knots or m/sec conversion where it appears to be the correct choice.

The data in this set focuses on daily temperature minimums, maximums, means, and differences for global station data. Several statistical indices are then calculated for each of these quantities, including smoothed and unsmoothed standard deviations and ensemble means. Certain quantities have also been interpolated onto a 2.5-degree mesh grid.

Summary of day for 78 Chinese stations For FGGE.

This dataset contains daily precipitation data for 1974 for stations in several African countries. These data were provided by ASECNA (Agency for Air Traffic Safety in Africa and Madagascar) as part of the GATE experiment.

First version of set. Combined with ds516.1, a later version, to create ds516.2, the preferred set to use.

A later version of the set. Combined with ds516.0 to create ds516.2, the preferred set to use.

This version is the merge of ds516.0 and ds516.1. Duplicate reports have been eliminated. Conflicting reports have been saved in a separate file. Bad reports have been saved in a separate file.

Daily and monthly rainfall for 2300 stations in 10 states in northeastern Brazil are contained in this dataset. Although the first data is for 1904, sixty percent of the stations do not report until the 1960s. The data was obtained from Jack Geisler, University of Utah.

This dataset contains summary of day information (maximum and minimum temperatures and precipitation amount) for Japanese stations from 1951 to 1989.

This dataset contains summary of day information for 108 U.S. WBAN-3 stations from 1948 to 1963.

This dataset contains surface temperature data from 7 U.S. WBAN stations for January 1948 to December 1978.

Station data from 255 U.S. surface stations are archived in this dataset. Station observation data include temperature, dewpoint, wind, visibility, weather, and clouds. Prio to October 1972, observation data are available at 6-hourly intervals, and from October 1972 on, they are available at 3-hourly intervals. The station reports also include 6-hourly precipitation amounts and daily maximum temperature.

Dennis Shea selected 370 U.S. and Canadian stations, with long records, from other DSS datasets. Includes monthly means.

This dataset contains hourly observations and summary of day information from Australian observing stations generally from 1957 to 1982, but some data go back as far as 1939.

This dataset contains Russian summary of day data for 223 Russian stations, beginning as early as 1880 and continuing through 1990. Information in each day's summary includes maximum, minimum, and average temperatures and preciptation total.

This dataset contains monthly summary information for stations in seven upper-Midwest states (Minnesota, Iowa, North Dakota, South Dakota, Montana, Wyoming, and Nebraska). Included in the summaries are temperature, precipitation, and snowfall statistics.

The Auxiliary datasets are datasets in ICOADS standard formats that will be included in the formal ICOADS database in the future. They are offered here, so that users have access to the data as soon as possible and don't need to wait for a formal ICOADS release update. These data have been quality checked for basic parameter ranges, but have not been cross checked for matching duplicates in ICOADS.

The Center for Ocean-Atmosphere Prediction Studies (COAPS) through their Research Vessel Surface Meteorology Data Center (RVSMDC) collect, quality control check, and archive high-resolution marine surface data taken aboard ocean research vessels. The measurements are made with modern and calibrated instrument packages with sampling on the order of one-minute frequency. This is a long-term archive for these data. These data and near real-time extensions can be obtained directly from the RVSMDC.

To support the International Comprehensive Ocean-Atmosphere Data Set (ICOADS) a representative hourly archive is derived from the high-resolution data. These data are available here as archive files and conveniently through ICOADS user data selection interface as 'ICOADS Auxiliary Data Selection'.

As part of international data exchanges this large ship observations archive has been provide to NCAR as a source for inclusion into COADS.

Meteorological and actinomtric (radiation) observations made aboard Russian Research Vesssels were digitized from cruise reports and logbooks at the All-Russian Research Institute of Hydrometeorological Information(RIHMI). The data cover 1937-2000 with the large majority in 1970-1992. There were 3753 documents (logbooks) with meteorological observations and 1125 documents with actinometrical observations digitized. This work was support by a NSF grant.

Surface observations, some XBT profiles, and some sub-surface station data from permanent observing ocean ships. Earliest data is 1945 and latest data is 1991. Added in June 2006 are station and line P profile data extracted from WOD05 by Boyer.

Ship logbook data digitized from the Kobe collection, by programs sponsored by the Japanese Meteorological Agency (JMA), and the Nippon Foundation.

Source data archives

• Arctic region Norwegian ship logbook data. Digitized in Russia, St. Petersburg, by Ecoshelf.
• Norwegian Logbook data from Wishman et al, circa 1995-1996

This dataset is comprised of various input datasets used to develop ICOADS. Some files are in original native format and other are in the ICOADS binary formats. Note, these data files are not recommended for general use. Individual observations and monthly summary statistics in either a 1-degree by 1-degree or 2-degree by 2-degree latitude by longitude boxes are the recommended user data products. Much more information about ICOADS is available at the project website.

ICOADS is the world's largest collection of marine surface in situ observations with more than 200 million records for 1784 through 2005. The collection is formed from many international data sources. The records are merged together and originate from ships (merchant, navy, research), moored and drifting buoys, coastal stations, and other marine platforms. Each report contains individual observations of meteorological and oceanographic variables, such as sea surface and air temperatures, wind, pressure, humidity, and cloudiness. The coverage is global and sampling density varies depending on date and position relative to major ocean shipping routes.

ICOADS is the result of a long-standing (beginning in 1981) cooperative project among the NOAA Earth System Research Laboratory, NOAA National Climatic Data Center, and NSF National Center for Atmospheric Research. Systematic quality assurance and processing into a uniform data format make the dataset a significant and often cited archive.

In an important, related dataset, ICOADS observations are statistically summarized on a monthly basis and in one-degree or two-degree latitude by longitude squares. The monthly summary statistics are also available , as is much more information on the ICOADS project website

ICOADS is the world's largest collection of in situ marine surface observations with more than 200 million records spanning 1784 through 2005. This dataset is the monthly summaries of the observations. The monthly time series are available at 2-degree (1800-2005) and 1-degree (1960-2005) spatial resolutions. After the observations are summarized the data are not interpolated or analyzed to fill data voids. The spatial coverage varies depending on the date and position relative to major ocean shipping routes. A monthly summary for an area box includes ten statistics (e.g. mean, median, standard deviation, etc.) for 22 observed and computed variables (e.g. sea surface and air temperature, wind, pressure, humidity, cloudiness, etc.).

ICOADS is the result of a long-standing (beginning in 1981) cooperative project among the NOAA Climate Diagnostics Center, NOAA National Climatic Data Center, and NSF National Center for Atmospheric Research. Systematic quality assurance and processing of many international data sources into a uniform data format make the dataset a significant and often cited archive.

The ICOADS observations are also available , as is much more information on the ICOADS project website

Marine surface data (ship obs., moored buoys, etc) are segregated from the GTS data stream at NCDC. The GTS records are encoded into IMMA format and collected into monthly files. The primary use of these data are input and cross-check referencing with other I-COADS data sources.

The GTS data stream from the VOSCLIM project ships. The BUFR formatted records are received from the UK Met Office. These data are primarily for the development of the ICOADS database.

Marine surface data are segregated from Global Telecommunication System (GTS) data stream at the National Center for Environmental Prediction (NCEP). The GTS records have global distribution and are from ships, buoys, and other platforms at frequencies native to each respective observing system. The data include many parameters from complete ship surface meteorological observations (wind, atmospheric pressure and and temperature, cloudiness, sea surface temperature, etc.) to limited variables from drifting buoy observations (sea surface temperature and atmospheric pressure). Beginning in March 1997 these data have been collected and archived primarily to support the development of the International Comprehensive Ocean-Atmosphere Data Set (ICOADS). Periodically, the GTS data are processed, quality assurance checked, combined with other data sources, and included in ICOADS.

These data may be suitable for some very specific studies, but it is generally recommended that ICOADS be used instead. The ICOADS observations and ICOADS monthly summary statistics are also available, as is much more information on the ICOADS project website

Simon A. Josey, Elizabeth C. Kent, and Peter K. Taylor compiled the SOC monthly mean flux atlas on a 1 degree by 1 degree grid. The fields have been derived from the COADS Release 1a (1980-93) dataset enhanced with additional metadata from the WMO47 list of ships.

The original version 1.1 atlas was completed in 1998. Additional work was done on the heat flux components in 2002. The air-sea heat flux was adjusted according to contraints determined from estimates of oceanic heat transport. Through the adjustment process the original imbalance of about 30 W/m^2 was reduced to about 2 W/m^2.

Both the original version 1.1 atlas and the adjusted air-fluxes are available in this dataset.

Monthly global grids of data, derived fluxes, and anomalies were prepared from the COADS data.

COADS observations (1945-1993) were used to derive climatologies and year-month anomalies for the basic observed variables and heat flux components. The data are analyzed to give full global coverage.

Grant Petty used the present-weather code values from COADS shipboard reports (1958-1991) to evaluate the frequencies and characteristics of global oceanic precipitation.

Monthly 2-degree by 2-degree and 5-degree by 5-degree wave statistics were computed for 1958-1997 (Global Waves) and 1964-1993 (North Atlantic Waves) respectively from COADS shipboard wave observations. Significant wave heights are determined for several different schemes.

As part of the GODAR (Global Ocean Data Archeology and Rescue) surface marine ship data were digitized in China. The data were received from NCDC on CD-ROM.

Observed vertical profiles of temperature from the worlds ocean.

Vertical temperature profiles of the ocean upper layer.

Ocean profile data of temperature, sailinity, and nutrients

Ocean profile observations for temperature and salinity

Expendable bathythermograph profiles at selected depth levels - instead of inflection points, or 5 meter intervals

High quality global ocean profile stations. Specially selected surface to bottom profiles intended for global ocean circulation studies. Approx. 12000 stations.

North Atlantic transient tracer profiles (tritium, helium-3, t, s, etc).

Observed ocean data taken on three separte surveys of a warm core ring north of the Gulf Stream during 1982

92 CTD oceanographic stations were collected to describe the crest of a Gulf Stream meander during 1988

Long time series of current meter data from 34N 70W in the North Atlantic, position designation Lotus, by WHOI

Ocean surface observations taken from tuna fishing vessels

Monthly river flow rate data for most of the globe except former Soviet Union countries (ds553.1) and US (ds550.1). This compilation was prepared by Byron Bodo (babodo@eol.ca), Jul 2001; it succeeds the Feb 2001 update. This is a combination of datasets from UNESCO, WMO GRDC, the RIVDIS by Vorosmarty (1998), and others. Changes for this version are primarily to documentation, however, there are updates to one Thai and two Bolivian gauges plus replacement of some UK provisional data; details are given in the documentation.

Average record length is about 26.7 years. About 2/3 of the records have from 10-100+ years of data.

The USGS offers the CD-ROM "Hydro-Climatic Data Network: Streamflow Data Set, 1874-1988". This consists of streamflow data measured at 1659 sites throughout the US and its territories. Daily mean, monthly mean, annual mean, minimum and maximum values of discharge are given for each water year. Stations selected distinguish human influences. For more information contact J.R. Slack (jrslack@usgs.gov).

This dataset contains monthly river discharge rates for 4,425 locations around the world except for the former Soviet Union. This dataset combines the UNESCO set (ds552.0) plus 3,455 gages, thus, improving spatial coverage over much of the world with major gains in Brazil, Australia and Argentina. The "new" guages have not been subjected to the rigorous comparisons of the benchmark UNESCO portion. The "new" gages locations, names, and drainage areas have been edited; and obvious errors (e.g., egregious extremes, replicates) removed from flow rates. Data sources and the edits are documented.

The USGS offers a CD-ROM "Hydro-Climatic Data Network: Streamflow Data Set, 1874-1988". This consists of streamflow data measured at 1659 sites throughout the US and its territories. Daily mean, monthly mean, annual mean, minimum and maximum values of discharge are given for each water year. Stations selected distinguish human influences. For more information contact J.R. Slack (jrslack@usgs.gov).

Monthly River flow rates for Russia and former Soviet Union countries prepared by Byron Bodo (babodo@eol.ca) whereby data were combined from ds552.0 (UNESCO and GRD sets) and ds553.0 (Russian set from US-USSR exchange project). Exact dates vary by station. See the inventory for detailed information. Data validation involved corrections which are documented. This is an update of the March 1999 version.

Monthly river flow rates for Russia and former Soviet Union countries in ds553.1 are augmented with data from Russia's State Hydrological Institute (SHI) and a few sites from the Global Hydroclimatic Data Network (GHCDN) . This compendium (v1.1 September 2001) includes 2458 gauges spanning all major river and oceanographic basins of the entire former Soviet Union. Exact dates vary by station. See the inventories for detailed information. Records are generally long: mean length is 36.3 yrs and 1,724 (70%) gauges have 25 or more years of data. 1,373 gages have data to 1985 and another 909 gauges also have some data from 1986-1995. 640 gauges represent watersheds of 10,000 or more sq km, and 1,055 gauges represent catchments of 1,000-9,999 sq km. The SHI additions have not been scrutinized as rigorously as ds553.1.

This dataset contains monthly riverflow data for stations (gauges) on Canadian rivers and streams.

Copy of CD - Global Streamflow Time Series, v.1.0 (a global hydro-climatic data network) Michael Dettinger (USGS) and Henry Diaz (CDC) (Journal of Hydrometeorology, I(4), 289-310, August 2000 This set has been included in ds552.1.

This data is obtained from the National Climatic Data Center (NCDC). "Information contained in this monthly data file are primarily those from the cooperative network, augmented by observations from principal climatological stations operated by the National Weather Service and other sites having highly trained observers."

The Global Historical Climatology Network (GHCN) is a comprehensive global surface baseline climate data set designed for monitoring and detecting climate change. Comprised of surface station observations of temperature, precipitation, and pressure, all GHCN data are on a monthly basis. GHCN is produced jointly by the National Climatic Data Center, Arizona State University, and Carbon Dioxide Information Analysis Center at Oak Ridge National Laboratory.

The Historical Climatology Network (HCN) was established in part to aid in the detection of regional climate change. Some of the HCN stations are first-order weather stations, but most are U.S. cooperative stations. To be included in the HCN, a station must have been active in 1987 and have had at least 80 years of temperature and precipitation data. The HCN database includes station histories, monthly temperatures, and total monthly precipitation data.

This dataset contains sunshine data for 240 U.S. stations (including Puerto Rico and nine Pacific islands) for the years from 1891 to 1984. The periods of record for individual stations varies widely with only 49 (20%) stations having periods of at least 90 years. Howerver, 192 stations (80%) have periods of at least 30 years.

In addition to the total monthly and annual measurements of sunshine duration, total monthly and annual maximum possible sunshine duration and monthly and annual percentages of possiblesummary

Global monthly station temperature and precipitation data from the U.S. Department of Energy

This world monthly surface station climatologicay has data for over 4700 different stations (2600 in more recent years). Data for some stations goes as far back as the mid-1700's. See decadal coverage for more detail. Most of the data was obtained directly from the National Climatic Data Center (NCDC), Asheville, North Carolina. However, much of the data prior to 1951 came from John Wolbach of Harvard College Observatory, who contracted to have this data key entered at NCDC. The first six months of 1961 were key entered at NCAR. Sharon Nicholson, Florida State University, provided African precipitation data to extend the records of over 250 stations. Dennis Shea, NCAR/CGD, has been a valuable source for data obtained directly from various countries.

World Weather Records for 1961-1990

This subset of the World Monthly Surface Station Climatology contains monthly surface station climatologies for U.S. stations for 1961 to 1970.

Monthly temperature and precipitation for U.S. stations for 1981 to 1983 from the Climate Analysis Center (CAC).

Sharon Nicholson at Florida State University compiled this dataset of monthly rainfall totals for African stations. The period covered is 1901 to 1984.

Monthly precipitation for South American stations for 1891 to 1983.

This dataset contains monthly mean rainfall amounts from Malaysia, islands in the central Pacific Ocean, and Hawaii.

This dataset contains 30-year normals of temperature and precipitation for U.S. stations. Also included are the sequential temperature and precipitation data that were used to compute the normals.

This dataset contains monthly precipitation amounts for Indian stations from 1901 to 1970.

This dataset contains monthly temperature and precipitation data for Canadian surface stations through 1989. The period of record for individual stations varies widely, but some stations have data as far back as 1891.

Dennis Shea built this from DSS' DS482.0.

This dataset includes all the monthly precipitation data in the ds564.0 NCDC Global Historical Climate Network and more, and has corrections for wind, wetting, and gauge type.

This dataset contains monthly precipitation amounts for Chinese stations for 1951 to 1980.

Monthly mean surface temperature and monthly accumulated precipitation from 160 land stations in China from 1951 to 1990.

This dataset contains monthly temperature and precipitation data for Chinese stations for 1951 to 1984.

This dataset contains monthly precipitation and maximum and minimum temperature data for 89 Chinese stations for 1965 to 1975.

This set includes a 60 station set with 14 variables and a 205 station set of monthly mean temp and monthly total precip.

This dataset contains summary of month data for 71 Mongolian stations for 1936 to 1983. Summary information includes monthly mean temperature and total monthly precipitation. These data were obtained from Phil Jones at the University of East Anglia.

Data from the ASEAN Climatic Atlas Project (ACAP)

This dataset contains monthly Antarctic surface station data for 1980 to 1989. These data were obtained from the University of Wisconsin.

This Global Atmospheric Research Programme (GARP) dataset was compiled from observations received operationally at the National Meteorological Center (NMC) and the Environmental Technical Applications Center (ETAC), from teletype-punched cards, and from special mailed in reports. Data for June 1970 include global upper air and surface observations from land stations, global surface marine data, and global aircraft reports.

The Line Islands Experiment of 1967 produced extensive data on disturbances in the equatorial trough zone. DSS has some surface and upper-air data from this experiment for February through April of 1967.

The Line Islands Experiment of 1967 produced extensive data on disturbances in the equatorial trough zone. DSS has flight-level aircraft data from flights in February, March, and April 1967 that were flown in conjunction with the experiment.

This dataset contains data from radiosonde observations made during the National Hail Research Experiment. Upper air data are available for the summers of 1972, 1973, 1974, 1976, and 1978.

During the summer of 1977, pibal and ship data were collected in conjunction with the Monsoon Experiment (MONEX). Included in this dataset are pibal data from four special Somalian stations and ship data from five Soviet ships in the Bay of Bengal, Arabian Sea, and equatorial Indian Ocean.

This dataset contains flight-level aircraft data for December 1978 from the Winter Monsoon Experiment (WMONEX).

Data from the Summer Monsoon Experiment, collected by NCAR, NASA, NOAA and France. The project area was over Saudi Arabia, the Arabian Sea, and the Bay of Bengal.

This dataset contains air-sea fluxes and sea-surface data obtained during TOGA COARE by aircraft, ship, buoy, and satellite. For more information about TOGA COARE, see the TOGA COARE Data Information System pages. For a listing of other TOGA COARE datasets at DSS, see the TOGA COARE page under Special Projects.

This dataset contains synoptic surface and upper air data and satellite images obtained during TOGA COARE. For more information about TOGA COARE, see the TOGA COARE Data Information System pages. For a listing of other TOGA COARE datasets at DSS, see the TOGA COARE page under Special Projects.

This dataset contains convection and mesoscale data obtained during TOGA COARE. For more information about TOGA COARE, see the TOGA COARE Data Information System pages. For a listing of other TOGA COARE datasets at DSS, see the TOGA COARE page under Special Projects.

This dataset contains large scale ocean data obtained during TOGA COARE. For more information about TOGA COARE, see the TOGA COARE Data Information System pages. For a listing of other TOGA COARE datasets at DSS, see the TOGA COARE page under Special Projects.

This dataset contains ocean mixing data obtained during TOGA COARE. For more information about TOGA COARE, see the TOGA COARE Information System pages. For a listing of other TOGA COARE dataset at DSS, see the TOGA COARE page under Special Projects.

A 51-year hydrological reanalysis between 1948 and 1998 over the continental US was recently completed at Climate Prediction Center of NCEP, NOAA. The project used Noah land surface model (LSM) with forcing data regridded from NCEP/NCAR Reanalysis data and the new CPC daily and hourly precipitation data. The best available orography, soil and vegetation types, etc. are used as the boundary conditions of this study. The output fields include all components of energy, water mass balance, snow cover and depth, and 4 layers of soil temperature, moisture (both liquid and frozen) below the ground. Three types of data files are archived in this dataset. They include the restart files, the forcing fields to run the Noah Land Surface Model, and the output files from the 1948-1998 runs. The output files are available on line for registerd users.

The daily restarting files and the hourly forcing fields are stored on NCAR's mass storage systems (MSS) only.

There are 24 variables including 8 energy components and skin temperature every 3 hours and 15 daily averages. See variable lists below for details. For the NOAH_h.grb (three hourly), NOAH_d.grb (daily accumulation or instantaneous) and NOAH_dd.grb (daily mean), different users may extract and generate sub-dataset for themselves.

You can find Noah LSM used in this study from: Noah LSM ver_2.5.2

The NCEP North American Regional Reanalysis products are on Eta 221 grid (32km) at 29 pressure levels. They were produced using Eta 32km/45-layer model. The input data includes all observations used in NCEP/NCAR Global Reanalysis project, and additional precipitation data, TOVS-1B radiances, profiler data, land surface and moisture data, etc. The output analyses are 3-hourly with additional 9 variables in the 3-hour forecasts to reflect accumulations or averages.

The NOAA FSL MAPS model data (including Model Output Reduced Data Set (MORDS) and Model Location Time Series (MOLTS) data) for the GCIP project. The MAPS model used for GCIP is an advanced, experimental version which is running on a 40-km grid with 40 vertical isentropic-sigma levels. Data are ingested into MAPS on a 1-hour assimilation cycle. Forecasts are run from MAPS out to 12-hour every 3 hours.

The operational NCEP Eta 212 grid (40km) model output is archived in this dataset. This is one of the model output archives for GCIP project. In addition to the MORDS (Model Output Reduced Data Set) and MOLTS (Model Output Location Time Series) analyses required by GCIP project, all upper air levels and their analyses, initialization, and forecasts are saved as well.

The CMC GEM (replacement of RFE) model outputs for GCIP project.

This dataset contains dropwindsonde data from the First GARP Global Experiment (FGGE) Special Observation Periods 1 (December 7, 1978 to February 26, 1979) and 2 (May 6, 1979 to July 25, 1979).

From the ALPine EXperiment, for FGGE

This dataset contains data from the Tropospheric Wind Earth Radio Location Experiment (TWERLE) constant-level balloons over the Southern Hemisphere for June 1975 to August 1976. The balloon measurements were taken at pressures near 150mb and include temperature and wind.

The Hydrologic Atmospheric Pilot Experiment (HAPEX) was a project designed to measure the hydrological budget and evaporation flux at the scale of a general circulation model grid square (104 km2 at the time). The HAPEX research area was in the southwest part of France, and the NCAR King Air participated by making routine daytime observations of heat, moisture, and momentum fluxes during a 67-day period May 9 and July 14, 1986. This dataset contains flight-level data from 20 King Air missions.

This dataset contains daily tropospheric and stratospheric analyses from the European Centre for Medium-Range Weather Forecasts (ECMWF) for the Australian Monsoon Experiment (AMEX) period of January 10-January 22, 1987. The grids are in GRIB Edition 0 format, and parameters include temperature, geopotential height, relative humidity, vertical motion, vorticity, and wind divergence on levels from 1000mb up to 10mb. There are also some surface variables available.

This is data extracted from various NCEP observed and analyzed datasets and then put in our RPTOUT format to support the Kuwait Oil Fires studies.

NOAA's National Environmental Satellite Services (NESS) has provided global longwave radiation analyses on a 5-degree latitude/longitude grid for the period from 19 April to 10 July 1971.

Other nephanalyses in DS038.0.

This dataset contains monthly nephanalyses of average daily cloud cover for the central Pacific Ocean. The data are on a 2.5-degree grid covering the Pacific from 30S latitude to 57.5N, and from 105E longitude to 282.5E. The analyses were provided by James Sadler of the University of Hawaii.

This dataset contains daily sea ice and rain rate data from the Nimbus-5 Electronically Scanned Microwave Radiometer (ESMR). The data, on a global 2.5-degree latitude/longitude grid, cover the 22-day period from 29 January to 19 February 1975.

This dataset contains NESS solar and infrared radiation data on global 2.5 degree latitude-longitude grids and 125x125 polar stereographic grids. The data is derived from data from the NOAA polar orbiting satellites.

This dataset contains 5-day mean brightness analyses from NOAA's National Environmental Satellite Services (NESS) for the period from June 1974 to September 1986. The data are on a 5-degree global grid.

NOAA's National Environmental Satellite Services (NESS) has provided visible and infrared brightness data from the Improved TIROS Operational System (ITOS) for 22 October to 24 October 1973. The data are on a 512x512 global grid.

This dataset contains visible and infrared brightness data on a global 2048x2048 grid from NOAA's National Environmental Satellite Services (NESS) for the 2-day period of 22-23 October 1973.

Data along the orbits for five brightness channels, Nimbus-5 SCR. Also includes daily global analyses of the channels.

This dataset contains global analyses of minimum albedo values over several periods between April 1969 and February 1970 from the Nimbus-3 satellite.

The Environmental Data and Information Service (EDIS) has provided global gridded visible satellite data from the Television and Infrared Observation Satellite (TIROS) for the period from January to June 1979 in support of the First GARP Global Experiment (FGGE).

Gridded analyses of OLR (Outgoing Longwave Radiation) and HRC (Highly Reflective Cloud) anomalies wherein the bias due to changing equatorial crossing times has been estimated and removed.

This dataset contains daily,pentad and monthly OLR data in which the bias from varying satellite equatorial crossing times (ECT's) has been estimated and then removed. This was done by starting with the twice daily 2.5 degree NCEP/NOAA OLR archive for 1974 thru 1999. The ECT biases caused by afternoon satellite orbital drift AND the abrupt transition from a morning satellite to an afternoon satellite were estimated separately and then these estimates were removed. Please see Estimating the Satellite Equatorial Crossing Time Biases for more details.

This dataset contains radiance data from the Nimbus-3 Satellite Infrared Spectrometer (SIRS) A and B instruments. The data are available for various periods between April 1969 and April 1971.

TIROS TOVS soundings. Basic NESS format (compact), not FGGE format. Both basic radiances and a retrieval sounding in each report, each 250 km.

This dataset contains global vertical temperature data from the NOAA Vertical Temperature Profile Radiometer (VTPR) for September 1974.

Eight infrared channels from NOAA's Vertical Temperature Profile Radiometer (VTPR) provide global vertical temperature data for the period from November 1972 to February 1979.

Subtarget radiance observations for November 1978, and January, July, and November 1979 from NESDIS.

This dataset contains global-coverage satellite data from NOAA's polar orbiter for January 1981 to January 1982. For more information about the Global Area Coverage (GAC) data, see section 3.1 of the NOAA Polar Orbiter Data User's Guide.

Basic TOVS satellite sounder data. Visible, infrared, U wave, stratospheric channels. NCAR has all this data for the FGGE year and later.

NASA maintains a graphic display of this data from 1979 to near present.

This dataset contains daily and monthly oceanic precipitation analyses on a 2.5-degree global grid. The data were constructed from the Microwave Sounding Units of seven TIROS-N series satellites. Data are available for the period between January 1979 and May 1994.

The daily data are considered to be not as reliable as the monthly data. Before using the daily data, it is highly recommended that you read the documentation associated with it.

A reprocessing of Advanced Very High Resolution Radiometer (AVHRR) Pathfinder atmosphere (PATMOS) results in a 20-year global climate dataset providing information on the earth's radiation budget. This dataset is the result of collaborative work between NOAA/NESDIS and NASA. The measurements from NOAA-7, NOAA-9, NOAA-11, and NOAA-14 satellites are combined in this dataset.

NOAA POLAR ORBITER GAC, 4KM DATA, 1989JAN01-1993NOV05. Data of the following periods are not available yet as of July 06, 1995. 1990JAN01 0000 UTC - 1990MAY24 0517 UTC 1992MAR04 1325 UTC - 1992MAR08 0551 UTC 1992MAR22 1131 UTC - 1992MAR26 0354 UTC 1993APR24 0900 UTC - 1993APR28 2200 UTC 1993JUN19 0000 UTC - 1993JUN24 0300 UTC 1993AUG15 0300 UTC - 1993AUG16 2300 UTC

Zonal Mean Methane and Nitrous Oxide Concentrations from the Nimbus-7 Stratospheric and Mesopheric Sounder (SAMS) Experiment

Total ozone observations from NASA's Nimbus-7 satellite.

Temperature Humidity Infrared Radiometer (THIR) radiances together with Total Ozone Mapping Spectrometer (TOMS) derived UV reflectivities, climatological temperature lapse rates, and current surface temperature and snow/ice information are used to generate cloud data. The cloud data include, total cloud amount, cloud amounts at high, middle and low altitudes, currus and deep convective clouds, and cloud and surface radiances.

Total-column ozone observations from NASA's Nimbus-4 satellite.

A subset of Synchronous Meteorological Satellite data for GATE

The Indian Meteorological Department has provided satellite data from the Indian National Satellit (INSAT) for the period April 1984 to March 1989. Data are available twice-daily until April 1988 when data become available 8 times daily.

This dataset contains global semi-monthly temperature distribution histograms derived from GOES satellite data. The data are on a 2.5-degree grid and span cover the period from December 1981 to October 1989.