Introduction

Here we document the CAMS reanalysis dataset, which, eventually, will cover the period January 2003 to near real time (NRT), though currently only data for the period 2003-2017  have been released. The CAMS reanalysis is the latest global reanalysis data set of atmospheric composition (AC) produced by the Copernicus Atmosphere Monitoring Service, consisting of 3-dimensional time-consistent AC fields, including aerosols, chemical species and greenhouse gases, though GHG fields will only be released in 2019. The data set builds on the experience gained during the production of the earlier MACC reanalysis and CAMS interim reanalysis.

The CAMS reanalysis was produced using 4DVar data assimilation in CY42R1 of ECMWF’s Integrated Forecast System (IFS), with 60 hybrid sigma/pressure (model) levels in the vertical, with the top level at 0.1 hPa. Atmospheric data are available on these levels and they are also interpolated to 25 pressure, 10 potential temperature and 1 potential vorticity level(s). "Surface or single level" data are also available.

Generally, the data are available at a sub-daily and monthly frequency and consist of analyses and 48h forecasts, initialised daily from analyses at 00 UTC.

The data are archived in the ECMWF data archive (MARS) and can be retrieved using the ECMWF Public Dataset service via the WebAPI (ECMWF Member State users can access the data using MARS directly, in the usual manner).

The IFS model and data assimilation system

The 4DVar data assimilation uses 12 hour assimilation windows from 09 UTC to 21 UTC and 21 UTC to 09 UTC.

The IFS model documentation for various model cycles can be found on https://www.ecmwf.int/en/forecasts/documentation-and-support/changes-ecmwf-model/ifs-documentation. The model used in the CAMS reanalysis includes several updates to the aerosol and chemistry modules on top of the standard CY42R1 release, which are listed below. Please note that 42r1 documentation is not available on the page, but the code for the earlier and later cycles is available for reference.

Aerosol model

• Updated aerosol optical properties, especially for organic matter (see Technical Memo 801 by Bozzo et al. 2017: "Implementation of a CAMS-based aerosol climatology in IFS").
• Bug fixes to sedimentation, which was unreasonably weak for some dust and sea-salt bins, with corresponding re-tuning of sea-salt scavenging.
• SO₂ dry deposition velocities updated to match those used in the chemistry scheme (from SUMO).
• New parametrisation of anthropogenic Secondary Organic Aerosol (SOA) production, proportional to non-biomass-burning CO emissions.
• More detailed SO₂ to sulfate aerosol conversion with dependence on temperature and relative humidity, and overall decrease in the conversion timescale especially at high latitudes.
• Increased sulfate dry deposition velocity over ocean.
• Mass fixer extended to aerosol species.
• Scaling of biomass-burning Black Carbon (BC) emissions using the ratio of BC AOD (CAMS interim reanalysis) / BC AOD (CAMS interim control run).
• 80% of SO₂ emissions are released in the two lowest model levels (as an update of tendencies) rather than at surface (fluxes)

Chemistry mechanism

The chemical mechanism of the IFS is an extended version of the Carbon Bond 2005 (CB05) chemical mechanism as implemented in CTM Transport Model 5 (TM5). In the CAMS reanalysis the model as documented in Flemming et al. (2015) and Flemming et al. (2017) is used with the following updates:

• Update of heterogeneous rate coefficients for N2O5 and HO2 based on clouds and aerosol
• Modification of photolysis rates by aerosol
• Dynamic tropopause definition based on T profile for coupling to stratosphere and tropospheric mass diagnostics
• Monthly mean VOC emissions calculated by the MEGAN model using MERRA reanalysed meteorology (Sindelarova et al., 2014) for all VOCs and for whole period 2003-2015 period.
• Bugfixes, in particular for diurnal cycle of dry deposition whose correction has decreased ozone dry deposition (about 15-20%)     
• The version number for the chemistry scheme is CHEM_VER=15

Greenhouse Gases

The model configuration for greenhouse gases is based on the specification of the following components documented in the listed papers below:

• Emissions for CO2 are documented in Agusti-Panareda et al. (2014), Massart et al. (2016).
• Bias correction for CO2 ecosystem fluxes based on the Biogenic Flux Adjustment Scheme is documented by Agusti-Panareda et al. (2016)
• Emissions and loss rate for CH4 is documented in Massart et al. (2014)
• Mass fixer configuration for CO2 and CH4 is documented by Agusti-Panareda et al. (2017)

Emission datasets

The emissions datasets used to produce the CAMS reanalysis are listed in Table 1. They include the MACCity anthropogenic emission, GFAS fire emissions, MEGAN biogenic emissions and several GHG emission datasets.

Table 1: Emission datasets used in the CAMS reanalysis

Data organisation and access

The data is listed in ECMWF's public data catalogue. To access the data use the ECMWF Web API with ‘dataset’:'eac4'.

Users with access to MARS can also browse the data on the MARS catalogue under class=mc and expver=eac4.

Stream:

• oper: sub-daily
• mnth: synoptic monthly means
• moda:monthly means of daily means

Type:

• an: analyses
• fc: forecasts

Levtype:

• sfc: surface or single level
• pl: pressure levels
• pt: potential temperature levels
• pv: potential vorticity level
• ml: model levels

Spatial grid

The CAMS reanalysis data have a resolution of approximately 80 km. The data are available either as spectral coefficients with a triangular truncation of T255 or on a reduced Gaussian grid with a resolution of N128. These grids are so called "linear grids", sometimes referred to as TL255.

Temporal frequency

For sub-daily data for the CAMS reanalysis (stream=oper) the analyses (type=an) are available 3-hourly. The daily forecast, run from 00 UTC, has 3-hourly steps from 0 to 48 hours for the 3D model level and pressure level fields, and hourly steps from 0 to 48 hours for the surface fields.

Monthly means

Several parameters are also available as synoptic monthly means, for each particular time and forecast step (stream=mnth) and as monthly means of daily means, for the month as a whole (stream=moda).

Monthly means for analyses and instantaneous forecasts are created from data with a valid time in the month, between 00 and 23 UTC, which excludes the time 00 UTC on the first day of the following month. Monthly means for accumulations and mean rates are created from data with a forecast period falling within the month. For example, monthly means of daily means for accumulations and mean rates are created from contiguous data with forecast periods spanning from 00 UTC on the first day of the month to 00 UTC on the first day of the following month.

Data format

Model level fields are in GRIB2 format. All other fields are in GRIB1, unless otherwise indicated.

Level listings

Pressure levels: 1000/950/925/900/850/800/700/600/500/400/300/250/200/150/100/70/50/30/20/10/7/5/3/2/1

Potential temperature levels: 300/315/320/330/350/370/395/475/600/850

Potential vorticity level: 2000

Model levels: 1/to/60, which are described at https://www.ecmwf.int/en/forecasts/documentation-and-support/60-model-levels.

Parameter listings 

The archive of available parameters can be browsed here.

Satellite Data

The atmospheric composition satellite retrievals used as input into the CAMS reanalysis are listed below. The following abbreviations are used in Table 1. TC: Total column, TRC: Tropospheric column, PROF: profiles, PC: Partial columns, ColAv: Column average mixing ratio, QR= quality flag given by data providers, SOE: Solar elevation, MODORO: Model orography, PRESS_RL= pressure at bottom of layer, LAT: Latitude.

Table 2: Satellite retrievals of atmospheric composition that were assimilated in the CAMS reanalysis

Validation reports

Validation Reports for the CAMS reanalysis can be found on the CAMS Quality Assurance website.

Guidelines

The following advice is intended to help users understand particular features of the CAMS reanalysis data:

• Users who want to use meteorological data only are advised to use the ERA5 meteorological reanalysis.
• CAMS data users please use the 'GEMS Ozone' (param 210203) and 'Total Column GEMS Ozone' (param 210206) fields. These are produced specifically for CAMS using the full tropospheric chemistry scheme, see also CAMS Global data: What is "GEMS ozone".

Known issues

At the time of writing (2017-11) we are aware of these issues with the CAMS reanalysis:

• Validation of AOD with Aeronet data has show there are some hot spots around outgassing volcanoes (in particular Mauna Loa and Mexico City) with high analysis AOD values that degrade the global average RMSE. If calculating global mean statistics it is advisable to exclude those two stations as unrepresentative. This is a side effect of model-resolution orography not resolving the height of the volcanoes that has been unmasked by recent enhancements to the SO2 oxidation scheme which improve aerosol on the global scale.
• During 2003 the ozone analysis has a degraded quality (bigger biases with respect to observations) in Arctic and Antarctic free troposphere because MIPAS and SCIAMACHY data of lower quality were assimilated.
• Between March-August 2004 no ozone profile data were available for assimilation. This affects the vertical structure of the ozone analysis and we see larger biases wrt ozone sondes, especially in the Antarctic.
• From 2013 onwards there is a larger seasonally varying bias in ozone in the free troposphere, particularly in the Arctic and Antarctic that is not seen in the control run. The reason for this bias is a change in the observing system, namely the change from 13-layer SBUV/2 data to 21-layer SBUV/2 data in July 2013 (see Table 2) that unfortunately has an impact on tropospheric ozone. A similar bias is seen in the NRT CAMS ozone analysis which also uses the 21-layer SBUV/2 data after 2013.
• During 2003 the seasonal cycle of the tropospheric column NO2 is not well represented because of the assimilation of SCIAMACHY NO2 data of degraded quality.

• The use of the NOx variable from the CAMS reanalysis (as well as from the CAMS interim re-analysis and the CAMS operational system) is not recommended. The user is advised to download NO and NO2 separately and to add them up. Please note that a conversion of the mass mixing ratios [kg/kg] to volume mixing rations / molar fractions [mol/mol] is needed to do this in a meaningful way.

This list will be updated as we become aware of further issues in the CAMS reanalysis.

How to cite the CAMS Reanalysis

Please acknowledge the use of the CAMS reanalysis as stated in the Copernicus C3S/CAMS License agreement:

• "Where the Licensee communicates to the public or distributes or publishes CAMS Information, the Licensee shall inform the recipients of the source of that information by using the following or any similar notice:

  'Generated using Copernicus Atmosphere Monitoring Service Information [Year]'.

• Where the Licensee makes or contributes to a publication or distribution containing adapted or modified CAMS Information, the Licensee shall provide the following or any similar notice:

  'Contains modified Copernicus Atmosphere Monitoring Service Information [Year]';

Any such publication or distribution shall state that "neither the European Commission nor ECMWF is responsible for any use that may be made of the information it contains."

References

A reference paper is available from www.atmos-chem-phys.net/19/3515/2019/ and further CAMS reanalysis references will be available from the ECMWF e-Library in the future. 

An ECMWF newsletter article 'The new CAMS global reanalysis of atmospheric composition' is available from: https://www.ecmwf.int/node/18821

Mailing list

To be kept informed of the latest news associated to the CAMS Reanalysis products, you may subscribe to the CAMS Global Reanalysis mailing list.

Related articles