The latest developments for the reactive gas modelling have focused on fully integrating the chemistry in the IFS complementing the on-line integration strategy already pursued for aerosol and greenhouse gases. The IFS including the modules for atmospheric composition is named Composition-IFS (C-IFS). C-IFS makes it possible (i) to use the detailed meteorological simulation of the IFS for the simulation of the fate of constituents, (ii) to simulate atmospheric chemistry globally at high resolution because of the computational parallel efficiency of the IFS, (iii) to use the IFS data assimilation system to assimilate observation of atmospheric composition and (iv) to simulate feedback between atmospheric composition and weather. Three different chemical schemes are being used: TM5, MOZART, and MOCAGE, with the implementation of TM5 currently being the most mature.
In this configuration, chemistry originating from the TM5 CTM, has been fully integrated into the IFS forecasting model. This version applies 54 tracers and 120 reactions and focuses on tropospheric ozone-CO-NMVOC chemistry. Stratospheric ozone is modelled with a linearized scheme. A more extensive description can be found in this document. This system runs semi-operational under expid fsd7 since November 2012.
Documentation coming soon!
In the early stages of the global atmospheric composition service development two chemistry transport models (CTMs), which had been in use by the scientific community, were coupled to ECMWF’s integrated forecasting system (IFS) using specific coupling software. Meteorological information is passed to the CTMs every hour, and the CTMs return concentration fields and chemical production and loss rates to the IFS for use in the data assimilation routines. This set-up has been in use throughout GEMS, MACC, and MACC-II (2007-2013).
This configuration employs the MOZART3 CTM with some extensions for tropospheric hydrocarbon and aerosol chemistry from MOZART4. The chemistry mechanism encompasses 115 species and 355 reactions and covers both tropospheric and stratospheric ozone and precursor chemistry. Further information can be found in the IFS-MOZART Model Description. This system has been run semi-operational under expid fkya from September 2009 till present and has been the backbone of the near-real-time assimilation and forecasting system and the reanalysis system.
In this configuration the TM5 CTM, version TM5-chem-3.2, is coupled to the IFS. TM5 applies 53 tracers and 107 reactions and focuses on tropospheric ozone-CO-NMVOC chemistry. Stratospheric ozone is constrained with climatological fields. Further information can be found in the IFS-TM5 Model Description. This system has been run semi-operational under expid fl52 between December 2011 and January 2013.
Documentation coming soon!
Next to the main IFS-CTM and C-IFS systems, MACC-II uses three other chemical data assimilation models that are focused on stratospheric chemistry. These modelling systems provide their own specific data products, but are also used as a service development mechanism. More information can be found below and in the Technical documentation of the MACC integrated stratospheric ozone service.
The BASCOE data assimilation system assimilates the offline dataset (level-2) retrieved (v3.3) from the Aura-MLS instrument. While delivered a few days later than the NRT stream, the offline dataset includes many species besides ozone.
Since August 2013, the NRT SACADA product used in MACC-III is based on the assimilation of both Aura-MLS NRT and offline data. This allows analysis with a 1-day lag only. As SACADA runs on a icosahedron grid, the output fields are interpolated onto a regular lat-lon grid. Note also that SACADA utilizes the GME NWP model to calculate its own 24h forecasts based on ECMWF 00:00 UTC analyses.
Short summery of the main product/model characteristics:
The NRT TM3DAM product provided in MACC-II is the ozone total column analysis of Envisat/SCIAMACHY (until March 2012) and GOME-2 (April 2012 - NRT) by the TM3 Data Assimilation Model. It is run in the following configuration:
More details and bibliographic references on TM3DAM can be found in the online TM3DAM documentation.
The IFS-LMD aerosol scheme is the scheme that was introduced to add aerosol modelling to the ECMWF IFS forecasting system. It is currently used for the daily analysis and 5-day forecast and was also used for the MACC reanalysis.
The initial package of ECMWF physical parameterizations dedicated to aerosol processes mainly follows the aerosol treatment in the LOA/LMD-Z model (Boucher et al. 2002; Reddy et al. 2005). Five types of tropospheric aerosols are considered: sea salt, dust, organic and black carbon, and sulphate aerosols. Prognostic aerosols of natural origin, such as mineral dust and sea salt are described using three size bins. For dust bin limits are at 0.03, 0.55, 0.9, and 20 microns while for sea-salt bin limits are at 0.03, 0.5, 5 and 20 microns. Emissions of dust depend on the 10-m wind, soil moisture, the UV-visible component of the surface albedo and the fraction of land covered by vegetation when the surface is snow-free. A correction to the 10-m wind to account for gustiness is also included (Morcrette et al. 2008). Sea-salt emissions are diagnosed using a source function based on work by Guelle et al. (2001) and Schulz et al. (2004). In this formulation, wet sea-salt mass fluxes at 80% relative humidity are integrated for the three size bins, merging work by Monahan et al. (1986) and Smith and Harrison (1998) between 2 and 4 µm. Sources for the other aerosol types which are linked to emissions from domestic, industrial, power generation, transport and shipping activities, are taken from the SPEW (Speciated Particulate Emission Wizard), and EDGAR (Emission Database for Global Atmospheric Research) annual- or monthly-mean climatologies. More details on the sources of these aerosols are given in Dentener et al. (2006). Emissions of OM, BC and SO2 linked to fire emissions are obtained using the GFAS system based on MODIS satellite observations of fire radiative power, as described in Kaiser et al. (2011). Several types of removal processes are considered: dry deposition including the turbulent transfer to the surface, gravitational settling, and wet deposition including rainout by large-scale and convective precipitation and washout of aerosol particles in and below the clouds. The wet and dry deposition schemes are standard, whereas the sedimentation of aerosols follows closely what was introduced by Tompkins (2005) for the sedimentation of ice particles. Hygroscopic effects are also considered for organic matter and black carbon aerosols. A detailed description of the ECMWF forecast and analysis model including aerosol processes is given in Morcrette et al. (2009) and Benedetti et al. (2009).
MODIS AOD data at 550 nm are routinely assimilated in a 4D-Var framework which has been extended to include aerosol total mixing ratio as extra control variable (Benedetti et al. 2009). A variational bias correction for MODIS AOD is implemented based on the operational set-up for assimilated radiances following the developments by Dee and Uppala (2008). The bias model for the MODIS data consists of a global constant that is adjusted variationally in the minimization based on the first-guess departures. Although simple, this bias correction works well in the sense that the MACC analysis is not biased with respect to MODIS observations. The observation error covariance matrix is assumed to be diagonal, to simplify the problem. The errors are also chosen ad hoc and prescribed as fixed values over land and ocean for the assimilated observations. The aerosol background error covariance matrix used for aerosol analysis was derived using the Parrish and Derber method (also known as NMC method; Parrish and Derber, 1992) as detailed by Benedetti and Fisher (2007). This method was long used for the definition of the background error statistics for the meteorological variables and is based on the assumption that the forecast differences between the 48-h and the 24-h forecasts are a good statistical proxy to estimate the model background errors.
The IFS-GLOMAP aerosol scheme is currently being implemented to improve the aerosol modelling within the ECMWF IFS forecasting system.
A new aerosol module "GLOMAP-mode" is being implemented into the IFS to upgrade on the mass-based scheme developed initially and referred to here as IFS-LMD (Morcrette et al., 2009). The GLOMAP-mode aerosol microphysics scheme (e.g., Mann et al., 2010; Mann et al., 2012) simulates the evolution of the particle size distribution, with explicit sources and sinks of particle number (e.g., via nucleation and coagulation) as well as mass. The scheme tracks the same component masses as the IFS-LMD scheme (sulphate, sea-salt, mineral dust and black and organic carbon) but calculates how their composition is distributed across the size range resolving internal mixtures and gas to particle transfer. Resolving these aerosol microphysical processes has been shown to improve the fidelity of simulated aerosol radiative forcings (Bellouin et al., 2013) and also allows to provide improved aerosol boundary conditions to regional air quality models, many of which include similar aerosol microphysics schemes.
The coupling of the IFS-GLOMAP simulated aerosol to the AOD data assimilation system is being carried out in MACC-II and will be posted here when it is in place.
The greenhouse gases CO2 and CH4 are implemented as simple tracers with prescribed fluxes at the surface and in the case of CH4 a parameterized loss rate due to OH and stratospheric radicals. The only exception is the coupling to the CTESSEL land biosphere model, which models the vegetation fluxes on-line as a function of the relevant model variables.
For the CO2 modelling in the IFS forecasting system, the land vegetation fluxes are modelled on-line by the CTESSEL carbon module (Boussetta et al., 2013). The anthropogenic fluxes are based on the annual mean EDGARv4.2 inventory (http://edgar.jrc.ec.europa.eu/overview.php?v=42, Janssens-Maenhout et al., 2012) using the most recent year available (i.e. 2008) with estimated and climatological trends to extrapolate to the current year. The ocean fluxes are from the Takahashi et al (2009) climatology and the fire fluxes are from GFAS (Kaiser et al., 2012). The meteorological fields are re-initialized every 24 hours (at 00UTC) using the operational ECMWF analysis and the CO2 fields are allowed to evolve freely from one 24-hour forecast to the next. Because the budget is currently not constrained by observations, the CO2 global bias can grow from one year to the next. In order to avoid this, the initial CO2 forecast fields are updated every year with the most recent atmospheric 3D fields from the MACC CO2 flux inversion system from LSCE (Chevallier et al., 2010) whenever these become available, typically in October. Methane fluxes are prescribed in the IFS using inventory and climatological data sets, consistent with those used as prior information in the CH4 flux inversions from Bergamaschi et al. (2009). The anthropogenic fluxes are from the EDGAR 4.2 database (Janssens-Maenhout et al, 2012) for the year 2008, i.e. the last available year. All the anthropogenic categories are based on annual mean values, except for rice, which has been modulated with a seasonal cycle of the Matthews monthly inventory for rice (Matthews et al., 1991). The wetland fluxes are from the Kaplan climatology described in Bergamaschi et al. (2007). The biomass burning emissions are from the MACC-II GFAS dataset (Kaiser et al., 2012). The other sources/sinks include wild animals (Houweling et al., 1999), termites (Sanderson et al., 1996), oceans (Houweling et al., 1999 and Lambert and S. Schmidt, 1993) and a soil sink (Ridgwell et al., 1999). For the chemical sink in the troposphere and the stratosphere, the climatological chemical loss rates from Bergamaschi et al. (2009) are used. These are based on OH fields optimised with methyl chloroform using the TM5 model (Krol et al., 2005) and prescribed concentrations of the stratospheric radicals using the 2-D photochemical Max-Planck-Institute model.
Greenhouse gases data from space are not yet available in near real time, mainly due to their partial reliance on auxiliary data. The assimilation system is therefore run with a 6 month lag in a so-called "delayed mode" (DM). It assimilates dry molar fraction of CO2 from the Thermal And Near-infrared Sensor for carbon Observation (TANSO) instrument and dry molar fraction of CH4 from TANSO as well and also from the Infrared Atmospheric Sounding Interferometer (IASI) instrument. The xCO2 product is provided by the Institute of Environmental Physics (IUP), University of Brenment (UB). The xCH4 TANSO product is provided by the SRON Netherlands Institute for Space Research using the proxy setup of the RemoTeC algorithm [Butz et al., 2010], a joint development between SRON and the Karlsruhe Institute of Technology (KIT). The xCH4 IASI product is provided by the Laboratoire de Meteorologie Dynamique (LMD) using a non linear inference scheme based on Multilayer Perceptron neural networks [Crevoisier et al., 2009 and 2013].
The MACC-II CO2 flux estimates are provided by the variational LMDZ flux inversion system.
The CO2 surface fluxes are estimated over more than three decades, from 1979 to 2013, at resolution 3.75 x 1.9 degrees (longitude-latitude) and 3-hourly, based on 131 CO2 mole fraction station records from three large databases:
The MACC-II CH4 flux estimates are provided by the 4D-VAR TM5 flux inversion system.
The MACC-II CH4 flux inversion system is based on the TM5-4DVAR inverse modeling system [Bergamaschi et al., 2009]. Observations from SCIAMACHY as well as high accuracy surface measurements from the NOAA global cooperative air sampling network are used in the inversion. The latter constrain significantly the surface mixing ratios in remote regions (ocean) and allow deriving corrections for potential small latitudinal or seasonal biases of the satellite data. 3D fields of CH4 mixing ratios from the TM5-4DVAR inversion are also available.
The MACC CH4 flux estimates are provided by the 4D-VAR TM5 flux inversion system [Meirink et al., 2008; Bergamaschi et al., 2013a]. Two different production streams are generated:
The reanalysis aims at the best possible consistency over the entire time series, in order to analyze trends and inter-annual variability (IAV) of CH4 emissions. The reanalysis has been initially performed for 2000-2010 [Bergamaschi et al., 2013a], and recently been extended until end 2012. The reanalysis includes inversions using only NOAA surface observations (scenario S1-NOAA: 2000-2012; MACC version ID: v10-S1NOAA_ra), and inversions using in addition also XCH4 satellite retrievals from SCIAMACHY (Iterative Maximum A Posteriori Version 5.5 (IMAP V5.5) [Frankenberg et al., 2011]; scenario S1-SCIA: 2003-2011; MACC version ID: v10-S1SCIA_ra).
Regular updates of global CH4 flux inversions are provided every 6 months (for period ~7-13 months before real-time). These 'Delayed‐Mode' CH4 inversions use generally both satellite and surface observations. Since beginning 2012, the GOSAT RemoteC PROXY v2.0 XCH4 retrievals [Schepers et al., 2012] are used. The model output from these GOSAT based delayed-mode inversions is available under the MACC version ID v10_an. Further details of the 'Delayed-mode' CH4 flux inversions are described in [Bergamaschi et al., 2013b] and [Bergamaschi and Alexe, 2014].
This data describes the N2O surface fluxes over 17 years, from 1996 to 2012, at 3.75° × 1.875° (longitude by latitude) and monthly resolution, based on air mole fraction records from 124 N2O sites plus aircraft, ship-based and ocean mooring records from:
These databases include both in situ measurements made by automated quasi-continuous analysers and irregular air samples collected in flasks and later analyzed in central facilities.
The flux inversion is based on a variational Bayesian inversion system, like the 4D-Var data assimilation system used in MACC-II, which allows the fluxes to be estimated at relatively high resolution over the globe. Fluxes and mole fractions are linked in the system by a global atmospheric transport model, which accounts for the loss of N2O in the stratosphere via photolysis and reactions with metastable oxygen atoms O(1D). A series of flux model simulations, flux inventories, and flux error models regularizes the solution to the flux inference problem. The uncertainty of the inverted fluxes is quantified from the Bayesian theory by a robust Monte Carlo method.