Monitoring atmospheric composition & climate

The ENS sub-project is one of the cluster of sub-projects dedicated to regional air-quality services for Europe.  ENS primarily focusses on the delivery and verification of the prototype operational European-scale regional NRT air-quality services.  In direct continuation from MACC, this service is based upon an esemble of forecasts performed at seven centres in Europe, recognised for their continuing experience in providing routine operational or pre-operational forecasts at the scale of Europe for several years.  The different assimilation and forecast suites are those employed in MACC (CHIMERE, EMEP, EURAD, LOTOS-EUROS, MATCH, MOCAGE and SILAM).  The individual forecasts will use ECMWF operational meteorological forecast data, the regional air quality assimilation systems developed and tested within EDA, as well as the chemical boundary conditions provided by GDA.  The data assimilation and forecast systems will also be used for building up a posteriori validated assessemnt services (EVA).

Every day, forecasts for each hour out to 96h ahead and analyses for the day before (with possibly more frequent updates, if timeliness of the NRT observational dataflow permits) will be produced by the seven designated centres.  The results will be processed centrally for computing ensemble and verification products at MF-CNRM and at ECMWF (as back-up), and will be displayed on the regional NRT products MACC-II web platform.  The main products that will be delivered by R-ENS are the following:

  • analyses and forecasts of key atmospheric pollutants (ozone, NO2, SO2, PM2.2, PM10, CO...) in the form of maps and numerical data from seven individual forecast systems and from European ensemble built upon these;
  • products tailored to meet users' common general needs (integrated AQ indicators, surface maps indicating probability of exceeding certain air quality thresholds,...);
  • starting at the beginning of the second year of the project, seasonal pre-operational birch pollen forecasts;
  • verification results in the form of maps and time-series over the sites regularly reporting observed values;
  • quarterley reports on individual and ensemble products verification against observations available in NRT (non-validated);
  • semestrail updates of the dossiers for each individual assimilation and forecast suite and for the ensemble.

The air-quality assessments provided by EVA will be on a yearly basis, and in due course the activity will evolve to provide longer-term assessments as longer-term datasets are built up.  Nebertheless, ENS, which provides near real-time analyses and forecasts of the regulatory air pollutant concentration fields in map form and as gridded data, is aimed at supporting air quality decision makers who have to deal with episodes as they occur.  The maps and data provided by ENS (by WP103 nad WP105 in particular) combine observations and simulations, and can be considered as the best reprsentation of air-pollution patterns over Europe.  They describe the extent of each episode and its evolution on an hour-by-hour and day-by-day basis.  They should help national authorities to assess the local or regional nature of the episode, adapt plans and actions for its management at the national or city level and improve communication to the general public.  These analyses and forecasts can thus be considered as support tools for short-term decision-making at the national level for the implementation of the AQ Directive, and their potential use in addition to the outputs of the EVA and POL sub-projects will be taken into account in MACC-II's interaction with the European environmental policy infrastructure, through WP134 of POL.

Beyond providing the daily services, an important new focus will be on the transition to full operations by the end of MACC-II.  In each of the regional production centres, we will identify points of weakness and bottlenecks, and win will aim to optimise data flows, timeliness and organisation of the implementation of numerical suites.  This is indeed crucial for the transition from routine prototype operational activities to fully operatioanl 24/7 services. The feasibility of an operational service certification (ISO 9001) will be assessed.

In addition, a few new RTD activities will be undertaken on the modelling of primary biogenic aerosol particles (more species of pollen) and of CO2, preparing future extensions of the service portfolio.  Research collaborations between the teams will be encouraged to narrow down current key uncertainties in air-quality forecastuing and to test and devlop new methodologies; as was the case in GEMS and MACC, it is important that regional activities in MACC-II continue to be at the forefront of international research in air quality, discussing and comparing frequently with North American colleagues in particular.  Lastly, as part of verification activities, specific attention will be paid to regional model performance in the Mediterranean area, through two Mediterranean partners running models at higher resolution (MM5-CAMx and a high resolution version of MOCAGE).  This activity will aim at documenting elements that reduce the skill of the core service models in this area and providing insight into possible improvements and the upgrade path; this will be done in linkage with the large CHARMEX research project on the Mediterranean area that involves jointly many European countries.

ENS has been structured into eight work packages, as described below.  Each work package is itself broken down into a number of tasks listed at the end of the following sub-sections.  The description of activities at the task level is presented in the corresponding work-package tables in Part A of this Description of Work.

WP101 ENS: Coordination, link with other project components, downstream services and other users

This work package groups coordination activities of the ENS sub-project.  Three main aspects will be covered.  Firstly, we will interact closeley with several MACC-II sub-projects that provide input data to the production of ENS services: GDA (provision of boundary conditions for selected reactive gases and aerosol components), OBS (NRT observational data flow, both in situ and satellite, for assimilation and verification), EMI and FIR (provision of anthropogenic and time-dependent fire emissions), as well as VAL (validation, covering in particular the evaluation of regional models with respect to altitude).  As sub-project leader, MF-CNRM will follow the work of the different ENS partners, in order to ensure homogeneous progress for the seven individual prototype operational production lines.

A second task will focus on interactions of ENS with EVA and EDA, involving the three respective sub-projects leaders.  Indeed , the three sub-projects rely on the same teams and have several interdependences, which will require frequent coordination.  As was the case in MACC, meetings outside of MACC-II assemblies will generally cover all three sub-projects as the same time.

Lastly, EAA will lead a task on continuous dialogue with key users of ENS services, allowing requests and requirements to be closeley followed.  The coordination activity will ensure a possibility of interactions in short loop, preceding more formal agreements (such as SLAs) and communication organised within the INT sub-project (in which EAA also participates).

ENS 1.1 Coordination of ENS work and interaction with OBS, EMI, FIR, GDA and VAL

ENS 1.2 Coordination and interaction with EDA and EVA

ENS 1.3 Coordination and interaction with downstream services and othr users

WP102 ENS: Model development including extension to CO(2) modelling

WP102 is devoted to the continuous development of models and regular update of the description of each individual suite.  It is recalled here that the assimilation part of the seven individual systems will be provided by the EDA sub-project.  Regular model enhancements (resolution, area covered, parameterizations, emssions updates, use of boundary conditions, etc.) are indeed key to steady progress of service quality.  WP102 is thus a small RTD component that allows teams to keep to the forefront of knowledge in air-quality forecasting research.

Following practices established in MACC, a dossier for each of the configurations of the seven models will be updated every six months and made available on MACC-II website, making an up-to-date reference for the production systems used daily for forecasts (WP103) and analyses (WP105).  This dossier will comprise:

  • the detailed description of the model version and assimilation system used, including a history of changes made;
  • a model development plan indicating the current upgrade and recovery actions decided and their schedule in time;
  • timeliness and completeness statistics of the daily delivery of products (from WP107);
  • model skill scores (from WP106 and WP107).

WP102 is also a background support activity for the ensemble production.  A limited and target research activity will take place on ensemble methodologies, in order to keep improving the range and quality of ensemble based products.  A simplified dossier for the ensemble production system will also be maintained and made avaialble on the MACC-II website.

Lastly, a new development on regional CO(2) modelling will be initiated within MACC-II.  The objective is to build up regional-scale greenhouse gas services, such as CO2 flux inversion.  Due to the high variability in space and time of CO2 fluxes over Europe, it is a priori interesting to investigate potential new regional services, complementing the ones provided for the global scale by GHG.  Also, many users of ENS services, such as national environment agencies, face increasingly the need to consider pollutants and greenhouse gas jointly (for instance to see combined effects of emission-abatement strategies), which further substantiate the interest of such an activity.  In MACC-II effort and ambitions will be limited to direct CO(2) modelling (but including both anthropgenic and biogenic surface fluxes), regionalmodel intercomparisons and confrontationwith the global models from GHG.  The added value of a regional CO(2) activity and expected benefits of developing new corresponding services beyond MACC-II (ENS but also and maybe primarily EVA) will be assessed.

ENS 2.1 Continuous development of the CHIMERE model

ENS 2.2 Continuous dvelopment of the EMEP model

ENS 2.3 Continuous development of the EURAD-IM model

ENS 2.4 Continuous dvelopment of the LOTOS-EUROS model

ENS 2.5 Continuous development of the MATCH model

ENS 2.6 Continuous development of the MOCAGE model

ENS 2.7 Continuous development of the SILAM model

ENS 2.8 Continuous development of ensemble processing

ENS 2.9 Joint modelling activities (including regional CO(2) and process-oriented studies

WP103 ENS: Air-quality forecasts

This task corresponds to the prototype operational provision of forecasts for key air-quality compounds with the seven individual model suites, which are developed and constantly updated within WP101.  Forecasts, up to 96h (an extra 24h forcast comapred to the situation in MACC is necessary to accomodate users' needs that require to have 72h of forecasts just after 00 UTC daily, in practice too early for the new daily forecast to be available) will be provided for a range of molecules and vertical levels in GRIB2 and NetCDF format.

  Core productsTarget list of additional products for boundary conditions
Molecules or aerosol component O3, NO2, SO2, CO, PM10, PM2.5 OH, NO, NO3, N2O5, HNO3, HCHO, PAN, isoprene, terpenes, total NMVOCs, SO4, NH3, NH4, BC, OA, NA (fine, coarse), DUST (fine, coarse)
Vertical levels above model ground (m) 0, 500, 1000, 3000 100, 250, 750, 1500, 2000, 2500, 3500, 4000

*********** add in table here *****************


As a baseline, only core products will be provided as part of the prototype operational service, as has been the case in MACC.  These core products will be verified at the surface on a daily NRT basis (WP106 and WP107); note also that the VAL sub-project has a complementary validation activity for above the surface (IAGOS aircraft, soundings, GAW/EMEP European high-altitude sites), but not necessarily in NRT and with a rather limited coverage of Europe.  The target list of aditional species covers those that are candidates for extension of the prototype operational service; they will be progressively implemented in test mode, and interaction with key users (WP101) will be needed to confirm usefulness.

Task 3.8 corresponds to ensemble processing, preformed at MF-CNRM: the individual model forecasts will be gathered and interpolated on the same higher resolution grid (0.1), and then ensemble processing techniques wil be applied to provide hourly ensemble forecasts over the range of 96h for the same species and vertical levels as the individual model forecasts.

ENS 3.1 European-scale prototype operational forecasts with the CHIMERE model

ENS 3.2 European-scale prototype operational forecasts with the EMEP model

ENS 3.3 European-scale prototype operational forecasts with the EURAD-IM model

ENS 3.4 European-scale prototype operational forecasts with the LOTOS-EUROS model

ENS 3.5 European-scale prototype operational forecasts with the MATCH model

ENS 3.6 European-scale prototype operational forecasts with the MOCAGE model

ENS 3.7 European-scale prototype operational forecasts with the SILAM model

ENS 3.8 European-scale prototype operational ensemble forecasts

WP104 ENS: Pollen forecasts

Biogenic particulate matter (pollen grains, enzymes and other genetic material released by plants, spores, etc.) contribute significantly to primary particulate matter mass, affect its size distribution, and interact with other pollutants.  Due to their seasonal character (the highest concentrations appear in spring, summer, and early autumn), they affect also the temporal variation of primary PM in the atmosphere.  The introduction of bioaerosols as intrinsic components of particulate matter in Europe is expected to improve significantly the forecast of PM in the region.  Also, it is recognised that it is necessary to consider exposure to both pollutants and bioaerosols in health impact studies.

Traditionally, pollen forecasts are based solely on local observations and do not consider transport of pollens from other regions, while measurements have shown evidences of transport of some bioaerosols over thousands of kilometers.  Following other recent pioneering work, in particular in the context of PROMOTE, a modelling component of primary biogenic aerosol particles (PBAP) has been started with MACC, with a view to extend in fine the service portfolio of regional services.  A flexible emissions subroutine has been developed as part of the SILAM model and has been coupled to several ENS forecasting systems.  Efforts focussed first on birch pollen.  It is expected in MACC-II to consolidate the coupling of all ENS models with the PBAP emissions routine developedand tested by FMI and, starting in the second year of the project, to demonstrate daily experimental birch ensemble forecasts during the pollinic season (with display on the MACC-II regional platform, WP107).  In parallel, FMI will carry out further developments, focussing on three new species affecting importantly health in Europe: grass, olive and ambrosia.

In addition, we will interact with potential users of the system and its products: national and international bodies responsible for allergenic forecasting and preparatory measures over specific regions, public-health authorities, health-care professionals and organisations, and the general public.  MACC-II to will link to them, in particular in the context of the COST ES0603 action (led by FMI, Dr. Mikhail Sofiev), ensuring no duplication of work.  For the work on PBAP, important and active collaboration will be maintained with the expert groups of the university of Turku (Dr. Hannah Ranta) and of the Medical University of Vienna (Pr. Sigfried Jaeger), especially regarding validation aspects.

ENS 4.1 Primary biogenic aerosol particle (PBAP) emissions module development

ENS 4.2 ENS regional models development for PBAP forecasting

ENS 4.3 Seasonal birch pollen pre-operational forecasts

WP105 ENS: Air-quality analyses

This task corresponds to the prototype operational provision of analyses for key air-quality compounds with the seven assimilation suites (developed within EDA).  Hourly analyses for the the day before will be delivered daily.  At first, this will be done once per day for the day before as in MACC.  If the timeliness of the observational dataflow permits, analyses could run only some hours after real time, in order monitoring maps faster and progressively.  The availability of individual analyses will directly condition the availability of the ensemble analysis products (task WP105.9).  Lastly, improved timeliness of observations and earlier running of the analysis stream will allow study of the impact and benefit of using analyses to base forecasts (instead of using simply the +24h forecast of the day before).

ENS 5.1 European-scale prototype operational analyses with the CHIMERE model

ENS 5.2 European-scale prototype operational analyses with the EMEP model

ENS 5.3 European-scale prototype operational analyses with the EURAD-IM model

ENS 5.4 European-scale prototype operational analyses with the LOTOS-EUROS model

ENS 5.5 European-scale prototype operational analyses with the MATCH model

ENS 5.6 European-scale prototype operational analyses with the MOCAGE model

ENS 5.7 European-scale prototype operational analyses with the SILAM model

ENS 5.8 European-scale prototype operational ensemble post-treatment of analyses

ENS 5.9 Joint work on the impact of assimilation on air quality forecasts

WP106 ENS: Evaluation of models and products

WP106 corresponds to the RTD component of individual model and ensemble products verification, and includes three main activities.

The first task is to update the prototype operational verification methodology, depending on the observational dataflow available and taking advantage of increasing knowledge on site types and from past periods.  This methodology will also closely follow research advances on air-quality model verification in the international community.

The second task corresponds to a specific evaluation work in the mediterranean area.  Indeed, many large cities are located at the coast and are prone to sever photochemical pollution spisodes.  Strong anthropogenic and biogenic emission, very active photochemistry, long-range transport and complex meteorological dynamics combine to make air-quality forecasts and hindcasts particularly challenging.  On top of routine surface monitoring, air quality in these cities has been the object of recent international measurement campaigns (Barcelona, Athens, Marseilles, ...) that have highlighted the complexity and shortcomings of current modelling tools, including ENS models.  Also, there has been much less validation and evaluation of routine air-quality forecasts for the Mediterranean than for nort-western Europe, and there is still much to learn in assessing in detail the preformance of the ENS models over the region through comparrison with observations and more dedicated models - an activity that was already started in MACC (task 0-INT2.1).  This work will contribute to the evaluation of the ENS products with a sytematic feedback regarding the skill of ensemble and individual models in this particular region.  Two modelling teams from Greece and Spain will study the specificities and current uncertainties regarding deterministic air quality forecasting in the Mediterranean area, focussing respectively ion the eastern and western Mediterranean.  The Mediterranean partners will contribute to the individual model and ensemble evaluation, using in particular higher resolution models, observations and their local expertise.

The third task corresponds, for each of the seven individual models and for the ensemble, to contribute every trimester to the quarterly model verification report, by adding comments (possibly on the basis of in-depth analysis and/or sensitivity tests) to the skill scores that are computed centrally (WP107).  Experience from GEMS and MACC regarding each model (with skill varying with season, daytime or night-time) will also be used to follow the evolution of performance.  This evaluation will be complemented by the analysis performed within the VAL sub-project, which will cover validation of the chemical boundary conditions (global components of MACC-II) and validation of regional model forecasts and analyses with respect to altitude (GAW sites, soundings and MOZAIC-IAGOS aircraft profiles).  For this purpose, ENS nad VAL partners will meet regularly at MACC-II assemblies or in ad hoc meetings.

ENS 6.1 Evaluation methodologies of individual and ensemble forecasts and analyses

ENS 6.2 Evaluation of ENS models in the Mediterranean area

ENS 6.3 Evaluation of CHIMERE forecasts and analyses

ENS 6.4 Evaluation of EMEP forecasts and analyses

ENS 6.5 Evaluation od EURAD-IM forecasts and analyses

ENS 6.6 Evaluation of LOTUS-EUROS forecasts and analyses

ENS 6.7 Evaluation of MATCH forecasts and analyses

ENS 6.8 Evaluation of MOCAGE forecasts and analyses

ENS 6.9 Evaluation of SILAM forecasts and analyses

ENS 6.10 Evaluation of ensemble forecasts and analyses

WP107 ENS: Display and dissemination of products and verification

WP107 will focus on the centralised part of the regional services.  It will involve mainly CNRM-MF, but with the collaboration of the EDA (RIUUK) and EVA (INERIS) leads.  The tasks will be to:

  • concentrate individual and ensemble model forecasts (WP103) and analyses (WP105), birch pollen forecasts (WP104) and high-resolution model outputs in the Mediterranean (WP106);
  • retrieve surface observations provided by OBS;
  • carry out verification, compute statistics and prepare graphics;
  • maintain and further develop the NRT regional product platform (integrated in the MACC-II platform), responding in particular to feedback from users (WP101.3, but also direct feedback received on the platform via a "contact us" option);
  • set-up more easy and user-friendly access to numerical data of the regional core services;
  • interact with the European Environment Agency to meet the evolving needs of the Eye-On-Earth application;
  • interact with PASODOBLE and other downstream services to meet their evolving needs;
  • prepare verification and assimilation statistics (departures, blacklists, etc.) for feedback to OBS on observation quality.

Being at the end of the chain, the ENS platform has a considerable importance on the visability, dissemination and reputation of MACC-II NRT regional services.  Thus major efforts will be devoted to making this platform both ergonomic (so that an occasional user will not be lost under massive loads of information) and comprehensive (so that expert users can find all needed information on the services, their quality, availability and timeliness).

ENS 7.1 Continuous development of the ENS products platform

ENS 7.2 Routine operation of the ENS products platform

ENS 7.3 Provision of numerical data to users

WP108 ENS: Transition to operations

WP108 is a new component compared to MACC: it is aimed at specifically organising the transition from prototype operational services, which requires end-to-end analysis for the seven individual production lines and for central processing of the NRT regional core services for the atmosphere.

The first year of MACC-II will be devoted to the analysis of the existing procedures set up by the different regional partners: the 8 production chains (7 individual models, central processing and ENS platform) will be described in full detail, assessing the level of operationality of each step and the corresponding risks for failure.  In particular, these activities will establish operationalization plans, compatible with the timing and resources of MACC-II.

The rest of the project will be devoted to the implementation of these operationalization plans, with the objective being ready for full operational status at the end of MACC-II.  During the last year, a task will study the opportunity of a certification (ISO9001) of the regional services, and document the steps and effort needed at each of the partners will be of benefit to the regional services production quality, compensating for the fact that the reqional services are effectively based upon a network of collaborating centres

ENS 8.1 End-to-end analysis of the individual and ensemble systems

ENS 8.2 Implementation of the operationalization plan for CHIMERE

ENS 8.3 Implementation of the operationalization plan for EMEP

ENS 8.4 Implementation of the operationalization plan for EURAD-IM

ENS 8.5 Implementation of the operationalization plan for LOTOS-EUROS

ENS 8.6 Implementation of the operationalization plan for MATCH

ENS 8.7 Implementation of the operationalization plan for MOCAGE

ENS 8.8 Implementation of the operationalization plan for SILAM

ENS 8.9 Implementation of the operationalization plan for ensemble and verification

ENS 8.10 Preparation for service certification