Monitoring atmospheric composition & climate
 
 
VAL

The validation (VAL) sub-project is central to the global services proposed for MACC-II.  It has strong interaction with the GRG, AER, GHG and GDA sub-projects, and interactions also with the ENS, EVA, FIR, EMI and OBS sub-projects to ensure routine documentation of the quality of the global services with contributions to the regional air quality sub-projects.

VAL is a new sub-project: there was no seperate validation project in MACC, where the validation work was integrated in the different sub-projects.  The aim of the new VAL project is to set up a unified, operational and user-orientated approach to validation.  By clustering the validation work in VAL the validation will become more independent from the modelling work.  VAL will nevertheless facilitate a smooth transition by continuing the validation activities from MACC.

Validation has to reach an operational status, which implies an automated production of verification/validation data and plots, and updats of validation reports with fixed intervals.  The ENS 3-monthly QA-QC dossiers are an example of this.  There are several considerations based on the experience in MACC:

  • The focus of VAL will be on the global sub-projects GRG, AER and GHG.  The production and presentation of validation results will be furthr harmonised.  Othr sub-projects will be involved to unify scoring approaches and presentation when appropraite.
  • Forecast validation has been quite limited, and will be explicitly  dealt with in VAL.
  • The collaboration between ECMWF routine monitoring activities (GDA) and the  validation teams will be intensified.  ECMWF is a partner in VAL.
  • Communication with the othr sub-projects is essential for VAL.  VAL partners will be encouraged to contribute to relevant sub-project meetings and sessions at the assembly.

The main goal for the VAL sub-project can be summerised in the following way: Every service, forecast and reanalysis should be accompanied on the MACC-II website by a validation report which provides essential information for users.  These validation reports should be regulary updated and should document the quality of the latest model/analysis version.  The work plan has been set up following the principle, and individual work packages are linked to MACC-II services (VAL.2 and VAL.3).  Apart from these core validation reports, VAL will also provide a set of deicated websites hosted by VAL partners with more detailed operational comparisons between the MACC-II service products and individual validation datasets.  Furthrmore verification statistics will be produced in an operational way, in collabration with GDA, and these too will be made available on the MACC-II website.  Quick-look validation work will be performed to provide feedback to the modelling teams in case of updates.  Inaccuracies in the fire emission and other input (from FIR and EMI) will be qualified based on their effects in the various global and regional atmospheric analyses and forecasts.

val_structure.png

Figure 10 VAL activities and interactions with other sub-projects

VAL will provide supplementary validation results for the regional air quality cluster.  In particular it will focus on concentrations above the surface and on the comparison of the global and regional scale analyses.  These validation results will be exchanged with the EDA, ENS and EVA sub-projects.

A  seperate work package is devoted to validation research.  This includes validatyion case studies, which will target both large international measurement campaigns as well as sudden large-scale events.  Examples of campaigns are POLARCAT, PEGASOS and SEAC4RS.  Examples of sudden events are the eruption of the Eyjafjallajokull volcano in April-May 2010, the Russian fires in July-August 2010, and the major intrusions of Saharan dust that occur from time to time.  When such events occur other sub-projects will be contacted to initiate rapid-response actions.  The validation methodology will be improved and documented in collaboration with other sub-projects.  In particular there is a need to define scores which can document the improvements of the system over time, and with respect to other modelling systems.  Two promising measurement networks, ceilometer and MAX-DOAS, will be studied and recommendations will be made for future operational use.  Links will be established with the NORS project, both for access to the NDACC ground-based remote-sensed validating data that will be provided by NORS, but also to draw on the results from the complementary validation of MACC-II's products that will be carried out by NORS.

To acheive its objectives, VAL has been structured into five work packages.  Each work package is broken down into a nymber of tasks:

WP81 VAL: Coordination

VAL.1.1 Coordination of VAL work

VAL.1.2 Interaction with the global, regional and emission sub-projects

WP82 VAL: Validation of the global atmospheric monitoring and forecasting service

VAL.2.1 Validation report

VAL.2.2 Validation websites

VAL.2.3 Real-time verification statistics

WP83 VAL: Validation of the reanalysis of the global atmospheric composition

VAL.3.1 Validation report

VAL.3.2 Validation websites

WP84 VAL: Supplementary validatyion of the regional air quality services

VAL.4.1 Supplementary provision of validation results for the regional air quality services

WP85 VAL: Validation research: Improvement of methodology and datasets, and case studies

VAL.5.1 Improvement of the validation methodology

VAL.5.2 Document on scoring methods

VAL.5.3 Analysis of special events

VAL.5.4 Ceilometer and MAX-DOAS