EuMetChem - COST Action ES1004

AQMEII & EuMetChem Workshop
8 May 2012, Utrecht, The Netherlands

Workshop was organized for setting up the second two continent (North America nd Europe) model evaluation activity focused on regional scale coupled meteorology/ air quality modeling systems. The workshop started from welcome speech (S. Galmarini, C. Hogrefe) to participants with explanation of motivation of the initiative.

09:05   Introduction                                                                                      
(S. Galmarini, C. Hogrefe)
Key Questions Facing the Climate Change / Air Quality Modelling Community (look for Alapaty et al. 2012 article); Challenges in addressing these questions; Moving to a coupled paradigm for regional models; Related activities in EU and NA;
+
Discussions on usefulness of Activity; win-win situation?, benefits, management (on how to sell the product) vs scientists (on a cutting edge of  research);  needs for AQ vs NWP communities;

09:55: A Summary of the AQMEII Phase 1 activity                                
(Efisio Solazzo)

Model activity and framework; Model predicted concentration and deposition; Ensemble system for evaluation of modelling results; List of models and countries in modelling community; Model output and variables: meteo + chem, hourly, monthly and quarterly data/fields; Monitoring networks (AIRS, CastNet, IMPROVE, NAPS, CapMON, AirBase, EMEP) with hourly, daily, weekly data; Spatial distribution of monitoring stations in EU and NA;  Check AE paper of 2011 Galmarini and S.T.Rao (check all listed papers at AQMEII website) Results from collective analysis; AQMEII and MOSAIC monitoring data from commercial air flights (from EU to NA; 11 airports in NA + 2 in EU); Results of comparison between the models and for vertical profiles of chem. species and meteorology, error correlations, coefficients of variations, fractional absolute difference - in airports; Comparison for NA vs EU domains for meteorology - wind speed, air temperature, relative humidity and precipitation; for species – dry and wet deposition, PM, PM10 (monthly averaged time series), comparison urban vs rural stations (error vs MFBias), PM precursors (NO2, SO2) (error vs MFBias), PM2.5 by model domains, Taylor diagram for correlation coefficients (for rural vs urban locations); PM components – NH4, SO4, etc.; comparison by sub-domains in EU vs NA for bias vs error (sum 2006).

10:45 – 11:15 Coffee Break

Look for poster presentations at ITM Conference

11:15: Ongoing analysis of AQMEII Phase 1 results                                
(C. Hogrefe)

Use of synoptic types for analysis;
FOR NA: identified 5 patterns (wind speed, sea level pressure, temperature, cloud fraction, observed anomalies) in 2006 (criteria: cor > 0.5, minimum 10 days) for ozone levels; analysis model bias (8hr daily mean O3) by patterns – May-Sep for NA – for different NA regions;
FOR EU: similar analysis for dominating patterns
Synoptic typing (was also done for UHam MEGAPOLI study, with LOTUS-EUROS, with Iberian Peninsula) helps to identify subsets of days with regionally homogeneous anomalies in temperature, wind speed and ozone; next step – expand analysis to PM2.5,

11:45: Presentation of Phase 2 – Introduction                                          
(S. Galmarini)

Selecting years (2010, 2006 – to leverage on existing database of model and measurements, to analyze the most recent year available, to harmonize with global community goals); identifying relevant events which could lead to a clear signal on feedback (forest fires, dust events, …); identifying processes of interest for classifying models and levels of coupling, variables to be requested to diagnose efficiency of processes of interest, collecting efficiently monitoring information for evaluating purposes;

11:55:  Collaboration with COST / ES-1004 EUMetChem                                  
(K.H. Schlünzen)

Background of the COST Action; Objectives; Main aim; Main topics; Working groups 1-4; Participating countries (already signed vs. non-signed yet; joint test case wanted by most participants (should allow to access feedbacks meteo/chem., models should be documented in model inventory, joint with AQMEII to share work (and avoid double work), details on processes interesting to EuMetChem in talk by D. Brunner);

12:00:  Joint Exercise
(S. Galmarini)

AQMEII coordinating group (consists of the coordinators; variable/process definitions; boundary conditions; two continent emissions; monitoring data, ENSEMBLE system; satellite observations)

12:10:  Selection of Case studies
(Paul Makar, S. Galmarini)

NA: initial criteria and episodes for NA case studies in 2006/2010 (from CARES DOE-NASA, CalNEX NOAA-NASA, TexAQS, BORTAS, IONS-06/ INTEX-B, etc.); NA surface monitoring networks (ozone, PM2.5, etc.); satellite info on forest fires; Saharan dust transport to SE USA; air flights HSRL system and data; also forest fires in Canada with plumes transported to USA; comparison with measurements fro PM2.5, O3; most important criteria; good satellite converge? Alignment with field campaign data?; what additional datasets are needed to be explored for selection;
EU: case studies post economic crises, harmonization; Summer 2010 Russian forest fires; Oct & Jul 2010 – Saharan dust events; domain size: lon 44W-65 E & lat 22-72 N, 0.2 deg resolution 543/545+1? x 250+1? grid points; volcanic eruption? In Iceland (what about emissions from air planes – traffic banned for several days);

12:45:  Coupled Processes and Variables of Interest                                            
(D. Brunner)

Coupled processes – effects of meteo on chem. (mostly considered in offline, potentially more accurate treatment in online) & effects of chem. on meteo (usually not considered in offline models, or only at synchronization intervals); interaction chains; Examples effects of Met on Chem for wind, temperature, cloud water, radiation; other meteo (H2O, soil moisture, lightning); Examples effects of Chem on Met for aerosols (listed direct, indirect, semi-direct effects); & radiatively active species (LW absorption + SW absorption by O3 in stratosphere); Common / global & EU & NA evaluation data (dataset/ network, description, parameters meteo. & chem.); summary of variable output – 2D fields, station output, profiles - for gas-phase species & aerosol parameters and meteorological parameters, also wet and dry deposition parameters (monthly), and emissions (quarterly);

13:20-14:40    Lunch Break

Look for poster presentations at ITM Conference

14:40:  Emission Inventories                                                           
(H. van Der Gon, M. Sofiev, C. Hogrefe)

EU, TNO: Anthropogenic emissions - from TNO, and Forest fires – from FMI; using “almost 2010” (=2009); see also George Pouliot et al. AE-2012 paper; plans in MACC-II for EU emissions; extend emission data to include 2008-2009; consistency in base year reported EI data; emission trends (but remember about technology changes and economical crisis) by country (EU+15+NO CH, EU12+HR, non-EU) groups; chemical composition of MACC PM emissions; SUMMARY: emissions data are expected to be available by end Jun 2012; Is the PM split useful? Use old or new 2006 EI; Updates based on further work in MACC will follow; TNO thinks that PM OC emissions are underestimated at present in EU, sensitivity of coupled system for 20-30% PM2.5 OC would be interesting?
EU, FMI: Emission from wild-land fires; IS4FIRES systems of FMI, fire emission database, fire emissions diurnal cycle, injection profile of wild-land fires emission; fractionation of the fire energy empirical formula for the total rate of emissions of FRP (using satellite data); structure of the integrated system for fires (IS4FIRES); fire emission database available from 2010+ with 10 km resolution and daily; SEVIRI – source of diurnal variation data; estimate top of the plume rise (new formula suggested); injection profile, zonal average; summary: available data – emission of PM, scaling to gaseous species, diurnal variation of the fire intensity, characteristics injection height.
NA, EPA: US emission trends (NOx - 28%, VOC - 24%, SO2 - 40%, PM2.5 - 14% reduction from 2006 till 2010); emissions supplied by 15 sectors; emissions processing (performed using SMOKE system; temporal allocation, speciation, spatial allocation – 12 x 12 km); point + biogenic + other sources; for 2010 – US power plant emissions, wildfires, mobile source emissions, other US sectors – projection from 2008 inventory, Canada – new 2009 inventory for Canadian emissions;
NOTE: there are no EU groups which will use NA emissions to make runs on NA domains

15:45:  Lateral Boundary Conditions Data                                               
(N. Savage)

Time period and frequency: 1 Dec 2005 – 31 Dec 2006; 1 Dec 2009 – 31 Dec 2010; temporal resolution - 3 hourly; netCDF format, vertical levels - all native grid levels from surface to 2.5 hPa, regular 1.125 deg lon-lat grid; EU bounding box of domain – 80E, 56W, 77 N, 16 N; gas phase chemical species (O3, CO, NO, NO2, NHO3, PAN, CH4, C2H6, ISOP, etc.); aerosols (sea salt and desert dust by size distribution in 3 bins, sulfate, etc.); supporting physical data – temperature on model levels and surface pressure (allow calculation of P on all model levels as data on a hybrid pressure grid); expected be ready during Sum 2012

15:50:  Grids, Formats and Database Infrastructure                               
(S. Galmarini)

Adoption of the same strategy as in AQME phase 1 (documents will be provided on how to proceed); Specifications: OAD over arching document / summary of data strategy; TSD technical specific documents, specific for any data format; distribution of software for data formatting in ENSEMBLE format;

Tentative schedule for phase 2:
EU domain emission inventories will became available in sum 2012; coupled models will be run for both years (2006+2010) to discern whether models are able to accurately simulate direct and indirect effects of aerosols on radiation; US DOE + US EPA + EU sponsored workshop to review state of the science in modelling interactions among aerosols, clouds, radiation precipitation (US, North Caroline, Oct 2012); AQMEII Phase 2 workshop – held in Sep 2013 (simultaneously with ITM Conference); research paper based on model comparison results; Lessons learned from AQMEII Phase 2 and follow-on activities will be discussed at workshop (jointly with NATO ITM in May 2015);

16:00 – 16:25 Coffee Break

Look for poster presentations at ITM Conference

16:25: Timeline of Phase 2 activity/                                                           
(S. Galmarini)

Sum/ Fal 2012 –  In reality model simulations could start (probably) after sum/fal and need to be delivered for evaluation by Jan/Feb 2013, note data submission may be staggered (gridded first, receptor locations – later);
May/Jun 2012 – Collective analysis at JRC starts;
Sum 2012 – Satellite, ozonesonde and MOZAIC  data ingesytion for ENSEMBLE;
Sep/ Oct 2012 – Collection of NA 2010 network data; key: finalize receptor locations – actual observations could be updated later until collective analysis starts;
Spr 2013 – Start negotiations with Atmos. Envir. special issue;
Spr 2013 – Workshop with COST, Mar?; or HARMO, Madrid, May?
Sep 2013 – Workshop;

Action Items:
Circulate Dominik spreadsheet to community to update processes/ variables/ datasets that may have been missed; finalize by end May 2012 (Copenhagen, EuMetChem meeting, CPH, DK); then will be added to model inventory (UHam); then modelers will be asked to then “describe” their model in the UHam database using this list of processes so we know what is in each model (could this also include each group nudging approach? – sensitivity simulations will be treated as separate models – modelers need to finalize their descriptions, otherwise their documentation is not visible – set same deadline as for data submission – each model submitting data to ENSEMBLE needs to be documented in the UHam database);
Create over-arching document (by Stef. Galmarini; end May/Jun 2012);
Monthly calls will be continued, at least, once per month till emissions are ready (then summer break);
Create workshop minutes, circulate to community (Christian/ Stefano)

17:15 Adjourn of the workshop