CITYDELTA WORKSHOP MATERA FREE UNIVERSITY BERLIN CONTRIBUTION: LONG TERM MODELLING OF OXIDANTS AND A

1
CITY-DELTA WORKSHOP MATERA FREE UNIVERSITY
BERLIN CONTRIBUTION LONG TERM MODELLING OF
OXIDANTS AND AEROSOLS IN THE BERLIN-BRANDENBURG
AREA CD II PRELIMINARY RESULTS

• R. Stern, A. Kerschbaumer
• Free University Berlin
• Institute of Meteorology
• Carl-Heinrich-Becker-Weg 6-10, D-12165 Berlin,
  Germany

2
REM/CALGRID (RCG)COMBINED 3-D REGIONAL/URBAN
MODEL

• TRANSPORT IN THE LOWER TROPOSPHERE
• PHOTOCHEMICAL TRANSFORMATIONS (CB4-Mechanism)
• Formation of O3 and other oxidants
• INORGANIC AEROSOL FORMATION (ISORROPIA-Module)
• Primary particles, fine and coarse mode PM2.5
  (EC, OC, Other) , PM10-PM2.5
• Secondary inorganic aerosols SO4, NO3, NH4, Na,
  Cl
• ORGANIC AEROSOL FORMATION (simplified
  SORGAM-Module of FORD)
• SOA production caused by anthropogenic and
  biogenic VOC-precursors
• DRY AND WET REMOVAL PROCESSES
• EMISSIONS INTERFACE FOR LONG TERM APPLICATIONS
• On the fly calculations of hourly anthropogenic
  and biogenic emissions, sea salt emissions, and
  wind blown dust (simple, very preliminary
  parametrization based on landuse and windspeed)

3
RCG-APPLICATION STATUS FOR CITY-DELTA II

• 6 Runs EUROPEAN SCALE (EMEP-Emissions) Base Run
  1999, 5 scenario runs, (0.5 lon, 0.25 lat),
  Boundary conditions (BC)
• 8 Runs URBAN SCALE BERLIN Base Run 1999, 7
  scenario runs (ca. 4x4 km2 )
• SCENARIOS
• Scenario 1 CLE 2010
  BC CLE
• Scenario 2 CLE MFR1 (NOX ,SO2, PM) BCMFR1
• Scenario 3 CLE MFR2 (VOC, CO, NH3) BC
  MFR2
• Scenario 4 MFR 2010
  BC MFR 2010
• Scenario 5 CLE MFR1 (NOX ,SO2, PM) BC CLE
• Scenario 6 CLE MFR2 (VOC ,CO,NH3) BC CLE
• Scenario 7 MFR 2010
  BC CLE

4
Large Scale RCG model area

• RESOLUTION 0.25 LATIDUDE, 0.5 LONGITUDE

5
Urban scale modelling domain approx. 300x300 km2
long lat grid , ? long 0.06250, ? lat 0.03125(
approx. 4x4 km2)
6
RCG-Model Configuration

• 5 vertical layers
• a 20 m thick surface layer
• two equal-thickness layers below the mixing
  height
• 2 above the mixing height and extending to the
  domain top at 2500 m.
• Results are for 10 m

7
LARGE AND URBAN SCALE DIAGNOSTIC METEOROLOGICAL
ANALYSIS SYSTEM

• Optimum Interpolation of synoptic and radiosonde
  data
• Vertical Resolution Isentropic surfaces 0 -10km
• Time resolution 3 h
• 3-D Wind, Temperature, Humidity, Density
• 2-D Cloud cover, Mixing Height, Rain, u, MOL,
  surface wind, temperature

8
URBAN SCALE EMISSIONS BERLIN-BRANDENBURG CD II

• Annual emission totals taken from the
  City-Delta II web page
• Time profiles taken from the City-Delta II web
  page
• VOC Profiles taken from the City-Delta II web
  page
• No plume rise for point sources
• Original data resolution ca. 2x2 km2,
  aggregated to model resolution of ca. 4x4 km2
• Major differences to CD I Berlin emissions
• factor of 2 more VOC emissions
• factor of 10 less SO2 emissions
• ca. 20 less PM emissions, ca. 10 less NOx
  emissions

9
(No Transcript)
10
(No Transcript)
11
BOUNDARY CONFIGURATION

• Taken from RCG-large scale run (EMEP-Emissions)
• Identical vertical configuration
• 57 species
• Hourly Update

12
REM/CALGRID TROPOSPHERIC O3 BACKGROUND USED FOR
URBAN AND REGIONAL APPLICATIONS
3-D monthly ozone climatology with resolution of
4 in latitude by 5 in longitude Jennifer A.
Logan Harvard University, Cambridge,
Massachusetts sonde data were used in
combination with surface and satellite data to
derive a gridded climatology for tropospheric
ozone
13
SOME EVALUATION RESULTSBASE CASE O3, NO2, PM10

• 4x4 urban scale CD I versus CD II
• CD II
• New emissions, new time profiles, new VOC
  profiles
• New meteorological data
• Parameterization of wind blown dust as function
  of landuse and windspeed (very preliminary !!)
• SOA added to the model
• New parametrization of sulfate formation

14
(No Transcript)
15
REM/CALGRID SIMULATION O3
16
No of days with max O3 8hmean gt 120µg/m3
17
(No Transcript)
18
REM/CALGRID SIMULATION NO2
19
(No Transcript)
20
REM/CALGRID SIMULATION PM10
PM10 observed 0.8 TSP observed
21
(No Transcript)
22
SUMMARY REM/CALGRID SIMULATION CD II

• Compared to CD I
• Higher O3 peak concentrations (higher
  background, higher VOC emissions in BB area)
• Higher PM concentrations (less emissions than in
  CD I, but SOA added, wind blown dust added)
• Slightly less NO2 concentrations (less emissions
  than in CD I)
• Compared to Measurements (urban and rural, no
  street)
• Overestimation of the observed O3 peak maxima
• Underestimation of observed PM concentrations in
  summer
• Underestimation of observed NO2 concentrations
• NO check whether a station is representative for
  a 4x4 km2 grid !!!!