Objective 2: Production of Lightning NOx ... Quantif

1
Effects of Urban-Influenced Thunderstorms on
Atmospheric Chemistry

• Kenneth E. Pickering
• Department of Meteorology
• University of Maryland
• HEAT Planning Workshop
• March 15, 2004

2
Outline

• Background chemical measurements, modeling for
  deep convection, urban plumes, lightning NOx
• HEAT proposed objectives, measurements,
  modeling strategies
• Possible activities

3
Effects of Deep Convection

• - Venting of boundary layer pollution
• - Transport of NOx, NMHCs, CO, and HOx precursors
  to upper troposphere
• - Downward transport of cleaner air
• - Transported pollutants allow efficient ozone
  production in upper troposphere
• - Results in enhanced upper tropospheric ozone
  production over broad regions
• Increased potential for intercontinental
  transport
• - Enhanced radiative forcing by ozone

4
Effects of Deep Convection

• - Lightning production of NO
• - Perturbation of photolysis rates
• - Effective wet scavenging of soluble species
• - Incorporation of pollution aerosols into
  precipitation processes
• - Nucleation of particles in convective outflow
• - In remote regions low values of O3 and NOx are
  transported to upper troposphere
• - Larger values of these species tranported to
  PBL where they can more readily be destroyed

5
Aircraft Measurements of Trace Gas Redistribution
in
Oklahoma PRESTORM June 15, 1985 MCC
CO
O3
Dickerson et al., 1987, Science
6
Pickering et al., 1990
7
Pickering et al., 1990
8
(No Transcript)
9
Kansas-Oklahoma Squall Line Cell
Goddard Cumulus Ensemble (GCE) Model with offline
tracer transport
10
June 10-11 PRESTORM Initial Conditions

• Altitude CO (ppbv) O3 (ppbv) NOx (pptv)
• 0-1.75 km 150 (245) 28 (64) 900 (2950)
• 1.75-2.5 135 28 607
• 2.5-5.0 106 35 280
• 5.0-8.1 67 43 97
• 8.1-10.3 76 60 218
• 10.3-trop. 65 75 308
• Urban BL values in parentheses Pickering et al.
  (1992)
• Representative of 45 km downwind of Oklahoma City

11
Vertically-averaged Ozone Production in Cloud
Outflow

• June 10-11 PRESTORM (4-15 km)
• Cloud-
• Undisturbed Processed
• Rural air 2.7 5.7 - 6.2
• Urban plume 2.7 9.4 - 9.9
• Values in ppbv/day Pickering et al. (1992)

12
LW Radiative Forcing - Clouds
LW Radiative Forcing - Clear
13
The effect of thunderstorms on local O3 can
be remarkable even at periphery of storm.
14
On the third day of a high O3 episode (June 24-26
1998), a line of thunderstorms passed just north
of the Fair Hill, MD monitor.
15
Production of NO by Lightning

• - Global production estimates range from
• 2 to 20 Tg N/yr due to uncertainty in global
  flash rate and in the production per flash
• Global flash rate estimated from OTD satellite
  measurement 44 flashes/s (Christian et al.,
  2003)
• Production per flash estimated from analysis of
  NO spikes in aircraft measurements, cloud-scale
  chemical transport modeling, or mass flux
  techniques
• Cloud-scale chemical transport models represent
  lightning either through explicit electrophysics
  or use of observed/parameterized flashes
• Models addressing other important questions
  production per CG flash vs. production per IC
  flash vertical distribution of lightning NOx at
  storm dissipation

16
July 12, 1996 STERAO-A Storm NE Colorado
17
(No Transcript)
18
Cloud-scale Chemical Model Results - July 12,
1996
Transport Only No chemistry or NO from Lightning
DeCaria et al., 2000
19
CG 460
CG 460
IC 345
IC46
Moles NO
Per Flash
CG 460
CG 460
IC 460
IC 690
Model-simulated vs. Measured NOx Profiles
For Four Lightning NO Production Scenarios
DeCaria et al. (2000)
For a 30-km flash, 460 moles NO/flash 1 x 1022
molec/m
20
(No Transcript)
21
Ozone Production for 24 hours Following Storm
42 x 42 km anvil region
Entire model domain
22
EULINOX - July 21, 1998
a.
b.
(Huntrieser et al., 2002).
23
1630 UTC
1653 UTC
Original Cell
Cell Splitting
1803 UTC
1734 UTC
Multicellular
Supercell
(Höller et al., 2000).
24
(No Transcript)
25
(No Transcript)
26
(No Transcript)
27
PCG PIC 250 moles/flash gives best agreement
with Falcon measurements at 8.5 km of mean NOx
3 ppbv
28
(No Transcript)
29
With 3 ppbv NOx in UT, ozone production is less
efficient than in STERAO-A case with 1.2 ppbv
Max. ?P(O3) 4 ppbv versus 10 ppbv in STERAO-A
case
30
HEAT Objectives

• Characterize and quantify convective transport of
  urban pollution from BL to UT
• Quantify lightning production of NOx
• Examine effects on UT chemistry (e.g., O3, HOx
  production)

31
Objective 1 Convective Transport

• Study transport and fate of urban pollutants
• Examine relative importance of convective
  motions, scavenging, and chemistry
• Measurements required vertical profiles of
  chemical mixing ratios before, during and after
  storm (CO, NO, NOx, NOy, O3, SO2, HOx, HC,
  peroxides, aldehydes, acetone, aerosols)
• Characterize inflow, outflow, and storm core (?)
• U. of WY King Air low level inflow, outflow
• WMI Lear Jet anvil outflow
• Chemical analysis of precip from mesonet
• CO, CO2 as tracers of air motion in storm

32
Objective 2 Production of Lightning NOx

• Quantify amount of NO produced per flash, per
  meter of flash channel, per thunderstorm, by
  different storm types
• Quantify amount produced by an IC flash vs. that
  produced by a CG flash and by different
  components of a flash
• Measurements required NO, NOx, NOy in low level
  inflow/outflow, in anvil outflow, and in storm
  core (?). Channel lengths and distributions from
  lightning mapping system, CG flashes from NLDN
• Analysis of flash and aircraft NO spike meas.
  chemical transport modeling mass flux analysis

33
Objective 3 Effects on UT Chemistry

• Examine effects of combination of pollution and
  lightning NOx on UT O3 and HOx chemistry
• Quantify relative contributions of boundary layer
  and lightning NOx to UT NOx mixing ratios
• Chemical transport modeling required
• To verify these models, chemical measurements
  needed in convective outflow plumes hours to days
  downstream

34
Possible Post-Mission Analysis and Modeling
Activities

• Analysis of relationships between flash data and
  observed NO spikes
• Cloud-resolving model simulations of chemical
  transport, wet scavenging, lightning NO
  production (parameterized, explicit) comparisons
  with measurements and between models
• Tests of convective transport and lightning
  parameterizations in regional models calculation
  of downstream ozone production