dry deposition in the United States Sarah Kavassalis and Jennifer - - PowerPoint PPT Presentation

dry deposition in the united
SMART_READER_LITE
LIVE PREVIEW

dry deposition in the United States Sarah Kavassalis and Jennifer - - PowerPoint PPT Presentation

The sensitivity of summer time surface ozone concentrations to dry deposition in the United States Sarah Kavassalis and Jennifer G. Murphy Department of Chemistry, University of Toronto Acid Rain Conference 2015, Rochester, NY., October 22 nd


slide-1
SLIDE 1

The sensitivity of summer time surface ozone concentrations to dry deposition in the United States

Sarah Kavassalis and Jennifer G. Murphy Department of Chemistry, University of Toronto

Acid Rain Conference 2015, Rochester, NY., October 22nd, 2015

slide-2
SLIDE 2

Why study surface ozone?

Pre-industrial surface ozone Present-day surface ozone

(1) The Royal Society, 2008. (2) Gerald Holmes, UDA-ARS Air Quality Program. (3) Kim et al., Applied Optics, 2001.

(1)

Ozone damage to vegetation

(2)

Leaves Grown in Low Ozone Environment vs High Ozone Environment

(3)

Ozone disruption of photosynthesis

slide-3
SLIDE 3

Ozone in CASTNET

Image: U.S. Environmental Protection Agency Clean Air Markets Division Clean Air Status and Trends Network (CASTNET)

slide-4
SLIDE 4

Ozone in CASTNET

Georgia Station (GAS153)

Data from: U.S. Environmental Protection Agency Clean Air Markets Division Clean Air Status and Trends Network (CASTNET)

slide-5
SLIDE 5

Ozone in CASTNET

Georgia Station (GAS153)

Data from: U.S. Environmental Protection Agency Clean Air Markets Division Clean Air Status and Trends Network (CASTNET)

slide-6
SLIDE 6

Ozone in CASTNET

Georgia Station (GAS153)

slide-7
SLIDE 7

Stratosphere- Troposphere Exchange

NO NO2

O3 CO, CH4, NMHCs NMHCs := Non-Methane Hydrocarbons

Processes Controlling Tropospheric O3

Emissions

O3

O(1D)

HO2, RO2 OH HO2 CO, O3

O3 Dry Deposition

O3

Photo- Chemistry

slide-8
SLIDE 8

Stratosphere- Troposphere Exchange

NO NO2

O3 NMHCs := Non-Methane Hydrocarbons

Processes Controlling Tropospheric O3

Emissions

O3

O(1D)

HO2, RO2 OH HO2 CO, O3

O3 Dry Deposition

O3

Photo- Chemistry

CO, CH4, NMHCs

S-T E: Governed by mixing Chemistry: O3 production is non-linear, dependent on emissions and meteorology Deposition: Highly dependent on surface composition

slide-9
SLIDE 9

North American PBL Ozone Budget

Summer-time planetary boundary layer (984–934 hPa) ozone budget over the Southeast and Midatlantic United States (95–75◦W and 28–40◦N) Racherla and Adams, The response of surface ozone to climate change over the Eastern United States, Atmos. Chem. Phys., 8, 871–885, 2008.

12.5 3.7 3.3 5.5

2 4 6 8 10 12 14

Chemical Production Chemical Loss Net transport Dry Deposition

Model Bu Budget (Tg Tg O3/t /three months)

slide-10
SLIDE 10

Ozone (ppb) Temp (° C)

  • Rel. Hum. (%)

Solar . (W/m2)

Data in CASTNET

Georgia Station (GAS153)

slide-11
SLIDE 11

Meteorological Controls on Ozone

  • 1.0 -0.5 0.0 0.5 1.0

p(Relative Humidity,Ozone)

  • 1.0 -0.5 0.0 0.5 1.0

p(Temperature,Ozone)

Summer p(RH,Ozone) Summer p(T,Ozone) Summer (June, July, August) observed midday (12-4pm) Pearson’s correlation coefficient of ozone versus relative humidity a) and b) temperature from 1987 to 2015 at CASTNET stations.

slide-12
SLIDE 12

Temp-O3 correlation

Davis et al., Atmos. Env. , 2011.

Ozone-Met Correlations in CMAQ

May 1st- September 30th Temperature-O3 correlation reasonable well captured by model

slide-13
SLIDE 13

May 1st- September 30th Temperature-O3 correlation reasonable well captured by model Relative Humidity-O3 correlation poorly captured by model.

Temp-O3 correlation RH-O3 correlation

Davis et al., Atmos. Env. , 2011.

Ozone-Met Correlations in CMAQ

slide-14
SLIDE 14

RH-O3 correlation

Park et al., ACP, 2014. Davis et al., Atmos. Env. , 2011.

Ozone-Met Correlations in CMAQ

A comparison of the simulated and observed hourly mean O3 dry deposition velocities. M3DRY is the deposition scheme used in CMAQ. Wesely is a popular alternative.

Relative Humidity-O3 correlation poorly captured by model.

slide-15
SLIDE 15

Ozone-Met Correlations: Role of Deposition?

vpair ~ vpsat

vpair << vpsat

vpsat vpsat

Low VPD High VPD

Georgia Station (GAS153), 2009

Vapour Pressure Deficit (VPD) = vp of H2O in air – saturated vapour pressure

slide-16
SLIDE 16
  • 1.0 -0.5 0.0 0.5 1.0

p(VPD,Ozone)

Ozone-Met Correlations: Role of Deposition?

vpair ~ vpsat

vpair << vpsat

vpsat vpsat

Low VPD High VPD

Summer p(VPD,Ozone)

Afternoon VPD and afternoon ozone are well correlated at most CASTNET sites in the summer. The correlation is stronger on average that that of temperature or relative humidity.

slide-17
SLIDE 17
  • 1.0 -0.5 0.0 0.5 1.0

p(VPD,Ozone)

Ozone-Met Correlations: Role of Deposition?

vpair ~ vpsat

vpair << vpsat

vpsat vpsat

Low VPD High VPD

Winter p(VPD,Ozone)

Afternoon VPD and afternoon ozone are poorly correlated at most CASTNET sites in the winter.

slide-18
SLIDE 18

Modelling Dry Deposition A Resistance Approach

Aerodynamic resistance Quasi-laminar resistance Ra Rb Soil resistance

  • 1. Turb

rbulent tr transport through atmosphere

  • 2. Mole

lecular dif iffusion through laminar sub-layer

  • 3. Uptake at

t th the su surface Rsoi

  • il

Rc Canopy resistance

Stomatal resistance

Rstom Rcut Rmes

Mesophyll resistance Cuticular resistance

slide-19
SLIDE 19

Modelling Dry Deposition A Resistance Approach

Aerodynamic resistance Quasi-laminar resistance Ra Rb Soil resistance

Cuticular resistance Stomatal resistance

Rsoi

  • il

Rstom Rcut Rc Rmes

Mesophyll resistance

soil cut mes stom c

R R R R R 1 1 1 + + + =

c b a d

R R R V + + = 1

SWC VPD Temp light phen stom

f f f f f R R

min

=

Jarvis multiplicative approach:

slide-20
SLIDE 20

Modelling Dry Deposition Jarvis Approach

Solar Radiation flight fVPD VPD ftemp Temperature SWC VPD Temp light phen stom

f f f f f R R

min

=

1 1 1

slide-21
SLIDE 21

Modelling Dry Deposition Jarvis Approach

Solar Radiation flight fVPD VPD ftemp Temperature SWC VPD Temp light phen stom

f f f f f R R

min

=

VPD fVPD  − = 02 . 1

CASTNET

1 1 1

slide-22
SLIDE 22

Sweetguma Norway Spruceb Beechc CASTNET Grapec Sunflowerd Potatoe VPD fVPD  − = 02 . 1 (a) Gunderson, 2002. (b) Karlsson , 2000. (c) Buker,2007. (d) Emberson, 2000. (e) Pleijel, 2002.

Modelling Dry Deposition fvpd

slide-23
SLIDE 23

Does our model work?

Vd (Our Model) Vd (CASNET)

slide-24
SLIDE 24

Does our model work?

R2 = 0.68 1:1 Vd (Our Model) Vd (CASNET)

slide-25
SLIDE 25

H(t) Δx Deposition

   

( )

t H O V dt O d

d dep 3 3

− =

Affect of Dry Dep on Summer Ozone

slide-26
SLIDE 26

H(t) Δx Deposition

   

( )

t H O V dt O d

d dep 3 3

− =

Affect of Dry Dep on Summer Ozone

slide-27
SLIDE 27

Affect of Dry Dep on Summer Ozone

75th precentile humidity for site vs 25th precentile humidity for site

Our Model’s Deposition Values CASTNET’s Deposition Values

(humid) (dry)

slide-28
SLIDE 28

Our Model’s Deposition Values CASTNET’s Deposition Values

Affect of Dry Dep on Summer Ozone

slide-29
SLIDE 29

Our Model’s Deposition Values CASTNET’s Deposition Values

Affect of Dry Dep on Summer Ozone

  • Vieno et al. (2010) associated a heat

wave with an extra 20 to 35 ppb of

  • zone due to the loss of the dry dep.

sink

  • Royal Society (2008) found ‘turning
  • ff’ deposition lead to a 19% increase

in daily mean ozone concentrations

  • Emberson et al. (2013) found

European exceedance days tripled under drought stressed conditions

slide-30
SLIDE 30

Conclusions

  • Midday Ozone and VPD are well correlated at

most CASTNET sites during the summer

  • Deposition of ozone to vegetation is sensitive to

VPD for many species of plants

  • The VPD-sensitive ozone sink can result in 5-

12ppb differences in day-to-day ozone concentrations

  • But: Ozone fluxes are highly sensitive to stomatal

resistance parameterization choices

slide-31
SLIDE 31

Limitations/Future Work

  • We don’t have a good characterization of what the

boundary height is doing which is really essential for deposition modelling.

  • Ongoing: We’re currently implementing our species-

dependent Ja Jarvis sch cheme in into GEOS-Chem

  • Currently: We’re mapping the entire canopy onto a

single representative leaf. Comparisons to real measurements are tricky.

  • Future: We’re planning more detailed canopy modelling

in in conjunction wit ith fie field campaign flu flux measurements.

  • Currently: We’re assuming well watered vegetation (no

drought stress) and no surface wetness effects as well as static species composition and LAI throughout time (not great assumptions)