Numerical simulation of space-time Numerical simulation of space-time distribution of gases and particles emitted distribution of gases and particles emitted

during 1998 Roraima fires during 1998 Roraima fires

Rodrigo GevaerdRodrigo Gevaerd11 Saulo FreitasSaulo Freitas22 – Karla Longo – Karla Longo22

11 Universidade de São PauloUniversidade de São Paulo22 Centro de Previsão de Tempo e Estudos Climáticos/INPECentro de Previsão de Tempo e Estudos Climáticos/INPE

Vegetation fires in Roraima during 1998

Burned areas by biome (km2) Barbosa e Fearnside

(2000)

ATSR and TRMM fire countsMarch -1998

smoke

El Niño• SST anomaly at Equatorial Pacific: + 3oC.• Weak convective activity on north of Brazil.• Positive Anomalies of temperature and negative of relative humidity on Roraima.

SST anomaly of 1997/98 years ~ + 4 K

Vegetation fires in Roraima during 1998Meteorological condition - (El Niño)

• Boa Vista: 30,6 mm set/97-mar/98

• 8.7% of climatological mean 352mm

• Relative humidity < 60 %

•Agricultural burning was the cause of initial ignition

Rainfall – March, 1998 (mm)

< 50 mm

LBA-CLAIRE 1998 Experiment: Flight-8, March, 26 LBA-CLAIRE 1998 Experiment: Flight-8, March, 26 19981998

Measurement of CO and COMeasurement of CO and CO2 2 at high troposphereat high tropospherewith biomass burning “signature”with biomass burning “signature”

10 km

CO CO – – COCO22

300 ppbv CO

Flight-8Flight-8 Guiana and SurinameGuiana and Suriname

Origin of air mass on Suriname on March, 26 1998

• Planetary Boundary Layer (1km)

East cost of EUA

(Krejci et al., 2004)

• Low troposphere ~ 3 km:

Africa

(Formenti et al., 2001)

• High Troposphere ~ 10 - 12 km:

Roraima

(Freitas et al., 2000)

Some sub-grid process involved at Some sub-grid process involved at gases/aerosols transportationgases/aerosols transportation

3D Eulerian model grid 3D Eulerian model grid boxbox

massinflo

w

massoutflo

w

x ~ 10-100 kmx ~ 10-100 km y ~

10-

100

km

y ~ 1

0-10

0 km

z ~

100 –

1000

m

z ~

100 –

1000

m

source

diffusion diffusion in the in the PBLPBL

convective transport convective transport byby

shallow cumulusshallow cumulus

convective transport convective transport byby

deep cumulus deep cumulus

dry dry depositiodepositio

nn

wet wet depositiodepositio

nn

Model top Model top 20-30 km20-30 km

Eulerian Transport Model :Eulerian Transport Model :CATT-BRAMS Atmospheric ModelCATT-BRAMS Atmospheric Model

• in-line Eulerian transport model fully coupled to the atmospheric dynamics• suitable for feedbacks studies• tracer mixing ratio tendency equation

where:• adv adv grid-scale advectiongrid-scale advection• PBL turb PBL turb sub-grid transport in the PBLsub-grid transport in the PBL• deep conv deep conv sub-grid transport associated to the deep convection including sub-grid transport associated to the deep convection including

downdraft at cloud scaledowndraft at cloud scale• shallow conv shallow conv sub-grid transport associated to the shallow convectionsub-grid transport associated to the shallow convection• W W convective wet removal convective wet removal • RR sink term associated with dry deposition or chemical transformation sink term associated with dry deposition or chemical transformation• Q Q source emission.source emission.

2.5 ,PMPBL deep shallowadv turb conv conv

s s s s sW R Q

t t t t t

Biomass burning emissions Biomass burning emissions estimateestimate

[ ][ ] . . . ,veg

f fireveg vegM a

carbon density data

land use data

fire product

emission & combustion factors

mass estimation

CO source emission (kg m-

2 day-1 )

Emission parameterization for this case study

total burned area given by Barbosa e Fearnside (2000)

diurnal variability of burned area given by the same time variation of TOMS Aerosol Index on Roraima.

combine fire counts from ATSR and TRMM to provide fire location

firefires count

( ) ( )

( )( )

q total

q

A t F t A

A tA t

N

Mean burned area estimate per fire count

Emission parameterization for this case study

[ ]

_0_

( )( , , ) ,fire

firesGrid Box

r tQ i j t B E A

V

Model grid-box daily

emissions estimate:

Total daily emissions of CO and PM2.5 (Tg)

40CO

4,5PM2.5

Regional daily emission of CO (Tg) 23-mar-1998

Parameterized Deep Convective TransportParameterized Deep Convective Transport

updraft detrainmentupdraft detrainment downdraft detrainmentdowndraft detrainmentenvironment environment

subsidencesubsidence

0)

)

(

(1 or

deep

b

bconv

du u udeepcon

u u d d

v

ud d

u z

zs mw ss s s

m z

s

zs

tss s

z

Vertical section at lat 10SVertical section at lat 10S

vertical vertical levellevel 11.5 km 11.5 km

CO (ppbv)CO (ppbv)

Deep Convective Deep Convective Transport of COTransport of CO

Heig

ht

(m)

Heig

ht

(m)

1 km1 km

10 km10 km

Model simulation of the main convective systemson 18-march-1998

Numerical simulation of transport of CO with Numerical simulation of transport of CO with CATT-BRAMS during CLAIRE flight 8CATT-BRAMS during CLAIRE flight 8

height 1km height 11,7 kmheight 1km height 11,7 km

Fire emissions Fire emissions andand trade wind trade wind

deep deep convectionconvectionsmoke in PBLsmoke in PBL

anticyclonic anticyclonic flowflow smoke at high smoke at high

tropospheretroposphere

CLAIRE CLAIRE flight 8flight 8

trade trade windwind

Numerical simulation of transport of CO with Numerical simulation of transport of CO with CATT-BRAMS during CLAIRE flight 8CATT-BRAMS during CLAIRE flight 8

Model validation using flight-8 CO profile

Model validation using flight-8 CO

profile

----- CO observation----- CO modeled

Thanks for your attention

Validação do modelo CATT-BRAMS (with Plume Rise)

Conclusões

• Validação do modelo de transporte CATT-BRAMS 3.3 e da parametrização de fontes emissoras.

• Utilização de avanços em elementos importantes do modelo atmosférico: parametrização de cúmulos (cúmulos rasos e técnica training) e representação de superfície (umidade do solo).

• Importância da queimada de grandes proporções no dia 17/março (pois não houve necessidade de um ambiente previamente poluído) e o sistema convectivo formado na tarde do dia 18/março.

• Testes iniciais da técnica plume rise.

Transporte de emissões de queimadas

• advecção em níveis baixos,

• cúmulos rasos,

• cúmulos profundos,

• advecção em níveis altos (jatos),

• formação de novas partículas,

• processos químicos,

• deposição seca,

• remoção úmida, etc.

Transporte em escala continental

Fonte: Krejci et al. (2003)

Plume caracteristics Características da pluma de fumaça em altitude

(“assinatura”)

Fonte: Andreae et al. (2001)

• elevados valores de ΔCO/ΔCO2 acima de 9 km, típicos de eventos

extremos de queimadas em savanas,

• ausência de halocarbonetos de origem antropogênica,

• elevada concentração de metilcloratos.

CATT-BRAMS model simulation setup

• instante inicial: 12UTC 17 março 1998• instante final: 12UTC 28 março de 1998• número de pontos: (290,242,43)• níveis de solo: 7• resolução horizontal: 30 km• resolução vertical no primeiro nível: 120 m• razão de incremento: 1.1• valor máximo de Δz: 750 m• passo de tempo: 30 s• campos atmosféricos: ECMWF• umidade do solo: Gevaerd e Freitas (2005)• radiação de onda curta: Harrington (1997)• radiação de onda longa: Chen e Cotton (1983)• parametrização de cúmulos: Grell (1993) – técnica training

Emissão de poluentes

Destino do carbono queimado em Roraima entre dez/97 e mar/98 (106 Ton):

Emissão bruta de gases do efeito estufa em Roraima entre dez/97 e mar/98 (106 Ton):

Modelo de transporte CATT-BRAMS 3.3

Advecção na escala da grade

Convecção rasa

Turbulência na escala sub-grade

Convecção profunda

Remoção úmida do PM25

Deposição seca e processos químicos

Emissão associada com a queima de biomassa

Validação do modelo CATT-BRAMS

Fatores envolvidos no transporte (STILT)

(a) Transporte advectivo pelos alísios (baixos níveis),(b) Transporte vertical promovido por sistemas convectivos,(c) Transporte advectivo seguindo um anticiclone em altos níveis.

COM CONVECÇÃO SEM CONVECÇÃO

Fatores envolvidos no transporte