atmosphere modelling group atmosphere modelling group
play

Atmosphere Modelling Group Atmosphere Modelling Group (with a - PDF document

Atmosphere Modelling Group Atmosphere Modelling Group (with a strong focus on new particle formation) (with a strong focus on new particle formation) University of Helsinki University of Helsinki Department of Physics Department of Physics


  1. Atmosphere Modelling Group Atmosphere Modelling Group (with a strong focus on new particle formation) (with a strong focus on new particle formation) University of Helsinki University of Helsinki Department of Physics Department of Physics Division of Atmospheric Sciences Division of Atmospheric Sciences kilometers ECHAM5- HAM S CADIS PENCIL- COUD cm meters MALTE S OS A UHMA seconds days months years

  2. kilometers M ichael Boy Sampo Smolander (group leader) Risto M akkonen Rosa Gierens ECHAM5- HAM K.V. Gopalkrishnan Natalia Babkovskaia S CADIS PENCIL cm meters CLOUD J ohanna Lauros Sanna-Liisa Luxi Zhou Sihto MALTE S OS A Anton Rusanen Henri Vuollekoski UHMA He Qingyang Chatriya Watcharapaskorn Ditte M ogensen seconds days months years The UHMA box model coagulation nucleation cloud droplet activation condensation Figure: Miikka Dal Maso

  3. UHMA – – University of Helsinki Multi University of Helsinki Multi- - UHMA component Aerosol model component Aerosol model MECCO – a method to estimate concentrations of condensing organics Vuollekoski, H., et al. Journal of Aerosol Science,41, 1080-1089, 2010. What is growing the particles?

  4. Basic idea of Markov chain Monte Carlo methods 0 � probability � 1 MECCO – a method to estimate concentrations of condensing organics UHMA

  5. Testing MECCO--UHMA: create perfect data UHMA MECCO—UHMA is working

  6. Field data needs to be smoothened Preliminary testing of with field data

  7. MALTE / SOSA / SCADIS MALTE / SOSA / SCADIS 3000 m Aerosol METEOROLOGY UHMA SCADIS CHEMISTRY EMISSIONS KPP MEGAN 15 m MCM 0 m MALTE – – Model to predict new Aerosol Model to predict new Aerosol MALTE formation in the Lower Tropospher Tropospher formation in the Lower Particle concentration and flux dynamics in the atmospheric boundary layer as the indicator of formation mechanism Lauros et al., Atmos. Chem. Phys. Discuss. 10, 20005-20033, 2010

  8. M ALTE (Hyytiä äl lä ä, M arch 2006) , M arch 2006) M ALTE (Hyyti Measurements: DMPS O 4 ] 2 (K = 5* 10 -13 cm 3 s -1 ) J = K * [H 2 S J = K org1 * � * [H 2 S O 4 ] * {[Mont.][O 3 ]} J = K org2 * � * [H 2 S O 4 ] * {[Mont.] [OH]} Vertical profile of particles with Vertical profile of particles with D p > 10 nm D p > 10 nm Kinetic nucleation (H 2 S O 4 ) Observation Organic induced nucleation

  9. SOSA – – Model to calculate the Model to calculate the SOSA concentrations of Organic vapours vapours and and concentrations of Organic Sulphuric Acid Acid Sulphuric Long term statistical comparison of different compounds

  10. Long-term data analyses: 2003-2008 dark green: days > 75 % of the yearly mean value between 9 am and 3 pm green: days > 60 & < 75 % of the yearly mean value between 9 am and 3 pm light green: days > 50 & < 60 % of the yearly mean value between 9 am and 3 pm � � H SO 1 � 2 4 Rp CS

  11. Modelling Atmospheric OH- - Modelling Atmospheric OH Reactivity in a Boreal Forest Reactivity in a Boreal Forest � OH-reactivity = loss rate of OH R OH = � k OH+X [X] Unit of OH-reactivity, R OH is [s -1 ] k OH+X is the rate coefficient [cm 3 mol -1 s -1 ] [X] is the concentration of chemical compound X Measured and Modeled OH- reactivity for August 2008 Modeled (blue) and 30 minute resolution, measured OH-reactivity (black) from the 13 th to the 27 th of August, 2008 at 14 meters.

  12. Modeled, Measured, and Missing OH- Reactivity 13-27Aug 13-18Aug 19-27Aug Modeled 2.5 s -1 2.3 s -1 2.6 s -1 Measured 6.5 s -1 8.6 s -1 5.1 s -1 4s -1 / 61% 6.2 s -1 / 73% 2.5 s -1 / 49% Missing We seem to be able to predict 30-50% of the OH-reactivity! Vertical profile of OH-reactivity [s -1 ] peak near ground during night night deposition and suppression of boundary layer

  13. Seasonally variation for 2008 Contributions from inorganic compounds, isoprene, methane, monoterpenes, and other VOCs. Fighting fire with fire: Could NO x emission be used to remove methane in a catastrophic clathrate release scenario?

  14. Clathrate gun hypothesis Temperature rise � Methane clathrate is crystalline solid which looks like ice, and in which a large amount of methane is trapped within a crystal structure of water � Clathrates naturally occur in permafrost and seabed. Clathrate Methane destablize release � Total reservoirs on earth ranges from 10 3 to 10 4 GtC. [Kennet et al. 2000] Large methane emission at ESAS • Greater than 80% of East Siberian Arctic Shelf (ESAS) bottom waters and greater than 50% of surface waters are supersaturated with methane regarding to the atmosphere. • The amount of methane currently coming out of ESAS is comparable to the amount coming out of the entire world's oceans. • [Shakhova et al., Science 5 March 2010 ]

  15. RCP database: By 2100, total methane emission may increase 300%. Osasto / Henkilön nimi / Esityksen nimi 7.12.2010 29 How is methane oxidized in atmosphere? CH 3 O NO O 2 OH OH OH CH 4 CH 3 O 2 HCHO CO CO 2 hv O 2 hv HO 2 OH OH CH 3 OOH Deposition 15.09.2010 30

  16. Adding NO x decrease methane..... CH 4 + OH + O 2 � CH 3 O 2 + H 2 O CH 3 O 2 + NO � NO 2 + CH 3 O CH 3 O + O 2 � HCHO + HO 2 HO 2 + NO � NO 2 + OH CH 4 +2O 2 +2NO � HCHO+H 2 O+2NO 2 2(NO 2 + hv � NO + O) 2(O + O 2 +M � O 3 +M) Net: CH 4 +4O 2 +2hv � HCHO+2O 3 +H 2 O R. P. Wayne, Chemistry of Atmosphere, 3rd edition, Oxford University Press, New York, 2000 Adding NO x has both cooling and warming effects

  17. A set of models were used to assess the effects of adding NO x � RF in CH 4 Radiative forcing (RF) is the change in net irradiance at the tropopause/top of atmosphere. CH 4 and NO x concentration were fixed at different values � Baseline scenario: • Unperturbed methane and NO x concentration level. � 10M/100M, 1N scenario: • 10/100 times present day methane concentration level ; • Unperturbed NO x concentration level. � 10M, 2N scenario: • 10 times present day methane concentration level; • 2 times present day NO x concentration level.

  18. After double NO x emission … � methane life time change is small � RF change due to ozone and aerosol indirect effect are comparable O 3 radiat radiative ive CD CDNC-alb NC-albedo relat edo related ed Scen Scenario ario CH 4 li CH lifetime (ye fetime (years) ars) forcing (W/ forcin g (W/m 2 ) radiat rad iative f ive forc orcing (W/ ing (W/m 2 ) 1M,1N 0.0 0.0 12 10M, 1N 0.76 2.06 22.2 -0.36 +0.34 10M, 2N 1.10 1.70 19.8 � Increase the methane by 10 in the atmosphere would result in a radiative forcing change of 2.514 W/m 2 . How big are the heating and cooling effects? � Scenario with CH 4 concentration EFFECT Magnitude (J/m 2 ) increased 10-fold CH 4 removal - 5.95×10 6 � Doubling of the NO X concentration for a year + 10.72×10 6 O 3 increase CDNC - 11.35×10 6 increase - 6.58 × 10 6 Net We do get a net cooling effect!!! But....

  19. Cooling effect insignificant Net cooling effect: -6.58 J/ m 2 � -0.21 W/ m 2 * year It is too small compared to the initial warming due to methane increase ( 2.51 W/m 2 ) as well as associated ozone warming (0.76 W/m 2 ) and aerosol indirect effects (2.06W/m 2 ). 0.21 W/m 2 << (2.51 + 0.76 + 2.06) W/m 2 Not an effctive way to save us! Elevated methane level leads to strong CDNC reduction

  20. CDNC reduction leads to positive aerosol indirect effects CDNC reduction leads to positive aerosol indirect effects

  21. PENCIL- -Cloud Cloud PENCIL Pencil code as a powerful tool for calculation of the turbulence coupled with an aerosol dynamic module to study cloud processes Scientific objectives Scientific objectives • studying the influence of turbulence on the aerosol dynamics and vice versa inside a cloud • investigating the activation of particles at the cloud boundary • quantifying the effect of particle production at the outflow of a cloud

  22. 2D aerosol + fluid dynamics model 2D aerosol + fluid dynamics model SCADIS SCADIS (SCA SCALAR LAR DIS DISTRIBUTION) TRIBUTION) ( SCADISis a high-resolution 3D model capable of computing the physical processes with both plant canopy and atmospheric boundary layer simultaneously Horizontal and Vertical Resolution – As per specific requirement

  23. TKE OVER HYYTIÄLÄ FOR ONE DAY ECHAM5- -Ham Ham ECHAM5 an aerosol- -climate modelling system climate modelling system an aerosol Past, present and future new particle formation

  24. #/cm 3 Future � Present � Past Condensation nuclei Cloud condensation nuclei (diameter > 3 nm) � (diameter > 70 nm) �

  25. Aerosol indirect effect (W/m 2 �� (anthropogenic effect: present-day/future compared to pre-industrial) �

  26. Aerosol indirect effect (W/m 2 ) (anthropogenic effect: present-day/future compared to pre-industrial)

Recommend


More recommend