Report on the Berlin 2 Open Access Conference - European ...

More documents

Recommendations

Info

ing different aspects of this topic on a range of spatial and temporalscales. Combined with the input data from the laboratoryexperiments this integrated approach allows a more completeview on physical, chemical and dynamical aspects of biomassburning plumes. The model hierarchy facilitates seamlesstransition from microscale to regional scale models and spansthe temporal scale from a few minutes to months and years.Among other aspects the modelling efforts address:• impact of biomass burning aerosol on cloud microphysicsand precipitation using detailed and parameterized model approachesincluding the ice phase• evolution of individual biomass burning plumes, consideringdynamical evolution and troposphere-stratosphere exchangeof emissions as well as chemical processes in youngplumes• effect of radiative transport on chemistry and dynamics ofthe atmosphere and determination of the role of biomass burningon the regional climate• effect of biomass burning on the atmospheric budgets oftrace constituents, water, and energy on the regional scaleSelected resultsOptical properties of the biomass burning aerosol: Modellingthe impact of vegetation fires on radiative transport requiresaccurate knowledge of the optical properties of biomass burningaerosol. Optical properties of particles are well describedby Mie theory and mainly depend on particle size, shape, andindex of refraction m. However, for biomass burning aerosolespecially m is not well known. The EFEU data set providesthe unique opportunity to investigate m for various fuel typesand burning conditions. As an example Fig. 4 shows 2-minuteaverages of the measured SSA (single scattering albedo = ratioof scattering and extinction coefficient) at a wavelength of550 nm of about 9-minutes old biomass burning aerosol froman oak (dry) fire; the corresponding burning conditions are alsoshown as dCO/dCO2 trace (crosses), where dCO and dCO2represent CO and CO2 concentrations above background levels.There was an obvious positive correlation between SSAand dCO/dCO2. In the smouldering phase at the beginningof the experiment (dCO/dCO2 > 8%) SSA was very high withvalues around 0.95. With the start of the flaming combustion(dCO/dCO2 < 8%) at 11:33 SSA showed a sharp decreaseFigure 4. Comparison between measured and modelled singlescattering albedo (SSA) for a 1-hour fire with dry oak as fuel.down to approximately 0.7 followed by a relatively steady increaseto just above 0.9 accompanied by an increase in dCO/dCO2.For comparison, Fig. 4 also shows Mie calculations of theSSA for three different black carbon (BC) mass fractions basedon the measured number size distributions under the assumptionof spherical particle shape which is justified by SEM pictures.Due to the significantly lower volatility of black carbon(BC) compared to organic carbon (OC) we performed the Miecalculations for coated spheres with a BC core (mBC = 1.75-0.44i) and an OC shell (mOC = 1.53-0.0i). Fig. 4 shows thatthe modelled SSA is fairly constant for each BC mass fraction.Hence, changes in size distribution alone cannot explain theobserved variations in SSA which indicates a significant role ofparticle chemistry and hence m. We found that BC mass fractionsof 2 and 20% represent the observed extreme SSA valuesof about 0.95 and 0.7, respectively, reasonably well. Thisis consistent with the expected higher BC content for moreflaming fires (lower dCO/dCO2 values) which produce highercombustion temperatures and consequently more soot. Hence,our results suggest that young biomass burning aerosol maybe modelled as BC spheres coated with a variable amount ofOC depending on the burning conditions. Similar Mie calculationswill be performed for all of the wood types investigated inthe EFEU laboratory campaign and then hopefully condensedinto a single parameterization at least for certain categories ofwood (e.g. northern boreal, tropical, peat, etc.).Drop nucleation ability of the aerosol particles: As a step forthe understanding of the influence of biomass burning aerosolon cloud formation, potential cloud condensation nuclei (CCN)were identified by determining the CCN/CN ratio in the laboratoryexperiments. This was done for different dry particle sizes(50, 100, 150, 250, and 325 nm) at different supersaturationsfrom 0.24 to 1.64 %. The experimental results are compared tocritical supersaturations calculated from extended Köhler theorywhich is also used in the detailed microphysical models appliedin the project. Fig. 5 shows the critical supersaturation forvarying e (soluble fraction, here ammonium sulphate) and APsizes (lines with different colors). CCN/CN ratios of 25 (circles),50 (squares), 75 (diamonds), and 100 % (triangles) are deducedfrom the measurements and compared to the theoreticalvalues. The black square on the red line means, e.g., that50 % of the 100 nm AP are activated at 0.46 % supersaturationwhich corresponds to a theoretical value of e = 0.08. One cansee that the soluble fraction generally is small (e < 10 % forabout 50 % of the particles) as it is expected for biomass burningAP and that e is slightly higher for smaller AP sizes (50 nmvs. 150 nm). Despite their small e, larger particles (150, 200nm) are activated at relatively low supersaturations that easilycould be reached in clouds. Obviously, the critical supersaturationsare very sensitive to e < 20 % especially for smaller sizes(50, 100 nm). Therefore, small changes in epsilon could highlyinfluence the number of drops in clouds affected by biomassburning. The CCN/CN ratios are quite similar for the differentwood types (spruce, pine, oak, and musasa) considered in theEFEU experiments.Modelling results from the regional model REMO: Thespatial distribution of smoke haze in Indonesia in 1997 canbe derived from the TOMS Aerosol Index (AI) data, shown asmonthly mean in the upper part of Fig. 6. During the smokehaze period maximum intensity in TOMS AI data is visible inthe provinces of South Sumatra, Jambi, Central and West Kalimantan,and on Irian Jaya, where the main peat fire emissionTHE EGGS 16
areas are located. Large scale transport from these source regionsto northern and western directions reaching into the IndianOcean occurred especially during September 1997, whenthe fire emissions were highest and rainfall and associated wetdeposition was lowest.The lower part of Fig. 6 presents the monthly mean atmosphericcolumn burden in mg(C)/m2 of total particulate matter(TPM) as determined by the regional model REMO. It shouldbe emphasised at this point that Fig. 6 displays only a qualitativecomparison of TOMS AI with REMO calculated TPM inorder to evaluate the simulated temporal and spatial distributionof smoke haze. The principle ability of the REMO modelto reproduce the spatial and temporal distribution of TPM isclearly visible from Fig. 6. A north-west transport of smokehaze dominated in September 1997. Compared to TOMS AIthe intensity of the smoke haze originating from Irian Jaya isrelatively low. These differences in modelled and observedsmoke haze distribution can result from the fact that TOMSFigure 5. Critical supersaturations (lines) and CCN/CN ratios(symbols) for different AP sizes and solubilities.AI cannot detect absorbing aerosols at altitudesbelow about 1 km because of theunderlying Rayleigh scattering.More info can be found at the project’swebsite:h t t p : / / p r o j e c t s . t r o p o s . d e : 8 0 8 8 /afo2000g3/S. Wurzler (1,5), M. Simmel, K. Diehl,T. Hennig, H. Herrmann, Y. Iinuma,K. Lehmann, A. Massling, F. Stratmann,A. Wiedensohler, G. Zech, K. Zeromskiene(1), R. Posselt (1,6), K. Hungershöfer(2), T. Trautmann (2,7), M. O. Andreae,D. Chand, U. Dusek, G. P. Frank,G. Helas, R. S. Parmar, O. Schmid, T.Winterrath, M. Welling (3), J. Trentmann(3,8), H. F. Graf (4,9), B. Langmann (4),F. Nober (9), C. Textor (4,10)Corresponding author: M. Simmel(simmel@tropos.de)(1) Leibniz Institute for TroposphericResearch, Leipzig(2) University of Leipzig(3) Max Planck Institute for Chemistry,Biogeochemistry, Mainz(4) Max Planck Institute for Meteorology,Hamburg(5) North Rhine Westphalia StateOffice for Environment(6) Dalhousie University, Halifax(7) German Aerospace Center,Oberpfaffenhofen(8) University of Washington, Seattle(9) University of Cambridge(10) LSCE, CEA/CNRS SaclayFigure 6. Comparison of TOMS AI data over Indonesia with REMO TPM results forSeptember 1997.This article has appeared in a similarform in the AFO2000 Newsletter.THE EGGS 17
Page 1: Impact of vegetation fires on compo
Page 4 and 5: appraisal forms. Each judge is expe
Page 6 and 7: European Mineral Sciences Initiativ
Page 8 and 9: image of the Earth-Moon system, tak
Page 10 and 11: Obituary:James Reed Holton (1938-20
Page 12 and 13: were drilled without measurement.Th
Page 14 and 15: Impact of vegetation fires oncompos
Page 18 and 19: Report on the Berl
Page 20 and 21: cloudiness. The complexity of inter
Page 22 and 23: A very useful reference book, espec
Page 24 and 25: according to Barbara S. Lollar who
Page 26 and 27: a list of learning objectives and e
Page 28 and 29: their institutions for access. We t
Page 30 and 31: 1st Alexander von HumboldtInternati

Report on the Berlin 2 Open Access Conference - European ...

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?