“The world according to….” The world according to….” “
Modeling surface trace gas exchanges: Modeling surface trace gas exchanges: What are the main limitations? What are the main limitations? Laurens Ganzeveld Max-Planck Institute for Chemistry, Mainz, Germany E-mail: ganzevl@mpch-mainz.mpg.de
Outline Outline • Results of modelling of surface trace gas exchanges at site- and global scale • What are the main limitations and how can we deal with/get around these? • Future: scenario analyses of impact of land cover and land use changes on atmospheric chemistry • Conclusion and outlook
Modeling surface trace gas exchanges Modeling surface trace gas exchanges SO 2 dry deposition velocity [cm s - -1 1 ] ] SO 2 dry deposition velocity [cm s 0.02 0.35 0.75 1.15
dt = ∂ c + ∂ c + ∂ c + ∂ c dc ∂ t ∂ t ∂ t ∂ t turb chem emiss dep ECHAM/SCM surface layer ~ 68 m chemistry turbulence Vegetation model crown-layer dry deposition ~ 0.5-15 m canopy-soil emissions layer Vegetation and wet skin fraction
Evaluation of multi- -layer model in SCM layer model in SCM Evaluation of multi Tropical rainforest, Manaus Manaus (Brazil), LAI=7, canopy height = 30 m (Brazil), LAI=7, canopy height = 30 m Tropical rainforest, Model Model Measurements Measurements R n [W m -2 ] 450 ~ 600 CC [0-1] 0.9 ~0.7 H [W m -2 ] 20 ~200 T air [K] 297 – 300 (32 m) ~296-301 (~39 m) u [m s -1 ] 0.5 – 2 (~39 m) ~1.5 – 2.5 (~39 m) q [g kg -1 ] 19 (~53 m) 15 (~39 m) 0 4 8 12 16 20 24 0 O3 flux [1e11 molec. cm-2 s-1] -1 -2 model -3 observ. model, 'dry day' -4 Time [hr]
Tropospheric NO NO x budgets Tropospheric x budgets Annual Annual
What are the limitations? What are the limitations? • Available measurements • Lack of understanding of controlling mechanisms • Characterization of surface properties and land use • Representation of micro- and PBL meteorology in models • Scaling issue: upscaling of measurements to the model resolution
• Lack of measurements: biased towards vegetated surface and only a small selection of gases (O 3 , SO 2 , NO x/y ) have been measured on a ”routinely” bases How to deal with /get around these limitations: How to deal with /get around these limitations: - wait patiently and hopefully the scientific career will last long enough - offer help: join field campaigns! - use model results for feedback to experimentalists O 3 soil resistance as a function of soil organic matter and soil moisture, July.
• Lack of understanding of controlling mechanisms � Vegetation: -Stomatal vs cuticular uptake, what is the fate within the plant tissue? Effects? -What controls the internal N concentrations? -How are the VOC emissions related to the nitrogen and carbon cycle? -Surface wetness? � Soils: -What controls the uptake by soils? organic material, soil moisture, soil pH, etc. � Water: -Role of plankton in controlling uptake/release of trace gases like O 3 /DMS NO 2 dry deposition velocity versus the stomatal resistance � Parameterisations become (too) complex and specific Parameterisations become (too) complex and specific [C NO2 = 41 ppbv], data by Johansson, 1987 � 5 Lehning at al., 2001: seasonal cycle in C 5 H 8 emission factors based on a parameterization of isoprene 4 synthase activity by fitting to observations. � Inconsistencies between parameterisations Inconsistencies between parameterisations � -1 ] 3 d [mm s VOC emissions versus dry deposition V 2 � Parameterisations are often scale dependent Parameterisations are often scale dependent � � Parameterisations are representative for present Parameterisations are representative for present- -day day � 1 conditions and might not be applicable to predict surface conditions and might not be applicable to predict surface 0 exchanges for future environmental conditions exchanges for future environmental conditions 0 1 2 3 4 5 s [mm s -1 ] Stomatal conductance K
We should get away from including too many different parameterisations to ations to We should get away from including too many different parameteris represent surface exchange processes of the different species in models. Instead, models. Instead, represent surface exchange processes of the different species in the development of more process- -based models that link the C and N cycle to the based models that link the C and N cycle to the the development of more process uptake (and effects of the uptake) of gases such as O 3 , SO SO x , etc., and VOC emissions uptake (and effects of the uptake) of gases such as O 3 , x , etc., and VOC emissions should be pursued. should be pursued. What is needed to develop process- -based models? based models? What is needed to develop process � More measurements in well constrained conditions (laboratory) to determine the role of the different substrates involved, e.g, soils versus vegetation . � Include parameters in field campaigns that are complementary in identifying the mechanisms: -energy balance and turbulence -photosynthesis -O 3 uptake -N uptake/release -VOC emissions
• Characterization of surface properties and land use January July 0 10
4 LAI=7 LAI=7 NO x flux [10 9 molec. cm -2 s -1 ] LAI=11 3 LAI=11 2 1 0 -1 -2 -3 00:00 06:00 12:00 18:00 24:00 Time [hr:min] What is needed? What is needed? More collaborations with GIS and remote sensing specialists: • We: What data are required for modeling? Soil properties, vegetation data, etc. • They: Data availability, domain (regional to global scale), resolution, and UNCERTAINTIES! ! UNCERTAINTIES
• Representation of micro- and PBL meteorology in models Sensible and latent heatfluxes, Harvard forest site. 100 0 H F /LF [W m -2] -100 -200 -300 -400 Max. observed. LE -500 -600 1 10 19 28 37 46 55 64 73 82 91 100 109 118 Time [Hr] H, original soil moisture LE, original soil moisture H LE
Model resolved PBL height for a soil moisture of 0.4 and 0.2 m, tropical rainforest, Manaus, April 1600 ws=0.4 Ws=0.4 1400 ws=0.2 Ws=0.2 1200 1000 PBL height [m] 800 600 400 200 0 0:00:00:00 0:06:00:00 0:12:00:00 0:18:00:00 1:00:00:00 Time [day:hr:min:sec]
•Turbulence exchange: - counter-gradient transport - intermittency - stability effects in canopy versus surface layer, nocturnal free-convection Tropical rainforest NOx flux [1e9 molec. cm-2 s-1] 16 11 6 1 1:03 1:06 1:10 1:14 1:17 1:21 -4 Time [day:hr] Leaf Area Index=8 LAI=4 comp. CNO2=0.5 ppbv nocturnal mixing each 2 hr
Nudging of SCM for model evaluation Nudging of SCM for model evaluation Friction velocity, BEWA site, July-august 2001 1 measured model, Trelax=2hr 0.8 model, Trelax=3hr u * [m s -1 ] 0.6 0.4 0.2 0 20.07.01 22.07.01 24.07.01 26.07.01 28.07.01 30.07.01 Time [dd.mm.yr] Air temperature, BEWA site, July-august 2001 30 25 T air [°C] 20 15 Tair Ts, Trelax=2hr Ts, Trelax=3hr 10 27.07.01 27.07.01 28.07.01 28.07.01 29.07.01 29.07.01 30.07.01 Time [dd.mm.yr]
Air temperature, BEWA site, July-august 2001 30 25 20 T air [°C] 15 10 5 10.07.01 15.07.01 20.07.01 25.07.01 30.07.01 Time [dd.mm.yr] observed model model, ws=wsmax
Short-wave radiation, BEWA site, July-august 2001 1000 800 Rg [W m -2 ] 600 400 200 0 10.07.01 15.07.01 20.07.01 25.07.01 30.07.01 Time [dd.mm.yr] observed model model, albedo=0.12
Simulated versus observed net radiation, BEWA site, July-August 2001. 1000 y = 1.0565x + 11.733 Observed Rn [W m-2] R 2 = 0.8332 750 y = 0.9394x + 5.2445 R 2 = 0.8289 500 albedo=0.17 albedo=0.12 250 albedo=0.12 albedo=0.17 0 0 250 500 750 1000 Simulated Rn [W m-2]
• Scaling issue: upscaling of measurements to the model resolution -Workshop with specialists from all kind of disciplines in which upscaling is involved didn’t result in a major breakthrough. Solutions? � Nested models/model hierarchy, site-scale – regional – global scale applications Example: SCM constrained with ECMWF analysed meteorology for interpretation of site observations using the process representation consistent with the global model ECHAM. � Using process-based models to avoid using scale-dependent parameterizations Bottleneck: are the required input data available at the site- as well as larger scales? Accuracy? Example: DNDC model for N soil emissions based on soil-redox potential
Land cover and land use changes Land cover and land use changes L arge Scale B iosphere A mazonia LBA ) L arge Scale B atmosphere experiment in A ( LBA iosphere- -atmosphere experiment in mazonia ( )
Lagrangian experiment with SCM to study de experiment with SCM to study de- -forestation forestation Lagrangian
C5H8 flux [1e 11 m ole c. cm -2 s- 35 'actual' 30 dLAI, deforest 25 C5H8 flux, deforest 20 1] 15 10 5 0 0 12 24 36 48 60 72 Time [hr] deforest - 'actual' 200 0.4 314 310 c(Cl-sun and Ct) 150 0.3 Rg [W m-2] 306 Tsurf [K] 100 0.2 302 50 0.1 298 0 0 294 0 12 24 36 48 60 72 0 12 24 36 48 60 72 Time [hr] Time [hr] Net radiation Cl-sun Ct 'actual' deforest deforest, reduced ws
Recommend
More recommend
