MODMEP results

Progress reports of the MODMEP-projects

MODMEP results

Development of size-resolved cloud models with detailed microphysics and complex multiphase chemistry
FEBUKO simulations
A. Microphysics
B. Multiphase chemistry

Development of adapted multiphase mechanisms and of tools for their implementation into complex models
Mechanism development and reduction
CAPRAM development
Reduction of CAPRAM 2.4 by the ISSA method

Coupling and time integration
Coupling scheme in SPACCIM
Operator splitting error

Spatial description of clouds in 3D models
VOF-method in cloud modelling

Conclusions and outlook


Development of size-resolved cloud models with detailed microphysics and complex multiphase chemistry
FEBUKO simulations (see also FEBUKO-Website)
Scenarios for parcel model runs are created from the first field campaign of the joint AFO2000 project FEBUKO. The chemical properties are derived from the chemical analysis of the aerosol particles. The presented results are obtained from initial data of October 27, 2001.
A. Microphysics

Liquid water content (left) and particle/drop number concentration (right) size distributions for a complete mountain passage at 9:00 h (top) and 11:00 h (bottom).

The figure compares the mass (left) and number (right) size distributions for 9:00 h and 11:00 h of event 1 for the whole mountain passage. The higher LWC corresponds to earlier drop formation for 9:00 h (top), larger drops, longer drop lifetime as well as larger cutoff sizes for activation. At 11:00 h, vertical velocity and supersaturation at cloud base are much higher leading to a sharp activation profile with a large gap between non-activated particles and activated drops. This gap is much less pronounced for the more gentle activation process at 9:00 h.

B. Multiphase chemistry

*Gas phase concentration profile of the OH and NO3 radicals as a function of traveling time from the upwind to the downwind station.*


*Aqueous and particle concentration profiles of the main S(VI) compounds, nitric acid and nitrate, as a function of traveling time from the upwind to the downwind station .*


*Concentration profile of the main iron compounds and of oxalic acid with dissociated forms.*

The simulations were performed with SPACCIM, considering detailed microphysics and complex multiphase chemistry. The chemical mechanism consists of the aqueous phase radical mechanism CAPRAM 2.4 coupled to the gas phase mechanism RACM. Phase transfer is implemented according to the approach by Schwartz, considering gas phase diffusion, mass accomodation coefficients and Henry's equilibrium. The presented results are for the cloud event taken place on the 27-th October, 2001, at 9 .00 UTC in the Thuringian forest. It can be seen from Figure 3 that at the respective cloud event a mixture of a night and day time radical chemistry is taking place. At the summit OH and NO3 reaches a concentration of about 3.0E+4 and 3.0E+6, respectively. From the upwind till the downwind station a sulfate production of about 4% was observed. Also, in the droplets a total nitrate production of about 10% was observed. At the beginning and at the end of the simulation, in the particles, due to a low pH value, HNO3 will undergo phase transfer to the gas phase. Most of iron can be found in the form of iron oxalato complexes. The sum oxalic acid and oxalate concentration undergoes relatively small variaions through the simulation.


Development of adapted multiphase mechanisms and of tools for their implementation into complex models
Mechanism development and reduction
CAPRAM development
(Chemical Aqueous Phase RAdical Mechanism)



*Characteristics of the CAPRAM mechanism series.*

Reduction of CAPRAM 2.4 by the ISSA method

Cloud chemical mechanism RACM/CAPRAM2.4: Schematic representation of ratio between full and general reduced mechanism produced by ISSA method contribution of scenarios to general reduced mechanism


*Deviations [%] between full and general reduced RACM/CAPRAM2.4 mechanism for target species at different scenarios.*


*Reduction results of ISSA method for cloud chemical mechanism RACM/CAPRAM2.4 and different scenarios.*


Coupling and time integration

Coupling scheme in SPACCIM

Coupling scheme between microphysics and multiphase chemistry codes The two models run separately, use their own step size control and exchange informations only every coupling time step. The microphysical variables Temp, LWC and q are interpolated linearly in time during the integration of the multiphase chemistry model. For the mass transport term, time average fluxes are generated during the microphysical integration over the coupling time interval.

Implicit integration of multiphase chemistry
BDF method with special sparse solver for linear systems (Wolke and Knoth, 2002) (modified Meis-Markowitz strategy, Schur complement approach, approximate matrix factorization).

Fully-coupled solution of model equations
Uniform description of microphysical and chemical processes, discontinuous Galerkin for discretisation of mass spectrum and explicit-implicit Runge-Kutta time integration schemes.

Operator splitting error
An accurate simulation of aerosol and droplet number concentrations and mass densities, respectively, are crucial for the calculation of e.g. the radiative properties of clouds, their life time, and precipitation rates. These quantities are also important for the energy budget and the hydrological cycle in larger scale models. Since the system of microphysical and multi-phase chemical processes is very complex and computer resources demanding there is a need to simplify the numerical simulations of this system. A common methodology to solve complex systems such as the above mentioned is the application of operator splitting.

Operator splitting errors for total drop number concentration and mean drop radius due to decoupling of coagulation/breakup from the other microphysical processes and multiphase chemical reactions. Results are shown for various decoupling intervals.


Spatial description of clouds in 3D models
VOF-method in cloud modelling
TRADITIONAL MODELS: Each integration step includes reinterpretation of the new states in the grid cells as regards the phase (whether cloud or non-cloud) --> Transport unavoidably leads to arbitrary vanishing of cloud phase (numerical diffusion additionally to conversion processes)	VOLUME-of-FLUID (VOF) METHOD: The phases are treated by separate variables and corresponding reference volumes. No re-interpretation of phases --> Transport does not affect the phases (possible changes by conversion processes only)

ADVANTAGES of VOF METHOD: - Advection and diffusion together are subjected the volume-of-fluid method - Cloud development and transport free from numerical dispersion - Enhancement of cloud intensity and life-time - Promotion of sparse grid resolution and low run-time


Conclusions and outlook
The simulations indicate that the developed microphysical as well as coupled models are helpful tools for the interpretation of FEBUKO measurements. It leads to an understanding of the complex processes underlying the observations at two valleys and a mountain station in a hill-capped-cloud experiment. On the other side, the comparison between model results and measurements entails many improvements in models, a high quality in model initialisation and an evaluation of the quality of simulation results. More detailed results are given in the FEBUKO presentations. The mechanism development performed its aim at a more complete chemical scheme for understanding cloud chemistry and aerosol-cloud interactions. The chemical mechanism is also completed to cover organic chemistry of compounds with more than two carbon atoms. Due to the resulting very complex mechanisms, within MODMEP also mechanism reduction techniques are developed and applied for the different versions of CAPRAM mechanisms. The SPACCIM model is reproducing the phase interactions and chemical conversions occurring in aerosol-cloud-interaction. SPACCIM accommodates mechanisms given in literature as well as different versions of CAPRAM, i.e. C2.3, C2.4 and in the very near future CAPRAM 3.0. It is investigated whether and which processes can be integrated decoupled with which loss of exactness and which gain of efficiency. A box model for a fully-coupled solution of the microphysical and chemical model equations is developed which uses the discontinuous Galerkin method for the discretisation of the mass spectrum and explicit-implicit Runge-Kutta time integration schemes. Customary microscale models for the simulation of cloud formation and transport generally adapt Eulerian grids. The advective transport in such a grid model implies steady distributions of the variables and smoothes any discontinuity by numerical diffusion. However, clouds distinguish from their environment by steep gradients in temperature, water content etc. This discrepancy in cloud modelling is broken by applying the volume-of-fluid (VOF) method to an existing 3D non-hydrostatic model. The developed multiphase mechanisms, the coupling scheme and the techniques for the treatment of multiphase chemistry can be integrated in other parcel or multi-dimensional modelling systems. Therefore, selected mechanisms, tools and modules will be provided for the community on the MODMEP Webpages in next time.

Last change: 2004/04/13