Advanced Particle Microphysics (APM)
The Advanced Particle Microphysics (APM) package was developed for implementation into GEOS-Chem at State University of New York (SUNY) at Albany (Yu and Luo [2009]). The APM model is optimized to accurately simulate secondary particle (SP, composed of sulfate, nitrate, ammonium, and SOA) formation and their growth to CCN sizes, with a higher size resolution for the size range of importance (\(1.2 - 120 nm\)): 30 bins, 10 additional bins for \(120 nm - 12 \mu m\)). The present version of the APM employs 20 bins for sea salt to cover the dry diameter size range of \(0.012 \mu m\) to \(12 \mu m\), and 15 bins for dust particles to cover size range of \(0.03 \mu m\) to \(50 \mu m\). Because of the large differences in the median sizes of black carbon (BC) and primary eorganic carbon (POC) from fossil fuel combustion and biomass burning, we employ two log-normal modes (one for fossil fuel and another for biomass burning) to represent hydrophobic BC and two other log-normal modes for hydrophilic BC. Similarly, 4 log-normal modes are used to represent hydrophobic and hydrophilic POC. The growth of nucleated particles through the condensation of sulfuric acid vapor and equilibrium uptake of nitrate, ammonium, and secondary organic aerosol is explicitly simulated, along with the scavenging of secondary particles by primary particles (dust, black carbon, organic carbon, and sea salt). The amounts of secondary species coated on primary particles (through condensation, coagulation, equilibrium uptake, and aqueous chemistry) are tracked.
Implementations of the aerosol optical properties look up table (Yu et al. [2012]) and radiation transfer (RF) model (Ma et al. [2012]; Yu et al. [2013]) enable GEOS-Chem/APM to derive aerosol direct radiative forcing and first indirect radiative forcing.
Author |
Institution |
|---|---|
SUNY Albany |
|
Gan Luo |
SUNY Albany |
Computational Information
The APM model contains a number of computationally efficient schemes:
Usage of pre-calculated lookup tables for nucleation rates and coagulation kernels;
Variable size ranges for particles of different types;
Variable bin resolution;
Variable and optimized time steps for the coagulation calculations;
The coating of primary particles by sulfate is tracked using one tracer (sulfate mass) for each type of primary particles;
Nitrate, ammonium, and SOAs associated with sulfate are calculated based on the equilibrium partition.
The above schemes enable the APM model to capture the main properties of atmospheric particles important for their direct and indirect radiative forcing while keeping the computational costs quite low.
In the study reported in Yu and Luo [2009] all simulations are running on 8-CPU Linux workstations with the 2.2 Ghz Dual Quad-Core AMD Opteron Processor 2354. The model system was compiled using OpenMP for running in parallel. The GEOS-Chem version used in the study (v8-01-03) had 54 species, and took 24.23 hours for one year full-chemistry simulations at 4x5 horizontal resolutions and 47 layers (GEOS-5 data). GEOS-Chem v8-01-03 with APM incorporated had 127 species (73 additional species: 40 for sulfate, 20 for sea salt, one for H2SO4 gas, 4 tracers for BC/OC from fossil fuel, 4 tracers for BC/OC from biomass/bio-fuel, and 4 for sulfate attached to dust, BC, primary OC, and sea salt particles). With full size-resolved microphysics (nucleation, condensation, coagulation, deposition, and scavenging) and chemistry, it took the model (127 species) 52.35 hours for the same year simulations on the same machine. In other words, the efficient schemes allow the increase in the computing cost per 100% increase in number of tracers (associated with particle size information) to \((52.35/24.23-1)/(127/54-1) = 86%\). Such a relatively small increase in the computing cost associated with full size-resolved microphysics is desirable and makes the future coupling of APM model with global climate model feasible.
Model Details
Particle Types and Representation
Particle |
Type |
|---|---|
Sulfate |
Secondary Particle (SP) |
Nitrate / Ammonium / SOA in equilibrium |
Secondary Particle (SP) |
Black carbon (BC) |
Primary Particle [1] |
Primary organic carbon (POC) |
Primary Particle [1] |
Dust |
Primary Particle [1] |
Sea salt |
Primary Particle [1] |
Notes
Particle or type |
Number of bins or modes |
Bin range |
|---|---|---|
Secondary Particles (SP) |
40 bins |
\(1.2 nm - 12 \mu m\) |
Black carbon (BC) |
2 log-normal modes for hydrophobic BC |
N/A |
Primary organic carbon (POC) |
2 log-normal modes for hydrophobic OC |
N/A |
Dust: |
15 bins |
\(30 nm - 50 \mu m\) |
Sea salt |
20 bins |
\(12 nm - 12 \mu m\) |
Microphysics
Nucleation
Nucleation (or new particle formation) is one of the key processes connecting gas-phase chemistry to aerosol microphysics and controlling number concentrations (and size distributions) of atmospheric particles. The APM model employs ion-mediated nucleation (IMN) (Yu [2010]) and binary homogeneous nucleation (BHN) (Yu [2008]) in term of look-up tables. Other nucleation schemes can also be included.
Based on IMN mechanism, sulfuric acid vapor concentration ([H2SO4]), temperature (T), relative humidity (RH), ionization rate (Q), and surface area of pre-existing particles (S) have profound and non-linear impacts on nucleation rates. The sensitivities of nucleation rates to the changes in these key parameters may imply important physical feedback mechanisms involving climate and emission changes, chemistry, solar variations, nucleation, aerosol number abundance, and aerosol indirect radiative forcing (Yu [2010]).
Growth
H2SO4 vapor concentration is a tracer and the condensation of H2SO4 on all particles is explicitly simulated. Many field measurements indicate significant contribution of secondary organic gases (SOGs)to the growth of secondary particles. A scheme to consider the oxidation aging of SOGs and explicit condensation of low volatile SOGs has been developed (Yu [2011]).
Coagulation
Coagulation is a process in which particles of various sizes and compositions collide with each other and coalesce to form larger particles. In the atmosphere, coagulation is an important process scavenging small particles and turning externally mixed particles into internally mixed particles. In the present model, the mass conserving semi-implicit numerical scheme is employed to solve the self coagulation of size-resolved sulfate and sea salt particles, as well as the scavenging of sulfate particles by sea salt, dust, BC, and POC particles.
Coagulation is the most time-consuming process among various size-resolved microphysical processes (nucleation, growth, coagulation, and deposition). The reason is that coagulation involves particles of different sizes and thus adds two additional dimensions (size of particle A and size of particle B) into 3-dimensional spatial grid system. For example, for 40 bins of sulfate, coagulation among sulfate particles is equivalent to solving 40x40 = 1600 reaction equations. To reduce the computing cost of 3-D sectional aerosol microphysics model, it is critical to optimize the number of bins and coagulation calculation.
Aerosol optical properties
The key particle optical properties needed for radiative transfer calculation include extinction efficiency (\(Q_{ext}\)), single scattering albedo (\(\omega\)), and asymmetry parameter (\(g\)). The absorption extinction efficiency (\(Q_{abs}\)) can be calculated from \(Q_{ext}\) and \(\omega\) as \(Q_{abs} = Q_{ext} \times (1 - \omega)\). The values of \(Q_{ext}\), \(\omega\), and \(g\) depend on wavelength (\(\lambda\)), core diameter (\(d_{core}\)), shell diameter (\(d_{shell}\)), and real (\(k_r\)) as well as imaginary (\(k_i\)) components of refractive index (\(k = k_{r} - k_{i}i\)) for both core and shell, and can be calculated with widely used Mie theory. The core-shell model of Ackerman and Toon, (1981), which can use either the volume averaged refractive indices or the shell/core configuration, is employed in our study.
To reduce computation cost for 3-D online calculation, we have designed and generated lookup tables so that \(Q_{ext}\), \(\omega\), and \(g\) values can be determined efficiently. According to the properties of aerosols resolved by APM, three lookup tables have been developed: the first for particles without solid absorbing cores (i.e., secondary particles, coated sea salt, and coated POC), the second for coated BC, and the third for coated dust. For coated BC and dust particles, the core-shell model assumes that BC and dust have a spherical core, surrounded by a spherical shell composed of all the other non- or less- absorbing secondary species and water. For hydrated (i.e., wet) secondary particles, coated sea salt particles, and coated primary organic particles, we set the core size to zero and use the volume-average of refractive index to calculate the optical properties of particles of given wet sizes. For refractive index of BC core, we use the value recommended by Bond et al. [2006] which is \(1.85 – 0.71i\). For dust core, a wavelength dependent parameterization of refractive index presented in Balkanski et al. [2007] is adapted. The volume-averaged refractive indices for species other than BC and dust are calculated based on the composition predicted by APM.
Details can be found in Yu et al. [2012].
Radiative transfer (RF)
Radiative transfer model is needed for aerosol radiative forcing calculation. Ma et al. [2012] integrated the radiation transfer (RF) model of the Canadian Center for Climate Modeling and Analysis (CCCma) with GEOS-Chem/APM and used the model to study aerosol direct RF (DRF).
A more recent comparison of DRF values (clear sky vs. all sky) based on GEOS-Chem/APM with those of other AeroCom models (discussion version of Myhre et al. [2013] indicates that CCCMa RF code may have underestimated the impacts of clouds on radiation. We found out that the underestimation is likely associated with the cloud overlapping assumption. The version of CCCMa RF code we integrated into GEOS-Chem does not contain the widely used McICA (Monte-Carlo Independent Column Approximation) scheme ─ a fast, flexible, approximate technique for computing radiative transfer in an inhomogeneous cloud field.
To properly take into account the impacts of clouds on aerosol DRF and more importantly to study aerosol first indirect RF (IRF), Yu et al. [2013] integrated the widely used Rapid Radiative Transfer Model for GCMs (RRTMG) (Mlawer et al. [1997]; Iacono et al. [2003], Iacono et al. [2008]) with GEOS-Chem/APM.
RRTMG, which contains the McICA scheme, is a broadband k-distribution radiation model (Mlawer et al. [1997]; Iacono et al. [2003]; Iacono et al. [2008]) that has been widely used in community models (such as WRF, CAM5, etc.). The RRTMG for shortwave (SW) (used in this study) can calculate fluxes and heating rates over 14 contiguous shortwave bands (\(820 - 50000 cm^{-1}\), or \(0.2 - 12.20 \mu m\)). The individual band ranges (in wavenumbers, \(cm^{-1}\)) are: 2600-3250, 3250-4000, 4000-4650, 4650-5150, 5150-6150, 6150-7700, 7700-8050, 8050-12850, 12850-16000, 16000-22650, 22650-29000, 29000-38000, 38000-50000, and 820-2600, with the last band coded out of sequence to preserve spectral continuity with the longwave bands.
GEOS-Chem/APM-RRTMG results have been included into the final manuscript on radiative forcing of the direct aerosol effect from AeroCom Phase II simulations (Myhre et al. [2013]) and another manuscript on host model uncertainties in aerosol radiative forcing estimates (:cite:t`Stier_et_al._2013`). GEOS-Chem/APM-RRTMG has also been employed to study the first aerosol indirect radiative forcing (Yu et al. [2013]).
Dust Particle Size Distribution
Model-simulated annual mean dust particle mass size distributions in Beijing, normalized to total dust mass concentrations. The distribution peaks in diameter of \(~3-5 \mu m\). Based on the size distributions, about 90% of DST4 (diameter range \(6 - 12 \mu m\)) is in PM10, and 30% of DST2 (diameter range \(2 - 3.6 \mu m\)) is in PM2.5.
Validation, Application, and Development
The first application of GEOS-Chem + APM focuses on predicting the number concentrations of particles in the troposphere. Significant amount of efforts have been devoted to validate the simulated global spatial distributions of particle number abundance, using a large amount of land-, ship-, and aircraft- based measurements. See following publications for details (full citations are given in the in the References section).
The model has been applied in a number of other studies and results have been reported in the following papers:
AEROCOM intercomparison
The AEROCOM-project is an open international initiative of scientists interested in the advancement of the understanding of the global aerosol and its impact on climate. We have made a commitment to represent the GEOS-Chem community to submit aerosol microphysics simulations based on GEOS-Chem + APM.
We submitted our size-resolved microphysics simulation results for A2-CTRL-2006 (GEOS-5, 2x2.5) to AEROCOM in January, 2011. The submitted results include OC and SOA mass concentrations which have been used in AEROCOM OA intercomparison. We also managed to submit AOD, AAOD, aerosol direct radiative forcing results to AeroCom in September 2011. The results have been in the following papers:
APM in other community models
The same APM model initially developed for GEOS-Chem (Yu and Luo [2009]) has been incorporated into WRF-Chem (Luo and Yu [2011]). GEOS-Chem/APM results provide initial and boundary conditions for WRF-Chem/APM simulations.