comprehensive databases with in situ measurement data (Meskhidze et al. 2013).
Example of such data base is the data base of the University of Miami (Mishchenko
et al. 2002). It is essential to gather long data base, to compare modeled data with
observations of the seasonal variability and trends.
Other subject is a choice of parameters used in models. The main and easiest to
apply is the wind speed at 10 m above sea surface. It is the simplest meteorological
parameter to use, because of direct relation between surface wave height or
whitecap generation. Science community still discuses about using other parameters
instead. The main other parameters are (Lewis and Schwartz 2004): Atmospheric
Stability, Wind Friction Velocity, Sea Water Temperature, Wave Phase Velocity,
Fetch, Salinity, Surface-Active Substances. Atmospheric Stability strongly affects
air mass movements so there is strong dependence for marine aerosol drops coming
from whitecaps. Wind friction velocity influences not only stability of atmosphere
but also whitecap ratio. This parameter is quite easy to obtain, so it is convenient
using it during source function estimation or model execution. Sea water temperature is strongly related to the kinetic viscosity of seawater so also existing of
whitecaps also should be related. The state of sea surface has its specific inertia.
Parameter which informs about state of the sea surface is wave phase velocity.
Other parameter connected with sea surface inertia is fetch. Fetch is the distance
over water that the wind has blown and may affect the wind spectrum. There is no
direct relation between aerosol emission and salinity. However, there are differences
in observations for brackish seas and open oceans with fresh water. Surface active
substances may change the sea state, roughness length, surface tension of the
seawater-air layer and affect on the lifetime of foam on the sea surface. Unfortunately, all these relations are very difficult to parameterize. The very interesting
parameterization was presented by Ovadnevaite et al. (2014) where the Reynolds
Number was used instead of all the above parameters. Advantage of such approach
is that Reynolds Number brings information about wind speed, kinematic viscosity
of water and indirectly: wave height, wind history, friction velocity or viscosity.
The new approach for determining SSA fluxes is presented by Grythe et al.
(2013). In this paper I have reviewed 21 SSA source functions known from the
literature. For each function a global SSA emission was described. In applying this
task the FLEXPART Lagrangian particle dispersion model was used. Additionally,
the authors proposed a new source function. This function, based on modeling
estimation determined the functional relation of the SSA emission versus such
parameters as wind speed (power dependence, *u
3.5 ) or sea surface temperature
and aerosol diameter (D p < 10 μm, lognormal relation enclosing 3 aerosol modes).
This is the first function determined using modeling estimation. The comparison of
the SSA emission obtained from all 21 source functions is presented in Table 2
(Grythe et al. 2013).
44
P. Markuszewski
Example of such data base is the data base of the University of Miami (Mishchenko
et al. 2002). It is essential to gather long data base, to compare modeled data with
observations of the seasonal variability and trends.
Other subject is a choice of parameters used in models. The main and easiest to
apply is the wind speed at 10 m above sea surface. It is the simplest meteorological
parameter to use, because of direct relation between surface wave height or
whitecap generation. Science community still discuses about using other parameters
instead. The main other parameters are (Lewis and Schwartz 2004): Atmospheric
Stability, Wind Friction Velocity, Sea Water Temperature, Wave Phase Velocity,
Fetch, Salinity, Surface-Active Substances. Atmospheric Stability strongly affects
air mass movements so there is strong dependence for marine aerosol drops coming
from whitecaps. Wind friction velocity influences not only stability of atmosphere
but also whitecap ratio. This parameter is quite easy to obtain, so it is convenient
using it during source function estimation or model execution. Sea water temperature is strongly related to the kinetic viscosity of seawater so also existing of
whitecaps also should be related. The state of sea surface has its specific inertia.
Parameter which informs about state of the sea surface is wave phase velocity.
Other parameter connected with sea surface inertia is fetch. Fetch is the distance
over water that the wind has blown and may affect the wind spectrum. There is no
direct relation between aerosol emission and salinity. However, there are differences
in observations for brackish seas and open oceans with fresh water. Surface active
substances may change the sea state, roughness length, surface tension of the
seawater-air layer and affect on the lifetime of foam on the sea surface. Unfortunately, all these relations are very difficult to parameterize. The very interesting
parameterization was presented by Ovadnevaite et al. (2014) where the Reynolds
Number was used instead of all the above parameters. Advantage of such approach
is that Reynolds Number brings information about wind speed, kinematic viscosity
of water and indirectly: wave height, wind history, friction velocity or viscosity.
The new approach for determining SSA fluxes is presented by Grythe et al.
(2013). In this paper I have reviewed 21 SSA source functions known from the
literature. For each function a global SSA emission was described. In applying this
task the FLEXPART Lagrangian particle dispersion model was used. Additionally,
the authors proposed a new source function. This function, based on modeling
estimation determined the functional relation of the SSA emission versus such
parameters as wind speed (power dependence, *u
3.5 ) or sea surface temperature
and aerosol diameter (D p < 10 μm, lognormal relation enclosing 3 aerosol modes).
This is the first function determined using modeling estimation. The comparison of
the SSA emission obtained from all 21 source functions is presented in Table 2
(Grythe et al. 2013).
44
P. Markuszewski