89 Chapter 5 Estimation of Aerosol radiative forcing Radiative forcing due to aerosols is one of the major uncertainties in estimating anthropogenic climatic perturbations (Charlson et al., 1992; Houghton et al., 1995; IPCC 2001). This is mainly due to the large scale spatial and temporal variabilities of aerosols and the lack of adequate database existing on their radiative properties (Russell et al., 1997; Bates et al., 2006; Bhawar and Devara, 2010). Aerosol radiative forcing is generally classified as direct and indirect in which the direct aerosol radiative forcing (DRF) is induced by scattering and absorption of solar radiation in a cloud free sky. Light scattered by aerosols results in a negative radiative forcing (cooling effect) and light absorbed by the particles lead to a positive forcing (heating effect) (Haywood and Shine, 1995; Ramanathan et al., 2001a, 2001b). On a global average basis, the DRF at the top of the atmosphere induces a negative forcing, offsetting the positive radiative forcing produced by greenhouse gases. For greenhouse gases, their concentration, distribution and radiative properties are well known, while the quantification of aerosols is still uncertain (IPCC, 2007). In the indirect radiative forcing aerosols modify the microphysical properties of clouds and thereby modulate the radiative properties of clouds and cloud life time (Twomey, 1977; Penner et al., 2004). If absorbing aerosols are present in higher atmospheric layers radiative heating on these layers can change the temperature gradient, and low level clouds are evaporated (Hansen et al., 1997). The gravity of this effect depends on the altitude of clouds (Satheesh, 2002; Satheesh and Moorthy, 2005). When a cloud layer is present above aerosols most of the incident
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
89
Chapter 5
Estimation of Aerosol radiative forcing
Radiative forcing due to aerosols is one of the major uncertainties in estimating
anthropogenic climatic perturbations (Charlson et al., 1992; Houghton et al., 1995; IPCC
2001). This is mainly due to the large scale spatial and temporal variabilities of aerosols
and the lack of adequate database existing on their radiative properties (Russell et al.,
1997; Bates et al., 2006; Bhawar and Devara, 2010). Aerosol radiative forcing is
generally classified as direct and indirect in which the direct aerosol radiative forcing
(DRF) is induced by scattering and absorption of solar radiation in a cloud free sky.
Light scattered by aerosols results in a negative radiative forcing (cooling effect) and
light absorbed by the particles lead to a positive forcing (heating effect) (Haywood and
Shine, 1995; Ramanathan et al., 2001a, 2001b). On a global average basis, the DRF at
the top of the atmosphere induces a negative forcing, offsetting the positive radiative
forcing produced by greenhouse gases. For greenhouse gases, their concentration,
distribution and radiative properties are well known, while the quantification of aerosols
is still uncertain (IPCC, 2007). In the indirect radiative forcing aerosols modify the
microphysical properties of clouds and thereby modulate the radiative properties of
clouds and cloud life time (Twomey, 1977; Penner et al., 2004). If absorbing aerosols are
present in higher atmospheric layers radiative heating on these layers can change the
temperature gradient, and low level clouds are evaporated (Hansen et al., 1997). The
gravity of this effect depends on the altitude of clouds (Satheesh, 2002; Satheesh and
Moorthy, 2005). When a cloud layer is present above aerosols most of the incident
90
radiation will be reflected back and only a small fraction interact with aerosols, while the
cloud layer is below the aerosol layer, the aerosols modify not only the incident radiation
but also the reflected radiation from the cloud layer below. The various radiative
mechanisms associated with cloud mechanisms are depicted in figure.5.1.
Figure 5.1: Diagrammatic representation of radiative mechanisms associated with cloud effects. The small black dots represent aerosols and open circles cloud droplets. CDNC refers to the cloud droplet number concentration and LWC the liquid water content (IPCC, 2007)
Aerosol radiative forcing at any layer is defined as the difference in the net fluxes
(down minus up) with and without aerosols, present in that layer. The effect of aerosols
on the top of the atmosphere (TOA) radiative flux is TOA radiative flux. Similarly, the
effect of aerosols on the surface radiative flux is surface (S) radiative forcing. The
difference between TOA forcing and surface forcing is atmospheric radiative forcing.
Hence
, , ,( ) ( )S TOA a a S TOA o o S TOAF f f f fΔ = ↓ − ↑ − ↓ − ↑ (1)
91
Where fa↓ and fa↑ denotes the downwelling and upwelling irradiance (W m-2) with
aerosols and fo↓ and fo↑ denotes the respective quantities (in W m-2) without aerosols
and,
( )TOA S ATMF F FΔ −Δ = Δ (2)
There are several methods for the estimation of aerosol radiative forcing. Aerosol
samples collected on filter papers are chemically analyzed to obtain the mass
concentration of various aerosol species. These are then converted to number distribution
and subsequently to optical depths using Mie scattering theories and forcing are
calculated using suitable radiative models (Satheesh and Srinivasan, 2006). Since the
surface aerosol properties are quite different from column aerosol properties the
estimated forcing magnitudes can cause large errors. Alternately the measured radiant
flux under clear sky conditions is subtracted from the stimulated aerosol free radiative
flux to estimate aerosol radiative forcing. Once the aerosol properties like AOD,SSA and
ASP are precisely retrieved aerosol radiative forcing can be computed with the aid of
efficient software tools like SBDART model (Ricchiazzi et al., 1998) or RRTM_SW
(Rapid Radiative Transfer Model Shortwave), (Clough et al., 2005). Aerosol radiative
forcing primarily depends on variables affecting the environment like surface albedo and
on the vertical distribution of aerosols. Several studies have revealed global radiative
forcing by individual aerosol components, such as sulphate (Penner et al., 1998),
carbonaceous aerosols (Chung and Seinfeld, 2002), sea salt (Gong et al., 1997), and
mineral dust (Tegen 2003). Multiple aerosol components have also been simulated
simultaneously in global models (Jacobson, 2001). In this section we present the
instantaneous aerosol radiative forcing and its sensitiveness to various parameters with a
simple analytical model as suggested by Haywood and Shine (1995). The direct Aerosol
92
Radiative Forcing in the short wave range (0.25 - 4µm) has also been estimated on clear
sky days in the month of April, August, October and December using SBDART model.
5.1 Simple analytical model
An aerosol layer can scatter and absorb solar radiation. Scattering aerosols cool
the atmosphere whereas the absorbing aerosols heat it. The overall effect of the aerosol
layer is decided by the surface reflectance and the nature of aerosols.
5.1.1 Theory
For an aerosol layer of optical depth τ and if we assume that the solar beam is
directly overhead, the fraction of the incident beam transmitted through the layer is .
The fraction of light getting reflected back in the direction of the beam
(1 )zr e ωβ−= − (3)
where ‘ω’ is the single scattering albedo and ‘β’ the back scattering fraction. The fraction
of light absorbed within the layer
(1 )(1 )za eω −= − − (4)
and the fraction scattered downward is
(1 )(1 )zs eω β −= − − (5)
The total fraction of radiation that transmits downward is
(1 )(1 )z zt e eω β− −= + − − (6)
93
If Rs is the reflectance of the Earth’s surface, the fraction of radiation reflected is Rs * t.
On the first upward pass of the reflected beam the fraction transmitted is also‘t’ and
hence The fraction of this beam reflected back to Earth is ‘r* Rs * t’, and transmitted
through the aerosol layer is t2 Rs. This process continues and for Rs and r < 1, the
change in reflectance due to the presence of aerosol layer is
2
1s
p ss
t RR r RR r
⎡ ⎤Δ = + −⎢ ⎥−⎣ ⎦ (7)
The sign of the change in planetary albedo due to presence of aerosol layer ∆Rp
determines whether the forcing is negative (cooling) or positive (heating). The key
parameter leading to cooling or heating is the SSA (ω) and the magnitude of ω at which
∆Rp = 0 defines the boundary between heating and cooling. For small τ it can be shown
that this boundary value ω crit is
2
22 (1 )
scrit
s s
RR R
ωβ
=+ − (8)
Figure 5.2 shows the dependence of ω crit on surface reflectance Rs and backscattering
factor β. For a mean surface albedo about 0.3, and for a representative value of the
spectrally and solar zenith angle averaged β of about 0.29, the critical value of ω is about
0.65. If Ta denotes the fractional atmospheric transmittance above the aerosol layer
which differ from unity due to Rayleigh scattering and absorption by ozone and other
gases of the atmosphere and F0 the incident flux, the change in the outgoing radiative
flux as a result of an aerosol layer underlying an atmospheric layer is
iaer = 5 User defined boundary layer aerosol. If iaer = 0, no boundary layer aerosol.
wlbaer = .340,.440,.675,.870,1.020 Aerosol optical depths are measured at five short band wavelength 340,440,657,870 and 1020 nm
gbaer = .412,.218,.147,.130,.110 AOD at wavelengths 340, 440, 675, 870 and 1020 nm
wbaer = 5*.9 Single scattering albedo 0.9
gbaer = 5*.8 Asymmetry parameter for the five wavelengths mentioned above, calculated from linear fit
zout=0,100 Get output at the surface and top of the atmosphere
iout = 10
One output record per run, integrated over wavelength. Output quantities are (integration by trapezoid rule) WLINF, WLSUP, FFEW, TOPDN, TOPUP, TOPDIR, BOTDN, BOTUP, BOTDIR
nstr =4 Number of discrete ordinates in DISORT (four polar angles and four azimuthal mode)
Table 5.4: Example of input.dat files for SBDART
105
where,
WLINF is the lower wavelength limit in microns.
WLSUP, the upper wavelength limit in microns.
FFEW, filter function equivalent width in microns.
TOPDN, total downward flux at ZOUT (100) km. in W/m2.
TOPUP, total upward flux at ZOUT (100) km. in W/m2.
TOPDIR, direct downward flux at ZOUT (100) km. in W/m2.
BOTDN, total downward flux at the surface. in W/m2.
BOTUP, total upward flux at the surface. in W/m2 and
BOTDIR, direct downward flux at the surface. in W/m2.
Additional input parameters like integrated water vapor amount (UW), integrated ozone
concentration (UO3), cloud parameters, cloud radius, etc. can be provided as input
parameters. The downward and upward SW flux at the surface and top of the atmosphere
were computed with and without aerosols. Thus Shortwave, clear sky radiative forcing at
the surface(S) and the top of the atmosphere (TOA) are estimated as
, , ,( ) ( )S TOA a a S TOA o o S TOAF f f f fΔ = ↓ − ↑ − ↓ − ↑ (21)
(∆FTOA - ∆FS) gives ∆FATM the net atmospheric forcing. This energy gets converted into
heat thereby resulting in atmospheric heating, which is the indicator of climatic impact of
aerosols. The atmospheric heating rate have been calculated (Liou, 2002) as
106
ATM
p
FT gt C P
Δ∂=
∂ Δ (22)
where T/ t is the heating rate, g the acceleration due to gravity Cp the specific heat
capacity at constant pressure and ∆P the change in atmospheric pressure.
5.2.3 Results and discussion
The aerosol radiative forcing in the short wave region ranging from 0.25 -4.0µm
under clear sky days have been computed for four representative months in 2010.
Calculations of aerosol radiative forcing have been performed separately, with and
without aerosols at hourly intervals and 24 hour averages have been taken to estimate the
direct radiative forcing. The four months April, August, October and December are the
representatives of summer, monsoon, post-monsoon and winter seasons. The radiative
forcing results and the atmospheric absorption translated into atmospheric heating have
April -1.5±0.4 -23.7±2.3 22.2±2.7 0.62 August -2.58±0.6 -14.80±1.3 12.22±1.9 0.34 October -2.68±0.4 -14.98±1.1 12.3±1.5 0.34
December 0.09±0.07 -18.12±1.8 18.91±1.89 0.53 Table 5.5: Aerosol radiative forcing over Kannur and corresponding heating rate/day ARF at any location is dependent on many parameters like total columnar AOD,
their vertical distribution, SSA, their size distribution, asymmetry factor, surface
reflectance, relative humidity and many other factors (George, 2001). The magnitude of
TOA forcing is slightly positive (0.09 W m-2) in December, and negative in April (-1.5
W m-2) August (-2.58 W m-2) and October (-2.68 W m-2).The surface forcing is negative
107
in all the four months. It varies from -14.80 W m-2 in August to -23.7 W m-2 in April.
Nearly equal forcing has been estimated in the Month of August and October. The
seasonal variation of aerosol forcing at the top of the atmosphere, atmosphere, and
surface is depicted in figure 5.4. The forcing at the surface leads to cooling of the
surface, but within the atmosphere, it results in heating. The significant difference
between the TOA and surface forcing is due to the absorptive properties of the aerosols,
and is a measure of the heating rate of the atmosphere. The maximum surface forcing in
the month of April may be attributed to maximum values of AOD during these months.
This variation in AOD is mainly due to the aerosol loading over this area from the
neighbouring polluted areas and also due to some local influences like firework festivals.
Almost equal values of aerosol forcing have been identified in the monsoon and post
monsoon seasons because the rain continues from June to November and wash out of
aerosols takes place. Moreover during these months the marine influence dominates and
more sea salts aerosols are injected into the atmosphere. In the month of December even
though the AOD values are low,the slightly low value of SSA make the aerosol forcing
positive at the top of the atmosphere.
The atmospheric absorption translates into atmospheric heating show that the
heating rate is 0.62Kday-1 in April and and is about half during ( 0.34Kday-1 ) monsoon
season. Even though the heating rate is small this is capable of influencing the monsoon
pattern (Manoj et al., 2010). Aerosol radiative forcing is a strong function of aerosol
optical depth. Hence it is significant to calculate the forcing efficiency that is the rate at
which the atmosphere is forced per unit optical depth. It is calculated by dividing the
forcing value by AOD at 500 nm and is an indicator of the forcing potential of the
composite aerosols. The values of forcing efficiency during different months are shown
in table 5.6.
108
Month
Radiative forcing efficiency (W m-2 τ-1)
TOA Surface Atmosphere
April -4 -63.2 59.2
August -9.34 -53.62 44.2
October -10.22 -57.17 49.9
December +2 -95.87 100 .
Table 5.6: Radiative forcing efficiency for different seasons
Month
Rad
iativ
e fo
rcin
g (W
/m2 )
-30
-20
-10
0
10
20
30Top of the atmosphere Surface
Atmosphere
April August October December
Figure 5.4: Seasonal variations in aerosol radiative forcing
The atmospheric forcing efficiency is maximum in the winter month December and
minimum in the summer month April, which may be attributed to the low value of AOD
in winter than summer. The forcing efficiency at the top of the Atmosphere is maximum
in the month October and minimum during the winter month December. This is due to
domination of sea salt aerosols during the monsoon season over this region.
109
5.2.4 Comparison with other geographical locations
Comparison of aerosol radiative forcing at different locations is shown in the table 5.7.
Location Period TOA (Wm-2)
Surface (Wm-2)
ATM (Wm-2) Reference
Mohal-Kullu
April-July -6.55 -40.51 33.96
Guleria et al., 2010
Aug.-Sept -6.89 -32.36 25.47
Oct.-Nov 10.62 -29.11 18.48
Dec.-Mar -10.36 -28.31 18.00
Vishakapatanam
Mar-May 3.99 -16.8 20.78
Sreekanth et al., 2007
June-Aug 2.36 -9.9 12.26
Sept.-Oct 0.7 -2.81 3.51
Nov-Feb 8.4 -35.78 44.18
Ahmedabad
April-May 8 ± 2 -41.4±5 48±7 Ganguly and Jayaraman 2006
June-Sept 14 ± 4 -41 ± 11 55.5 ± 15
Oct-Nov -22 ± 3 -63 ± 10 40 ± 11
Dec-Mar -26 ± 3 -54 ± 6 28 ± 9
Trivandrum
April-May 0.3 to -1.4 -37.4 to 34.2 37.6 to 32.8 Babu et al.,
2007 June-Sept -1.4 to -2.6 -26.9 to 24.4 25.5 to 21.8
Oct-Nov -1.5 to -2.8 -30.2 to 27.8 28.7 to 25
Dec-Mar 4.1 to 1.8 -48.9 to 44.8 52.9 to 46.6
Table 5.7: Aerosol radiative forcing at different locations
Comparing our results with those reported from other geographical areas of India,
it is seen that TOA forcing is negative in Mohal Kullu, mostly negative in Trivandrum,
whereas it was positive over Vishakapattanam for all the seasons. In Ahmadabad it is
positive during summer seasons and negative during post monsoon and winter seasons.
The magnitude of TOA forcing, is more or less comparable over the four locations,
except at Ahemadabad. The minimum and maximum values of TOA, surface and
atmospheric forcing are lower than that at other regions except that at Vishakapattanam.
110
This variation is attributed not only to the diversity of aerosols, but also to the surface
reflectance and weather conditions, which in turn highly influence the behavior or
aerosols, also.
Aerosol radiative forcing during the dust events over New Delhi indicates a
consistent increase in surface cooling ranging from -39 W m-2(March) to -99 W m-2
(June) and an increase in heating of the atmosphere from 27 W m-2(March) to123 W m-2
(June) (Pandithurai et al., 2008). This was attributed due to the rise in AOD from 0.55
(March) to 1.2 (June) at 500nm and the decrease in SSA from 0.84 to 0.74. The results
indicate a strong influence of absorbing aerosols over these regions during summer.