atmosphere showing areas of major absorption bands by atmospheric species. ... the earth's surface more than the red and the sky appears to be blue. Rayleigh scattering ... the UV-Visible range is presented in Figure. 3. The most ... In general, the most important light absorbing particles are those of âelemental carbonâ also ...
VISIBILITY AND PARTICULATE AIR POLLUTION Philip K. Hopke Center for Air Resources Engineering and Science Clarkson University Potsdam, NY 13699-5708
INTRODUCTION Airborne particulate matter has a number of adverse effects on human health and the environment. One of the major environmental impacts comes through the interaction of particles with light leading to reduced ability to see objects at a distance. Historically, “visibility” has been defined as “the greatest distance at which an observer can just see a black object viewed against the horizon sky.” An object is usually referred to as at threshold contrast when the difference between the brightness of the sky and the brightness of the object is reduced to such a degree that an observer can just barely see the object. This measure of visibility is also called the visual range. The visual range is primarily limited by the concentration of airborne particles that can scatter and absorb light. Figure 1 shows the various ways that light interacts within the atmosphere to affect the ability to observe an object at a distance. Thus, it is useful to understand the nature of solar radiation and its relationship between visibility, atmospheric gases, and the concentrations of airborne particulate matter in various size ranges.
SOLAR RADIATION The sun behaves as a black body radiator with a temperature of approximately 5250EK. The spectrum of solar energy at the top of the atmosphere and penetrating the atmosphere to sea level is depicted in Figure 2. The black body radiation is modified primarily by the absorption of light by ozone, water, and CO2 and to a lesser extent by NO2. The amount of light at any position on the earth is a function of location, time-of-day and day-of-the-year.
Figure 1. Important factors involved in seeing a scenic vista are outlined. Imageforming information from an object is reduced (scattered and absorbed) as it passes through the atmosphere to the human observer. Air light is also added to the sight path by scattering processes. Sunlight, light from clouds, and ground-reflected light all impinge on and scatter from particles located in the sight path. Some of this scattered light remains in the sight path, and at times it can become so bright that the image essentially disappears. A final important factor in seeing and appreciating a scenic vista are the characteristics of the human observer. Figure taken from Malm (1999).
Figure 2. Solar spectrum at the top and bottom of the atmosphere showing areas of major absorption bands by atmospheric species.
EXTINCTION The attenuation of light is defined by Beer’s law that can be written as
Where I is the light intensity traversing a distance x, I0 is the incident light intensity, and the fractional attenuation of light per unit distance is known as the light extinction coefficient, bext. The light extinction coefficient, bext, is expressed in units of one over length, for example inverse kilometers (km-1) or inverse megameters (Mm-1). Extinction occurs from four different processes: scattering of light by molecules, absorption of light by molecules, scattering of light, and absorption of light by particles. Each of these processes involve the interaction of the electromagnetic energy in the light with matter.
The light-extinction coefficient, bext (expressed
as inverse megameters, 1/Mm), is the sum of
where scattering by gases in the atmosphere, bsg, is described by the Rayleigh scattering theory [vandeHulst, 1957] and will be referred to as Rayleigh scattering. Scattering by particles, bsp, is caused by both fine and coarse aerosol species and is the largest contributor to total light extinction in most locations [Malm et al., 1994]. Absorption due to gases, bag, is primarily due to nitrogen dioxide (NO2) and is assumed to be negligible because almost all monitoring sites are in rural locations [Trijonis and Pitchford, 1987]. Absorption by particles, bap, is caused primarily by carbon containing particles.
Scattering of Light by Molecules Molecules are small compared to the wavelengths of visual light. Typical gas molecules have diameters of the order of a few tenths of a nanometer and the wavelength of visible light ranges from 400 to 700 nm. Thus, as the light propagates through the atmosphere, the electric field across the molecule is essentially constant in strength, but changing as the wave moves. Thus, the dielectric molecule is subjected to a time-varying field that gives rise to oscillations within the molecule resulting in changes in the direction of the light. This elastic scattering process is called Rayleigh scattering and provides the fundamental limits to the transmission of light through the air. The intensity of light scattered by a single small particle from a beam of
unpolarized light of wavelength ë is given by:
where R is the distance to the particle, è is the scattering angle, n is the refractive index of the particle, and d is the diameter of the particle. It can be seen that there is a strong wavelength dependence to the scattering such that blue (400 nm) light is scattered more effectively than red (700 nm) light. Thus, the sunlight is scattered in the atmosphere such that the blue is directed to the earth’s surface more than the red and the sky appears to be blue. Rayleigh scattering determines the maximum possible visible range and depends on the density of the air. Thus, visual range increases with altitude as the pressure decreases.
Absorption by Gases The only gas with significant absorption in the range of visible light is NO2. The absorption spectrum of NO2 in the UV-Visible range is presented in Figure 3. The most significant absorption is in the 400 to 500 nm range and at typical ambient concentrations, absorption by NO2 does not contribute significantly to the atmospheric extinction. However, the absorption of light below 430 nm is important as a source of atomic oxygen that can then react with molecular oxygen to produce ozone in the
Figure 3. Absorption spectrum of NO2 in the range of 200 to 800 nm. Figure drawn using the data of Bogumil et al. (2003).
lower troposphere.
Scattering by Particles The largest source of light reduction in the atmosphere is scattering by airborne particles. In this case, the scattering is by particles that have diameters of the same order as the
wavelengths of light. This scattering is commonly termed Mie scattering after Gustav Mie who first report solving Maxwell’s equations to the interaction of an electromagnetic wave with an isotropic, homogeneous, dielectric sphere. It cannot be solved in closed form as was the case for Rayleigh scattering. For angular scattering, with I0, representing the intensity of light incident on a particle and I(è) being the intensity scattered into unit solid angle at an angle of è (è=0 is forward scattering):
where á = ðd/ë and Mie theory permits calculation of the factors i1(è) and i2(è). The subscripts 1 and 2 represent the perpendicular and plane polarized cases, respectively. Figure 1 shows a typical plot of these scattering integrals as a function of á for a fixed refractive index. To obtain the scattering from a distribution of particles having the same refractive index, this equation can be summed over a series of K discrete, small wavelength intervals.
where Nk is the number of particles per unit volume in size bin k. Figure 4 gives the volumespecific light scattering efficiency in units of ìm-1 as a function of particle diameter. The light scattering coefficient is derived by multiplying the volume-specific light scattering efficiency factor by the volume concentration. The mass specific light scattering efficiency can be obtained by dividing the values for the curves by the density of the PM. It can be seen that the scattering is strongest in wavelengths near the wavelength of light and decreases to a relatively constant small value for large particles. Thus, it is the particles in the 0.1 to 1.0 ìm size range that are most effective is scattering light. Mie scattering is relatively insensitive to the wavelength of light and thus, cloud droplets
scatter all wavelengths approximately equally and appear to be white.
Figure 4. Volume-specific light-scattering efficiency as a function of particle diameter for various refractive indices and a wavelength of 550 nm.
Absorption by Particles The refractive index for a material is actually a complex number where the real portion governs the scattering and the imaginary portion governs the light absorption.
where i =
`. In general, the most important light absorbing particles are those of “elemental
carbon” also referred to as black carbon, light absorbing carbon, graphitic soot and a number of other names. There is also light absorption by other particles. For example, Asian dust has a strong yellow color and is often referred to as yellow sand or yellow dust. Saharan dust is red in color. Depending on the elemental composition of the particle, it can have light absorption resulting in color. However, generally the absorption by these dust particles is significantly weaker than that from black carbon. A major problem in assessing the light absorption properties of elemental carbon is the fact that the mass absorption coefficient is dependent on the
source of the particulate matter and how much atmospheric processing it has undergone from its emission into the air. Thus, the value of bap is quite variable and hard to estimate a priori.
VISUAL RANGE According the Koschmieder (1924a,b) theory, the visual range of an object viewed against the horizon sky, VR (km), is inversely proportional to the horizontal extinction coefficient, bext (km-1), bext = K/VR. where K is the Koschmieder constant. The Koschmieder constant, K, depends on the contrast threshold sensitivity (2-5%) of the human eye as well as on the inherent contrast of the visible objects against the horizon sky (Middleton, 1952). If a threshold contrast of 2% is assumed, and bext is in Mm-1, the Koschmieder constant becomes 3.912. The limitations in visual range estimates include the observers' visual acuity, the number, configuration, and physical and optical properties of the visible targets. The observer's subjectivity imposes a random component on the observed signal. The lower contrast of real targets compared to black objects imposes a systematic underestimate of visual range. In addition, visibility is reported in quantized units, depending on the availability of visible targets, i.e. an observation of 10 miles means that the visual range is greater than 10 miles. Thus, the reported visual range is always an underestimate of the actual visual range compared to ideal black target conditions.
VISUAL RANGE, MASS AND SPECIES CONCENTRATIONS The visual range has been found in a number of cases to be directly related to the fine particle mass concentrations. Figure 5 shows the results of Chow et al. (2002) based on data reported by Samuels et al. (1973). According to Samuels et al. (1973), there was a direct correlation between particle mass concentration, light scattering, and visibility. Husar et al. (2000) were able to develop a reasonable estimate of fine PM mass (haze) from visibility data on a worldwide basis using visibility data from airports. They propose additional filters in three broad categories: (1) filters to assess the validity of individual data points, (2) filters based on statistics for specific stations, and (3) filters based on spatial analysis of the data.
Figure 5. Plot of bext as estimated by the Koschmeier equation compared to the measured fine particle mass. (from Chow et al., 2002). Malm, 2000 included calculated aerosol light extinction for each of the five major fine fraction particle (PM2.5) components, coarse fraction mass (PM10-2.5), and Rayleigh scattering by gases and summed them for an estimate of total light extinction in Mm-1 using the following algorithm: bext = 3 x f(RH) x [Sulfate] + 3 x f(RH) x [Nitrate] + 4 x [Organic Mass]+ 10 x [Elemental Carbon] + 1 x [Fine Soil] x 0.6 x [Coarse Mass] + 10
where f(RH) is a water growth factor that is a function of RH. Using this algorithm, Malm compared the measured bext with that predicted by this equation. Figure 6 presents these results. Pitchford et al. (2007) have suggested a revised algorithm that is more consistent with the recent atmospheric aerosol literature and reduces bias for high and low light extinction extremes. The revised algorithm differs from the original algorithm in having a term for estimating sea salt light scattering from Cl ion data, using 1.8 instead of 1.4 for the mean ratio of organic mass to measured organic carbon, using site-specific Rayleigh scattering based on site elevation and mean temperature, employing a split component extinction efficiency associated with large and small size mode sulfate, nitrate and organic mass species, and adding a term for nitrogen dioxide (NO2) absorption for sites with NO2 concentration information. The revised IMPROVE
algorithm contains f(RH) curves for small- and large-mode ammonium sulfate that are also applied to small and large mode ammonium nitrate. The revised algorithm is given as
bext = 2.2 x fS(RH) x [Small Sulfate] + 4.8 x fL(RH) x [Large Sulfate] + 2.4 x fS(RH) x [Small Nitrate] + 5.1 x fL(RH) x [Large Nitrate] + 2.8 x [Small Organic Mass] + 6.1 x [Large Organic Mass] + 10 x Elemental Carbon + 1 x Fine Soil + 1.7 x fSS(RH) x [Sea Salt] + 0.6 Coarse Mass + Rayleigh Scattering (Site Specific) + 0.33 [NO2 (ppb)]
Figure 6. Scatter plot of the original IMPROVE algorithm estimated particle light scattering compared to the measured values. The results from applying this revised algorithm are shown in Figure 7. It can be seen that there is a reduction in the bias, but there is a somewhat lower R2 value. In order to utilize this algorithm, substantial additional data is required including NO2 and Cl concentrations and a more complex relative humidity. Thus, it will not always be possible to apply this approach depending on available data.
Figure 7. Scatter plot of the estimated bext based on the revised Pitchford et al. (2007) algorithm compared to the IMPROVE measured data. CONCLUSIONS Prior work has suggested that it is possible to develop relationships between measured visual range and fine particle mass concentrations. In addition, bext can be estimated with reasonable accuracy given a sufficiently complete chemical characterization of the PM2.5 such as is available with ion beam analysis.
REFERENCES Bogumil, K., Orphal, J., Homann, T., Voigt, S., Spietz, P., Fleischmann, O.C., Vogel, A., Hartmann, M., Bovensmann, H., Frerick, J., Burrows, J.P., 2003. Measurementsof molecular absorption spectra with the SCIAMACHY pre-flight model: Instrument characterization and reference data for atmospheric remote sensing in the 230-2380 nm region, J. Photochem. Photobiol. A.: Photochem. 157, 167-184.
Chow, J. C., Bachmann, J. D., Wierman, S. S. G., Mathai, C. V., Malm, W. C., White, W. H., Mueller, P. K., Kumar, N., Watson, J. G., 2002. Visibility: science and regulation. J. Air Waste Manage. Assoc. 52, 973-999.
Husar, R.B., Husar, J.D., Martin, L., 2000. Distribution of continental surface aerosol extinction based on visual range data, Atmospheric Environ. 34, 5067-5078.
Koschmieder, V.H., 1924a, Theorie der horizontalen Sichtweite, Beitr. Phys. frei. Atmos. 12, 3353.
Koschmieder, V.H. 1924b. Theorie der horizontalen sichtweite – II. Kontrast und sichtweite; Beitr. z. Phys. d. fr. Atmosph. 12, 171-181.
Malm, W. C., 2000. Spatial and seasonal patterns and temporal variability of haze and its constituents in the United States. Report III. Fort Collins, CO: Cooperative Institute for Research in the Atmosphere, Colorado State University. Available: http://vista.cira.colostate.edu/improve/Publications/Reports/2000/2000.htm
Malm, W.C., Sisler, J.F., Hu!man, D., Eldred, R.A., Cahill, T.A., 1994. Spatial and seasonal trends in particle concentration and optical extinction in the United States. Journal of Geophysical Research 99, 1347-1370.
Mie, G., 1908. Beiträge zur Optik trüber Medien, speziell kolloidaler Metallösungen, Annalen
der Physik 330, 377-445.
Pitchford, M., Malm, W.C. , Schichtel, B., Kumar, N., Lowenthal, D., Hand, J., 2007. Revised Algorithm for Estimating Light Extinction from IMPROVE Particle Speciation Data J. Air & Waste Manage. Assoc. 57:1326–1336,
Samuels, H. J., Twiss, S., Wong, E. W., 1973. Visibility, light scattering and mass concentration of particulate matter: report of the California Tri-City Aerosol Sampling Project. Sacramento, CA: State of California, Air Resources Board.
Trijonis, J.C., Pitchford, M., 1987. Preliminary extinction budget results from the RESOLVE program. In: Bhardwaja, P.S. (Ed.), APCA Transactions 10: Visibility Protection. Pittsburgh, PA.
van de Hulst, H.C., Light Scattering by Small Particles, Wiley, New York, 1957.
