Modified from Mankin et al. 1999 and James Crawford (personal communication).
10.2.1 In-Situ Observations of Gases
Mass spectrometry (MS). In this method, the chemical species present in air samples are ionized, then an electromagnetic field is applied that separates ions according to their charge-to-mass ratios. The detection of specific lines in the mass spectrum provides quantitative information on the chemical composition of the air injected in the instrument. If the chemical species to be measured is selectively ionized by charge transfer of injected positive or negative ions, the instrument is called a chemical ionization mass spectrometer (CIMS). For example, acids such as H2SO4, HNO3, or HCl can be ionized by charge transfer of reagent SF6–. Charge transfer from positive water ions (H+H2O) is called proton transfer reaction–mass spectrometry (PTR-MS) and provides a method to measure the atmospheric abundance of a wide range of organic species.
Gas chromatography (GC). In this method, air samples are injected in a narrow tube (GC column), and the chemical species are separated as they flow through the column and interact differently with the material in the column (Figure 10.1). The species are identified by their retention time in the column, and their concentrations are determined by a detector in the output stream (such as mass spectrometer, flame ionization, electron capture, thermo-ionic detectors). This technique is commonly used to measure organic species. If the detector is a mass spectrometer, the method is referred to as GC-MS.
Figure 10.1 (a) Schematic representation of a gas chromatograph. Reproduced with permission from Lagzi et al. (2013). (b) Example of a chromatogram for a mixture of hydrocarbons as a function of the respective chemical species retention time [minutes] in the GC column, showing the intensity of peaks and their chemical identification.
Electrochemical ozonesondes. These small, lightweight balloon-borne instruments are routinely used to measure the vertical profile of ozone. The device contains electrodes immersed in an aqueous solution of potassium iodide (KI). When ozone enters the sensor, iodine molecules (I2) are formed:
2 KI + O3 + H2O → I2 + O2 + 2 KOH
(10.1)
The conversion of iodine into iodide at the cathode of the instrument produces a weak electrical current proportional to the mass flow rate of ozone through the cell. By measuring this current and the rate at which air enters the cell, ozone concentrations can be derived. The Brewer–Mast (BM) ozonesonde operates with a single electrochemical cell that includes electrodes between which a small electrical potential is applied to prevent current flow in the cell unless free iodine is present. The ozonesonde referred to as electrochemical concentration cell (ECC) is made of two separate cells each containing slightly different concentrations of a KI solution, and connected by an ion bridge. No external electrical power is required since the driving electromagnetic force is provided by the difference in the KI concentrations. The instrument must be calibrated and temperature corrections must be applied.
Tunable-diode laser spectroscopy. Many methods to derive the concentrations of chemical species, either in the laboratory or in the atmosphere, are based on the analysis of the spectral signature resulting from the interaction of the species with radiation. In tunable-diode laser (TDL) spectroscopy, the light source is provided by a TDL whose emission wavelength is adjusted to match a characteristic absorption line of the target gas in the path of the laser beam. The gas concentration is derived from the measurement of the light intensity detected, for example, by a photodiode.
Resonance fluorescence. This method is based on the measurement of resonance radiation scattered by gas molecules that are irradiated in an instrument at a wavelength corresponding to a particular electronic transition for the gas. The induced fluorescence spectrum is analyzed to deduce the gas concentration. The concentration of OH, for example, can be derived from the laser-induced fluorescence (LIF) associated with the A2Σ+ (v′=0) → X2Π (v″=0) electronic transition near 308 nm.
Chemiluminescence. The detection of nitric oxide (NO) is based on reaction with ozone present in excess in a reaction chamber:
NO + O3 → NO2 + O2 + hν
(10.2)
The radiative emission produced by this reaction covers a broad spectrum (600–3000 nm with a maximum intensity around 1200 nm) and can be detected by a sensitive photoelectric device. Its intensity is proportional to the concentration of NO. In the presence of a mixture of NO and NO2, a measurement of the total NOx concentration can be achieved by converting NO2 to NO on a catalytic surface upstream of the reaction chamber. Similarly, a measurement of total reactive nitrogen oxides (NOy, including NOx and its oxidation products) can be made by catalytic reduction to NO followed by measurement of the NO concentration by chemiluminescence.
10.2.2 In-Situ Observations of Aerosols
In-situ sampling of aerosols is generally performed through an inlet that transports the particles to a collector or detector. In the ideal case, inlets should draw the totality of the particles in a specified size range. In reality this is not the case because particles deviate from the inlet airflow. Much effort has been devoted to the design of efficient isokinetic inlets that minimize this effect. The problem is particularly difficult in the case of aircraft sampling because of the fast and complex airflow surrounding the fuselage.
Total aerosol concentration. The total aerosol mass concentration can be determined by collecting particles on a filter and weighing the filter under controlled temperature and humidity conditions. Another technique is the β-gauge that measures the attenuation of β-radiation through a particle-laden filter. The attenuation, caused by electron scattering in the filter media, is proportional to the total number of atomic electrons, and provides therefore information on the total mass density of the sample. Aerosol number concentrations can be measured by growing particles by condensation in a supersaturated environment until they are large enough to be detected optically. This is the approach used in condensation nuclei (CN) counters.
Aerosol size distribution. The most common instrument used for counting and sizing particles is the optical particle counter. This instrument measures the amount of light scattered by individual particles as they flow through a tightly focused beam of light. The scattered light is directed to a photodetector. The size of the particles can be derived from the resulting electrical signal on the basis of a calibration curve. Differential mobility analyzers use an electric field to separate particles according to their mobility, which is a function of particle size. In the cascade impactor, the aerosol size distribution is measured by injecting air into a device containing a cascade of plates (impactors) around which the airflow is deviated. The largest particles do not follow the curvilinear air streamlines passing around the first impactor and, instead, hit the detection plate. This process, which involves increasingly narrowing nozzles, is repeated several times to extract from the beam the gradually smaller particles (see Figure 10.2).
Figure 10.2 Schematic representation of a cascade impactor device. The airflow is accelerated as the gas passes through several gradually narrowing nozzles. Smaller particles remain in the flow, while larger particle with higher inertia hit the collection plates. Increasingly smaller particles are trapped by subsequent collection plates.
Reproduced with permission from Lagzi et al. (2013).
Aerosol composition. The chemical composition of the aerosol can be determined by collecting particles on filters and subsequently analyzing the filter substrate. This analysis can be done by aqueous or organic extraction, in which the chemical species of interest are dissolved. The composition of the liquid sample is then determined by various techniques. Particles on the filter can also be heated and volatilized, and the resulting gases analyzed by gas chromatography or other methods (see Section 10.2.1).
Electron microscopy provides information on particle morphology and elemental composition. In this method, particles collected on a filter are irradiated by electrons under vacuum conditions. The X-ray energy spectrum produced
by the interactions of electrons with the particles provides information on the elemental composition of the particles. A limitation is that this method will evaporate any aerosol water and other semi-volatile species.
Mass spectrometry is increasingly used to determine the chemical composition of individual particles. This enables high-frequency measurements of aerosol composition and provides size distribution information for particles with different chemical signatures.
10.2.3 Remote Sensing
Remote sensing instruments are based on the collection of spectroscopic data along a selected atmospheric line of sight. The spectra are interpreted in terms of the species concentration integrated over the line of sight, with different levels of spatial resolution depending on the instrument and the species observed. Remote sensing can be performed from the ground, aircraft, and satellites, and can use either passive or active data collection methods.
Passive remote sensing. In passive methods, the radiation source is external to the instrument and is provided by the Sun, another star, the Moon, or the Earth and its atmosphere (infrared). A detector such as a spectrometer or a radiometer captures the electromagnetic radiation from the radiation source after it has propagated through the atmosphere along an optical path. The intensity, spectral distribution, and polarization of the measured radiation provide information on atmospheric concentrations over the optical path. Dobson spectrophotometers (Figure 10.3) were developed in the mid-1920s by G. M. B. Dobson to investigate atmospheric circulation by measuring changes in atmospheric ozone, they are now deployed globally to verify observations from satellites. The instrument derives total column and vertical profiles of ozone by measuring the direct UV radiation from the Sun, the Moon, or the zenith sky for different wavelength pairs. The total ozone column is derived from the contrast in atmospheric absorption between 305.5 nm (strong ozone absorption) and 325.4 nm (minimal ozone absorption). The vertical profile is measured using the 311.4 and 332.4 nm wavelength pair at high solar zenith angles. The measurement is based on the Umkehr effect (Götz et al., 1934), which describes the reversal (Umkehr in German) of the curve that represents the log-intensity ratio of the scattered light as a function of the solar zenith angle. This reversal, which is observed for a solar zenith angle of approximately 88 degrees when the wavelengths are 311.4 and 332.4 nm, results from the existence of an ozone maximum in the stratosphere.
Figure 10.3 Schematic representation of the Dobson ultraviolet spectrophotometer for the measurement of ozone. Entering radiation from the Sun, the Moon, or the zenith sky is reflected by a right-angle prism and falls on slit S1. The beam is then decomposed by a first spectroscope (lens L1, prism P1, and mirror M1), which reflects the radiation back to the focal plane of the instrument. Fixed slits S2, S3, and S4 isolate the different nearby wavelengths. A second spectroscope with reversed dispersion (lens L1 and mirror M2 and prism P2) recombines the light onto a photomultiplier. A chopper alternatively allows radiation at the two wavelengths to reach the detector. The ozone column is determined from the measurement at two or more pairs of wavelengths.
Reproduced from Komhyr and Evans (2008).
Instruments based on passive differential optical absorption spectroscopy (DOAS) measure the concentrations of gases along the light path by application of the Beer–Lambert law (see Section 5.2.4). When the Sun is used as the light source and the broad spectral signal associated with atmospheric scattering is removed from the observed spectrum, the remaining signal has spectral signatures representing absorption lines of atmospheric molecules. Difference between online and offline wavelengths measures the concentration of the absorber. Active DOAS systems using their own light source can measure the integrated concentrations of chemical species along the light path between the instrument and a reflector that may be located several hundreds of meters away. The multi-axis differential optical absorption spectroscopy (Max-DOAS) measurement technique (Hönninger et al., 2004) retrieves vertical profile information by combining measurements of scattered sunlight from multiple viewing directions. This retrieval requires a detailed radiative transfer model. Ground-based Max-DOAS instruments are highly sensitive to absorbers in the lowest few kilometers of the atmosphere.
Filter radiometers, often used for spacecraft observations, measure the radiative emission of the atmosphere or the transmitted solar radiation within a particular spectral band determined by a wavelength selection device (filter). The detectors are often cooled to limit the interferences from the radiation emitted by the instrument itself. Figure 10.4 shows a cross-section of the ozone mixing ratio in the upper troposphere and lower stratosphere measured by a spaceborne multi-channel limb scanning infrared radiometer. In the gas-filter correlation radiometry (GFCR) method, the incoming radiation passes through a so-called correlation cell that is filled by the target gas and acts as a spectral filter. The difference between the signal recorded from a broadband detector and the signal emerging from the correlation cell characterizes the amount of the target gas in the atmosphere.
Figure 10.4 Cross-section of the ozone mixing ratio [ppbv] measured by the 26-channel High Resolution Dynamics Limb Sounder (HIRDLS) at 20°–70°N on November 5, 2007. The data show the intrusion of ozone-rich air masses in the vicinity of the jet stream (thin full lines) and gradual dilution as the air penetrates further into the troposphere. This ozone pattern is associated with the presence of a double tropopause (black dots) at mid-latitudes.
Courtesy of W. Randel and J. Gille, NCAR.
Fourier transform infrared (FTIR) spectroscopy measures the thermal IR radiation emitted by the Earth’s surface and atmosphere with a Michelson interferometer, in which the incoming radiation is split into two beams by a half-transparent mirror. The first beam is directed to and reflected by a fixed flat mirror, while the second one is reflected by another flat mirror that is continuously moving along the axis of the incoming beam. As the two beams recombine, their phase shifts produce interference patterns. The resulting signal, called an interferogram, recorded as a function of the position of the moving mirror, represents the Fourier transform of the atmospheric spectrum. From there, the spectrum can be derived with high resolution. Almost all molecules have an IR spectrum from vibrational–rotational transitions and FTIR spectroscopy is therefore a versatile tool to detect a wide range of species (Table 10.1). The main limitation is interference from high-concentration species such as H2O and CO2. In satellite applications, limited vertical profile information can be obtained in nadir (downlooking) observations by exploiting known vertical gradients in temperature, and further vertical resolution can be obtained in limb observations at different angles.
Microwave instruments. Sensors operating in the microwave wavelength range of 0.1 to 10 cm (3–300 GHz) measure thermal emission from molecules to derive information on atmospheric parameters and chemical composition, particularly in the upper atmosphere (Kunzi et al., 2011). In this method, the observed shape of spectral lines emitted by the chemical species is fitted with the shape calculated for a specified vertical distribution of the emitter.
Active remote sensing. The most common active remote sensing technique is the light detection and ranging instrument, known as lidar (Figure 10.5), which emits a coherent light beam (often pulses at a given wavelength) to the atmosphere. A small fraction of the light is scattered by atmospheric molecules or aerosol particles back to the receiver, and the vertical distribution of the scatter can be derived by timing the return. This allows for much higher vertical resolution than passive methods. Different lidar devices account for different types of scattering processes: Rayleigh, Mie, or Raman (see Section 5.2.4). The first two processes do not change the frequency of the incident photon except by a possible Doppler shift. In the third process (Raman lidar), the wavelength of the scattered radiation is slightly shifted as a result of energy exchanges between the incident radiation and atmospheric molecules. Measurement of the spectral shift allows identification of the scattering species and determination of its ve
rtical profile.
Figure 10.5 Schematic representation of a lidar system. The light produced by a laser beam directed upward is scattered by atmospheric molecules or particles. Backscattered photons are collected by a telescope and the intensity measured by a detector.
Reproduced with permission from Lagzi et al. (2013).
If two pulses at different wavelengths are produced, e.g., by frequency multiplication of the laser output, the intensities of the two return signals are differently affected by the absorption of atmospheric species. The method, called differential absorption lidar (DIAL), allows the measurement of the vertical profile of ozone. Measurements made by an airborne DIAL instrument shown in Figure 10.6 highlight the complex distribution of ozone and aerosols in the troposphere, including small-scale features.
Figure 10.6 DIAL measurements of ozone mixing ratios [ppbv] (a) and aerosol scattering ratio at 591 nm (b) along a NASA DC-8 flight on July 8, 2008 from Cold Lake, Alberta, Canada to Thule, Greenland. The measurements were made during the ARCTAS field campaign.
Data from J. W. Hair, NASA Lidar Applications Group, NASA Langley Research Center. Source: NASA (www.science.larc.nasa.gov).
10.2.4 Measurement of Surface Fluxes
The vertical surface flux Fz of a chemical species can be determined directly by the eddy correlation method, in which the species number density n and the vertical wind velocity w are measured concurrently at the same location:
(10.3)
Here, the measurements need to be made within the surface layer (lowest ~50 m of the atmosphere) for the flux to be representative of the surface (see Box 8.4). The usual platform is a tower extending ~10 m above the surface or canopy top. A useful flux measurement must temporally average the instantaneous flux wn over a representative collection of turbulent eddies, as represented by the averaging overbar in (10.3). The averaging time is typically about one hour. w and n can be decomposed as the sums of their time-average and fluctuating components (Section 8.2):
Modeling of Atmospheric Chemistry Page 54