The Atmosphere

The atmosphere is composed of nitrogen, oxygen and argon, a variety of trace gases, and particles or aerosols from a variety of sources. Reactive, trace gases have short mean residence time in the atmosphere and large spatial and temporal variations in concentration. Many trace gases are removed by reaction with hydroxyl radical and deposition in rainfall or dryfall at the Earth's surface. The upper atmosphere, the stratosphere, contains ozone that screens ultraviolet light from the Earth's surface. Chlorofluorocarbons released by humans lead to the loss of stratospheric ozone, which might eventually render the Earth's land surface uninhabitable. Changes in the composition of the atmosphere, especially rising concentrations of CO₂, CH₄, and N₂O, will lead to climatic changes over much of the Earth's surface.

Atmospheric hydroxyl radical (OH) abundances from ground-based ultraviolet solar spectra: an improved retrieval method

The Fourier Transform Ultraviolet Spectrometer (FTUVS) instrument has recorded a long-term data record of the atmospheric column abundance of the hydroxyl radical (OH) using the technique of high resolution solar absorption spectroscopy. We report new efforts in improving the precision of the OH measurements in order to better model the diurnal, seasonal, and interannual variability of odd hydrogen (HO(x)) chemistry in the stratosphere, which, in turn, will improve our understanding of ozone chemistry and its long-term changes. Until the present, the retrieval method has used a single strong OH absorption line P(1)(1) in the near-ultraviolet at 32,341 cm(-1). We describe a new method that uses an average based on spectral fits to multiple lines weighted by line strength and fitting precision. We have also made a number of improvements in the ability to fit a model to the spectral feature, which substantially reduces the scatter in the measurements of OH abundances.

Simple Monte Carlo methods to estimate the spectra evaluation error in differential-optical-absorption spectroscopy

Differential-optical-absorption spectroscopy (DOAS) permits the sensitive measurement of concentrations of trace gases in the atmosphere. DOAS is a technique of well-defined accuracy; however, the calculation of a statistically sound measurement precision is still an unsolved problem. Usually one evaluates DOAS spectra by performing least-squares fits of reference absorption spectra to the measured atmospheric absorption spectra. Inasmuch as the absorbance from atmospheric trace gases is usually very weak, with optical densities in the range from 10(-5) to 10(-3), interference caused by the occurrence of nonreproducible spectral artifacts often determines the detection limit and the measurement precision. These spectral artifacts bias the least-squares fitting result in two respects. First, spectral artifacts to some extent are falsely interpreted as real absorption, and second, spectral artifacts add nonstatistical noise to spectral residuals, which results in a significant misestimation of the least-squares fitting error. We introduce two new approaches to investigate the evaluation errors of DOAS spectra accurately. The first method, residual inspection by cyclic displacement, estimates the effect of false interpretation of the artifact structures. The second method applies a statistical bootstrap algorithm to estimate properly the error of fitting, even in cases when the condition of random and independent scatter of the residual signal is not fulfilled. Evaluation of simulated atmospheric measurement spectra shows that a combination of the results of both methods yields a good estimate of the spectra evaluation error to within an uncertainty of ~10%.

Improvement of differential optical absorption spectroscopy with a multichannel scanning technique

Differential optical absorption spectroscopy (DOAS) of atmospheric trace gases requires the detection of optical densities below 0.1%. Photodiode arrays are used more and more as detectors for DOAS because they allow one to record larger spectral intervals simultaneously. This type of optical multichannel analyzer (OMA), however, shows sensitivity differences among the individual photodiodes (pixels), which are of the order of 1%. To correct for this a sensitivity reference spectrum is usually recorded separately from the trace-gas measurements. Because of atmospheric turbulence the illumination of the detector while an atmospheric absorption spectrum is being recorded is different from the conditions during the reference measurement. As a result the sensitivity patterns do not exactly match, and the corrected spectra still show a residual structure that is due to the sensitivity difference. This effect usually limits the detection of optical densities to approximately 3 × 10(-4). A new method for the removal of the sensitivity pattern is presented in this paper: Scanning the spectrometer by small wavelength increments after each readout of the OMA allows one to separate the OMA-fixed pattern and the wavelength-fixed structures (absorption lines). The properties of the new method and its applicability are demonstrated with simulated spectra. Finally, first atmospheric measurements with a laser long-path instrument demonstrate a detection limit of 3 × 10(-5) of a DOAS experiment.

Diode-laser-based sum-frequency generation of tunable wavelength-modulated UV light for OH radical detection

Greater than 1.5 microW of tunable wavelength-modulated 308-nm radiation was generated by sum-frequency mixing of 90 mW of 835-nm diode-laser output and 1.5 W of 488-nm Ar(+)-laser output in a beta-barium metaborate crystal. Hydroxyl radicals formed in a discharge-flow reactor were detected by use of the generated UV beam by direct absorption, wavelength-modulation absorption spectroscopy, and laser-induced fluorescence. Wavelength modulation with second-harmonic detection yielded an estimated minimum detectable absorbance of 8.5 x 10(-6) at a 1-Hz bandwidth. This absorbance level corresponds to an OH detection sensitivity of 11.7 parts per trillion for a 1-m path length at 1 atm and 298 K.

The oxidizing capacity of the earth's atmosphere: probable past and future changes

The principal oxidants in the lower atmosphere are ozone (O3) and two by-products of O3 photodissociation, the hydroxyl radical (OH) and hydrogen peroxide (H2O2). A number of critical atmospheric chemical problems depend on the earth's "oxidizing capacity," which is essentially the global burden of these oxidants. There is limited direct evidence for changes in the earth's oxidizing capacity since recent preindustrial times when, because of industrial and population growth, increasing amounts of O3 precursor trace gases (carbon monoxide, nitrogen oxides, and hydrocarbons) have been released into the atmosphere. The concentrations of O3 and possibly H2O2 have increased over large regions. Models predict that tropospheric O3 will increase approximately 0.3 to 1% per year over the next 50 years with both positive and negative trends possible for OH and H2O2. Models and the observational network for oxidants are improving, but validation of global models is still at an early stage.

An Intercomparison of Tropospheric OH Measurements at Fritz Peak Observatory, Colorado

The hydroxyl radical (OH) controls the lifetimes and therefore the concentrations of many important chemical species in Earth's lower atmosphere including several greenhouse and ozone-depleting species. Two completely different measurement techniques were used in an informal intercomparison to determine tropospheric OH concentrations at Fritz Peak concentrations by chemical analysis; the other used spectroscopic absorption on a long path. The intercomparison showed that ambient OH concentrations can now be measured with sufficient sensitivity to provide a test for photochemical models, with the derived OH concentrations agreeing well under both polluted and clean atmospheric conditions. Concentrations of OH on all days were significantly lower than model predictions, perhaps indicating the presence of an unknown scavenger. The change in OH concentration from early morning to noon on a clear day was found to be only a factor of 2.