Vertically constrained CO2 retrievals from TCCON measurements

Partial column-averaged carbon dioxide (CO₂) mixing ratio in three tropospheric layers has been retrieved from Total Carbon Column Observing Network (TCCON) spectra in the 1.6 μm CO₂ absorption band. Information analysis suggests that a measurement with ～60 absorption lines provides three or more pieces of independent information, depending on the signal-to-noise ratio and solar zenith angle. This has been confirmed by retrievals based on synthetic data. Realistic retrievals for both total and partial column-averaged CO₂ over Park Falls, Wisconsin on July 12, 15, and August 14, 2004, agree with aircraft measurements. Furthermore, the retrieved total column averages are always underestimated by less than 1%. The results above provide a basis for CO₂ profile retrievals using ground-based observations in the near-infrared region.     

大气CO2卫星遥感中植物叶绿素荧光影响的校正

大气CO₂卫星遥感监测的关键在于高精度,而植物叶绿素荧光存在与大气散射相似的光谱特征,干扰大气散射相关参数的反演结果,从而影响CO₂的反演精度。因植物叶绿素荧光强度微弱且影响特征与大气散射高度相关而难以准确校正。鉴于现有大气CO₂卫星遥感精度不足,以及植物叶绿素荧光对大气CO₂反演存在不可忽视的影响程度和复杂性,设计了O₂-A、1.6和2.06 μm CO₂三个光谱带协同的参数化校正方法。首先通过对大气散射采取基于光子路径长度概率密度函数（PPDF）的参数化建模,降低叶绿素荧光与大气散射参数的光谱相关性;其次,针对叶绿素荧光强度弱,难以准确反演的问题,基于轨道碳观测器（OCO-2）的叶绿素荧光产品构建了5km分辨率的全球叶绿素荧光先验信息库,增强CO₂与叶绿素荧光同步反演中对叶绿素荧光的先验约束,提高叶绿素荧光的反演精度。通过O₂-A、1.6和2.06 μm CO₂三个光谱带的协同同时反演大气散射、叶绿素荧光和大气CO₂,并结合叶绿素荧光丰富的先验信息,能够较准确剥离大气散射和叶绿素荧光而提高大气CO₂的反演精度。选择叶绿素荧光强度明显较高的柱总碳观测网络（TCCON）中的Park Falls（45.945°N,90.273°W）站点开展验证,对该站点近4年每年8月份的温室气体观测卫星（GOSAT）数据进行反演,发现植被叶绿素荧光导致GOSAT XCO₂反演结果系统偏低2×10-6(ppm)左右,通过本文的方法进行校正,反演结果的低估程度有明显改善,最大低估由2.58 ppm降低到1.49 ppm,且离散程度也有一定程度的改善,明显控制了CO₂反演中叶绿素荧光的影响,对于实现1%（~4 ppm）的CO₂反演精度来说,提供了有力支撑。     

Evaluation of errors from neglecting polarization in the forward modeling of O2A band measurements from space, with relevance to CO2 column retrieval from polarization-sensitive instruments

Sensitivity studies have been performed to evaluate the errors resulting from ignoring polarization in analyzing spectroscopic measurements of the O₂A band from space, using the Orbiting Carbon Observatory (OCO) as a test case. An 11-layer atmosphere, with both gas and aerosol loading, and bounded from below by a lambertian reflecting surface, was used for the study. The numerical computations were performed with a plane-parallel vectorized discrete ordinate radiative transfer code. Beam and viewing geometry, surface reflectance and aerosol loading were varied one at a time to evaluate and understand the individual errors. Different behavior was observed in the line cores and the continuum because of the different paths taken by the photons in the two cases. The errors were largest when the solar zenith angle was high, and the aerosol loading and surface reflectance low. To understand the effect of neglecting polarization on CO₂ column retrievals, a linear error analysis study was performed on simulated measurements from the OCO spectral regions, viz. the 1.61 and 2.06 μm CO₂ bands and the O₂A band. It was seen that neglecting polarization could introduce errors as high as 10 ppm, which is substantially larger than the required retrieval precision of ∼2 ppm. A variety of approaches, including orders of scattering, spectral binning and the use of lookup tables are being explored to reduce the errors.     

The Nimbus 7 LIMS version 6 radiance conditioning and temperature retrieval methods and results

The Nimbus 7 Limb Infrared Monitor of the Stratosphere (LIMS) experiment operated for just over 7 months in 1978/1979, as planned. Its version 5 (V5) dataset was archived for public use in 1983. Subsequently, several important improvements were realized about the LIMS instrument, its forward radiance model, and the line parameters (from HITRAN 92 or 96) for the gases that emit in its broadband spectral channels. In addition, the orbital attitude knowledge for LIMS was refined by comparing the slopes of the observed radiance profiles from its two CO₂ channels within the upper stratosphere through lower mesosphere. This approach also overcomes the usual requirement for absolute radiance calibrations in flight. The band model approximations employed for CO₂ and interfering O₃ and based on HITRAN for the 15-μm region are adequate for accurate retrievals of temperature versus pressure profiles T(p). Consequently, a reprocessed or Version 6 (V6) profile dataset was generated using these improved radiance conditioning methods (or Level 1) and the T(p) retrieval algorithm (or Level 2), and those updates are described briefly. In particular, the approach for V6 used all the radiance data ( vertical spacing), and the V6 retrievals are based on all the measured radiance profiles along an orbit. Precision and accuracy of the V6 T(p) data are improved over that for V5, most noticeably in the mesosphere. Profiles of geopotential height have also been generated as part of the V6 dataset from which one can analyze its horizontal gradients for the wind fields and higher dynamical quantities, rather than mapping the temperature fields first. The entire LIMS V6 profile dataset was archived at the NASA Goddard Distributed Active Archive Center (DAAC) in September 2002. The high quality of these V6 data demonstrates more fully the promise of the broadband, limb-infrared remote sensing technique for operational measurements of middle atmosphere T(p) from Earth-orbiting satellites.     

Aerosol influence on polarization and intensity in near-infrared O2 and CO2 absorption bands observed from space

We study the intensity and degree of linear polarization of reflected solar radiation at the top of the atmosphere within two carbon dioxide bands and one oxygen absorption band in the near-infrared. In particular, we are interested in the sensitivity of the degree of linear polarization and intensity to changes of aerosol and cirrus cloud layer heights, microphysical properties, and surface albedo. For the simulations we use spectral response functions representative of the Orbiting Carbon Observatory (OCO). Inside the O₂A band at 760 nm and strong CO₂ band at 2060 nm we find a strong influence of the aerosol and cirrus cloud layer height on the degree of linear polarization. An increase of the aerosol or cirrus cloud layer height can lead either to a decrease or increase of the polarization within the band, depending on the microphysical and optical properties of the scatterers, surface albedo, and absorption strength in the bands. The results for the O₂A band also indicate that even over land OCO enables an estimation of the height of an aerosol or cirrus cloud layer. Inside the weak CO₂ band at 1610 nm the influence of aerosol or cirrus cloud layer heights is lower as compared to the O₂A band and CO₂ band at 2060 nm, due to the relatively stronger surface influence. Here an increase of aerosol or cirrus cloud layer height leads to an increase of the degree of linear polarization even in case of low surface albedo and for weakly polarizing scatterers. For the weak CO₂ band at 1610 nm we also study the influence of the aerosol or cirrus cloud layer height on the column CO₂ estimate and the errors resulting from ignoring polarization in simulations of backscatter measurements by space-based instruments such as OCO. Depending on the surface albedo, misinterpretations of the height of atmospheric scatterers might strongly affect the column CO₂ estimates.     

Sensitivity of multi-angle photo-polarimetry to vertical layering and mixing of absorbing aerosols: Quantifying measurement uncertainties

The impact of tropospheric aerosols on climate can vary greatly based upon relatively small variations in aerosol properties, such as composition, shape and size distributions, as well as vertical layering. Polarimetric measurements have been advocated in recent years as an additional tool to better understand and retrieve the aerosol properties needed for improved predictions of aerosol radiative forcing on climate. The goal of this study is to introduce a formal approach to assessing the sensitivity of both intensity and polarization signals to absorbing aerosol layering, explicitly accounting for instrument measurement uncertainties. If ignored, sensitivity to aerosol height can introduce biases in aerosol property retrievals at short (ultraviolet or blue) wavelengths; if properly exploited, it may enable the extraction of some basic information on aerosol profiles. Employing a vector successive-orders-of-scattering (SOS) radiative transfer code, we conducted modeling experiments to determine how the measured Stokes vector elements are affected at 446 nm (blue band) by the vertical distribution, mixing and layering of smoke and dust aerosols under the assumption of a simple Lambertian surface and predefined aerosol microphysical properties. We find that smoke and dust vertical layering, if ignored, can introduce biases in radiometric and polarimetric aerosol property retrievals for aerosol optical depth (AOD) above 0.3 (polarimetric) and AOD above 0.5 (radiometric), and should, therefore, be accounted for in retrievals at high aerosol loadings.     

Tunable laser spectroscopy of CO2 near 2.05 μm: Atmospheric retrieval biases due to neglecting line-mixing

Inclusion of line-mixing into spectroscopic models is required for precise and accurate dry-mole-fraction column CO₂ retrievals from a satellite instrument. We measure first-order line-mixing parameters of thirteen ¹²CO₂ 20⁰1₃←00⁰0 CO₂ transitions near 2.05 μm using tunable laser spectroscopy. We also report line position measurements of twelve ¹³CO₂ and ¹⁸OCO transitions in this spectral region. By incorporating line-mixing into a simple atmospheric model, we show how biases in CO₂ retrievals can be generated if line-mixing is ignored.     

气溶胶对大气CO_2短波红外遥感探测影响的模拟分析

气溶胶引起的光学路径长度改变是影响高分辨率近红外光谱反演大气CO_2浓度的重要误差源.本文利用高精度大气辐射传输模式模拟中国碳卫星观测,结合CALIPSO(Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations)卫星的气溶胶廓线产品研究了不同特性的气溶胶对卫星观测光谱的影响.模拟结果显示:气溶胶散射引起的光学路径长度改变与气溶胶类型、模态以及垂直分布密切相关;城市型和海洋型气溶胶对观测光谱影响很大;多层分布的积聚模态大陆型和海洋型气溶胶在光学厚度小于0.3时,会引起5%以内的负辐射变化,随光学厚度不断增加会引起正的辐射变化;主要以粗粒子模态存在的气溶胶在不同的垂直分布情况下均会引起辐射的负变化,从而造成CO_2浓度的高估;另外,随气溶胶分布高度变高,负的辐射变化程度会逐渐减小.     

Channel selection using information content analysis: A case study of CO2 retrieval from near infrared measurements

A major challenge in retrieving CO₂ concentrations from thermal infrared remote sensing comes from the fact that measurements in the 4.3 and 15 μm absorption bands (AIRS or TES) are sensitive to both temperature and CO₂ variations. This complicates the selection of absorption channels with maximum CO₂ concentration information content. In contrast, retrievals using near infrared (NIR) CO₂ absorption bands are relatively insensitive to temperature and are most sensitive to changes of CO₂ near the surface, where the sources and sinks are located. The Orbiting Carbon Observatory (OCO) was built to measure reflected sunlight in three NIR spectral regions (the 0.76 μm O₂ A-band and two CO₂ bands at 1.61 and 2.06 μm). In an effort to significantly increase the speed of accurate CO₂ retrieval algorithms for OCO, we performed an information content analysis to identify the 20 best channels from each CO₂ spectral region to use in OCO retrievals. Retrievals using these 40 channels provide as much as 75% of the total CO₂ information content compared to retrievals using all 1016 channels in each spectral region. The CO₂ retrievals using our selected channels have a precision better than 0.1 ppm. This technique can be applied to the retrieval of other geophysical variables (e.g., temperature or CH₄), or modified for other instruments, such as AIRS or TES.     

Non-Voigt line-shape effects on retrievals of atmospheric ozone: Line-mixing effects

A theoretical approach is proposed to model line-mixing (LM) effects on absorption coefficients of O₃ perturbed by N₂ and air. It uses state-to-state rotational cross-sections calculated with a semi-classical approach and two empirical parameters, which enable switching from the state space to the line space. The first, associated with couplings within Q branches is deduced from a room temperature far-infrared spectrum. The second, governing line-couplings between R (or P) lines, is determined from a spectrum measured in the ν₁+ν₂+ν₃ band. The model developed is then successfully compared with measurements performed at room temperature for a relatively large range of pressure (0.7-8 atm) and in four different bands (from 3 to 300 μm). Accurate predictions are, in particular, obtained in the 10 μm (ν₁, ν₃) region, which is widely used for remote sensing purposes. Consequences of LM effects on retrievals of ozone atmospheric volume mixing ratios are then studied using simulated atmospheric spectra. The results show that LM leads to systematic spectra fit residuals and errors on the retrieved ozone amounts, which are small but might be detectable in measured atmospheric spectra.     

Angular distribution of the photoelectron spectrum of CO2, COS,CS2, N2O,H2O,and H2S

The photoelectron spectra of the triatomic molecules CO₂, COS, CS₂, N₂O, H₂O, and H₂S have been measured as a function of the angle θ between the direction of the incoming photon and outgoing photoelectron. The photoelectron spectra have been measured with a double-focusing electrostatic electron spectrometer to which has been attached a chamber containing a gas discharge lamp that can be freely rotated. (The photon source used was the 21.22 eV He I resonance line). From the dependence of intensity as a function of θ the angular parameter β was determined for each ionization band observed in the photoelectron spectra. A correlation was noted between the values of β and the molecular orbitals relative to the contributions of oxygen and sulfur atomic orbitals. Individual β values were also obtained for most of the vibrational bands seen in the photoelectron spectra. In most cases the vibrational structure showed little or no change in the angular parameter for a given electronic state. In certain cases, however, such as the fourth ionization band in CS₂, CO₂, and COS, rather sizeable changes in β were observed for the different vibrational bands.     

Retrieval of aerosol physical and chemical properties from mid-infrared extinction spectra

We report several results that validate the accuracy of a retrieval method for the determination of a number of aerosol particle properties from their mid infrared (600-6000 cm⁻¹) extinction spectra. These properties include the number density, chemical composition, phase, size distribution, and to some extent, shape. The approach is based on information obtained in laboratory studies of micron-sized particles using the aerosol flow tube (AFT) technique. We report here experiments in which our method is used to measure a variety of aerosols including SiO₂ micro-spheres as well as solid NaCl, (NH₄)₂SO₄, ice and liquid water particles. The uncertainties in the retrieved aerosol properties associated with the particle shapes (spheres, spheroids, cylinders, hexagonal and rectangular prisms) as well as the effect of variations in the spectral range were evaluated. To assess the accuracy of the retrieved size distributions and particle shapes, the properties calculated from infrared spectra were compared with corresponding properties determined using alternative methods. We used scanning electron microscopy (SEM) for solid (NH₄)₂SO₄ and NaCl aerosols and direct particle imaging with an optical microscope assembly for liquid water aerosols. On the basis of the validation results, we discuss the boundaries of applicability of the most popular spectral model, single scattering by spherical, homogeneous aerosol particles.     

Line shape of the far-wing beyond the band head of the CO2ν3 band

Carbon dioxide is one of the most important trace gases in the terrestrial atmosphere. The spectral data required in remote sensing are the spectral parameter of each absorption line and a line shape model. This paper describes the absorption properties of CO₂ near 2400 cm⁻¹; these properties are of interest to those in the atmospheric temperature sounding field. The shape of the far-wing of N₂- and O₂-broadened CO₂ lines was investigated in the 2200-2500 cm⁻¹ spectral region in a temperature range of atmospheric interest (230-318 K). We focused on the higher rotational quantum number of the R-branch in the ν₃ band, where the effect of the far-wing is enhanced. The effect of the far-wing has been studied extensively by others, since the CO₂ν₃ band is known to exhibit sub-Lorentzian behavior. Here, we show the observed spectra along with calculated spectra for five temperatures. We used first-order line-mixing and the χ-factor, which accounts for the effect of the far-wing, to create the calculated spectra. Our results provide new knowledge of quantum interference of the spectral line in the ν₃ band of CO₂.     

Absolute frequency measurement of CO2 laser lines with a femtosecond laser comb and new determination of the CO2 molecular constants and frequency grid

Absolute frequency measurements of a CO₂ laser stabilized on saturated absorption resonances of CO₂ laser lines are reported. They were performed using a femtosecond-laser frequency comb generator and two laser diodes at 852 and 782 nm as intermediate oscillators, with their frequency difference phase-locked to the CO₂ laser. Twenty ¹²C¹⁶O₂ laser lines in the P and R bands at 9 μm were measured with a relative uncertainty of a few 10⁻¹² limited by the CO₂ frequency reproducibility. A new determination of the CO₂ molecular constants was obtained from these data and previous measurements in the 10 μm band. The CO₂ frequency grid was also calculated, with an improvement of two orders of magnitude compared to the previous grid of Maki et al. [J. Mol. Spectrosc. 167 (1994) 211].     

Atmospheric modelling using FASCOD to identify CO2 absorption bands and their suitability analysis in variable concentrations for remote sensing applications

Recent climate studies have proven that both temperature and CO₂ content of the earth's atmosphere followed a regular 100,000-year cycle of change and that they are closely correlated. Moreover, the observed increase of CO₂ in the atmosphere exceeds the predicted values extrapolated from historical data. Other than industrialization and rapid urbanization, geo-natural hazards such as leakage from hydrocarbon reservoirs and spontaneous combustion of coal contribute a considerable amount of CO₂ to the atmosphere. Several researchers have studied the possibilities and reliabilities of atmospheric CO₂ retrieval by the point-based method (nearly accurate but much localized) and globally (wider observation but many uncertainties). Radiative transfer codes, such as FASCOD (Fast Atmospheric Signature Code) with the HITRAN (High-Resolution Transmission) spectral database can simulate atmospheric transmission and path radiance with customized gas composition (CO₂, water vapour, CO, etc.) and concentration in order to understand the phenomena in a specific wavelength region. In the present study, a number of atmospheric models were constructed with different CO₂ concentrations (ppmv) with a combination of water vapour and other atmospheric gases such as CO, CH₄, N₂O, SO₂, etc., to find out the interference patterns of these gases over CO₂ absorption bands. The transmission features of these gas combinations were analysed by partial least-squares regression models. These models show that the most suitable CO₂ absorption bands are located around 2 μm, such as 1.998 and 2.001 μm. The spectral information derived from different concentrations of CO₂ can be fitted in multivariate models to predict the CO₂ concentration from spectral information in a controlled environment. Furthermore, the present study explores the sensitivity of some available remote sensing sensors in variable CO₂ concentrations for use in real world.     

Infrared spectroscopy of solid mixed ammonia–water and acetylene–water aerosol particles

Mixed water aerosols are important components of planetary and lunar atmospheres. In this work, we use rapid-scan Fourier transform infrared (IR) spectroscopy to study solid ammonia–water and acetylene–water aerosol particles formed in a bath gas cooling cell at 78 K. With this set-up, we record time-dependent extinction spectra of particle ensembles in the mid-IR to monitor changes to the internal structure of the aerosol particles. Both ammonia–water and acetylene–water were found to form molecularly mixed structures. The mixing is observed by monitoring the profile for the ammonia ν₂ band and the acetylene ν₅ band, both of which are sensitive to particle properties. Depending on the injection conditions, the mixed particles form either immediately after sample injection or after a short mixing period of several tens of minutes. We confirm the formation of mixed particles by comparing the experimental spectra with spectra calculated with the vibrational exciton model.     

Line-widths and band strengths for the 9·4- and 10·4-μm CO2 bands

Using a stabilized CO₂ laser, the band strengths, self- and N₂-broadened half-widths, and their variation with rotational quantum number have been measured for the 9·4 and 10·4-μm CO₂ bands. In addition, transmittance calculations have been carried out using these values for comparison with low-resolution laboratory spectra. Good agreement was obtained. Evidence is also presented indicating that the far wings of the 15-μm CO₂ bands are sub-Lorentzian.     

Updated database plus software for line-mixing in CO2 infrared spectra and their test using laboratory spectra in the 1.5-2.3 μm region

In a previous series of papers, a model for the calculation of CO₂-air absorption coefficients taking line-mixing into account and the corresponding database/software package were described and widely tested. In this study, we present an update of this package, based on the 2008 version of HITRAN, the latest currently available. The spectroscopic data for the seven most-abundant isotopologues are taken from HITRAN. When the HITRAN data are not complete up to J″=70, the data files are augmented with spectroscopic parameters from the CDSD-296 database and the high-temperature CDSD-1000 if necessary. Previously missing spectroscopic parameters, the air-induced pressure shifts and CO₂ line broadening coefficients with H₂O, have been added. The quality of this new database is demonstrated by comparisons of calculated absorptions and measurements using CO₂ high-pressure laboratory spectra in the 1.5-2.3 μm region. The influence of the imperfections and inaccuracies of the spectroscopic parameters from the 2000 version of HITRAN is clearly shown as a big improvement of the residuals is observed by using the new database. The very good agreements between calculated and measured absorption coefficients confirm the necessity of the update presented here and further demonstrate the importance of line-mixing effects, especially for the high pressures investigated here. The application of the updated database/software package to atmospheric spectra should result in an increased accuracy in the retrieval of CO₂ atmospheric amounts. This opens improved perspectives for the space-borne detection of carbon dioxide sources and sinks.     

Infrared spectral radiative properties of high temperature gas with multi-component

The shock layer of high-speed missiles is a highly uneven gas medium with multi-component and high temperature. Its spectral radiation effects degrade and distort the remote sensing images of infrared sensors. The exact numerical calculating method of the shock wave thermal radiation noise has been developed in this paper. With the multi-temperature model and the rectangular grid recursive method of ray tracing, atomic-molecular absorption and emission spectra of CO₂, H₂O and N₂ were obtained along the line of sight along the seeker detectors. Based on image degradation evaluation criteria and the wind tunnel experimental results, how images blurred by the proposed model were verified. The relations between the shock wave thermal radiation noise and flow parameters were also analyzed. For the 3–8 μm infrared band, shock wave thermal radiation noise is little effected by flight altitude, and close relations with Mach number, as the empirical formula given.     

Evaluation of nonisothermal band models for H2O

Comparisons of predicted and measured radiance spectra for nonuniform hot H₂O gas samples are reported. The predicted spectra are generated with band models formulated in a companion paper, and band model parameters constructed to handle optical path temperature variations from 200 to 3000°K. The model identified in the companion paper as the intuitive derivative approximation is shown to be superior to the traditional Curtis-Godson approximation in treating general optical paths along which high degrees of nonuniformity occur.