Search

Unlike rasterization, where one can rely on pixel quad partial derivatives, an alternative approach must be taken for filtered texturing during ray tracing. We describe two methods for computing texture level of detail for ray tracing. The first approach uses ray differentials, which is a general solution that gives high-quality results. It is rather expensive in terms of computations and ray storage, however. The second method builds on ray cone tracing and uses a single trilinear lookup, a small amount of ray storage, and fewer computations than ray differentials. We explain how ray differentials can be implemented within DirectX Raytracing (DXR) and how to combine them with a G-buffer pass for primary visibility. We present a new method to compute barycentric differentials. In addition, we give previously unpublished details about ray cones and provide a thorough comparison with bilinearly filtered mip level 0, which we consider as a base method.

The U.S. Department of Labor can take certain regulatory actions to facilitate the adoption of lifetime income programs by sponsors of defined contribution retirement plans. This article outlines the simple steps the DOL can take, without the need of legislative change, that would support plan sponsors in these efforts.

We propose a powerful hardware architecture for pixel shading, which enables flexible control of shading rates and automatic shading reuse between triangles in tessellated primitives. The main goal is efficient pixel shading for moderately to finely tessellated geometry, which is not handled well by current GPUs. Our method effectively decouples the cost of pixel shading from the geometric complexity. It thereby enables a wider use of tessellation and fine geometry, even at very limited power budgets. The core idea is to shade over small local grids in parametric patch space, and reuse shading for nearby samples. We also support the decomposition of shaders into multiple parts, which are shaded at different frequencies. Shading rates can be locally and adaptively controlled, in order to direct the computations to visually important areas and to provide performance scaling with a graceful degradation of quality. Another important benefit of shading in patch space is that it allows efficient rendering of distribution effects, which further closes the gap between real-time and offline rendering.

When rendering effects such as motion blur and defocus blur, shading can become very expensive if done in a naive way, i.e. shading each visibility sample. To improve performance, previous work often decouple shading from visibility sampling using shader caching algorithms. We present a novel technique for reusing shading in a stochastic rasterizer. Shading is computed hierarchically and sparsely in an object-space texture, and by selecting an appropriate mipmap level for each triangle, we ensure that the shading rate is sufficiently high so that no noticeable blurring is introduced in the rendered image. Furthermore, with a two-pass algorithm, we separate shading from reuse and thus avoid GPU thread synchronization. Our method runs at real-time frame rates and is up to 3 × faster than previous methods. This is an important step forward for stochastic rasterization in real time.

Stochastic sampling in time and over the lens is essential to produce photo-realistic images, and it has the potential to revolutionize real-time graphics. In this paper, we take an architectural view of the problem and propose a novel hardware architecture for efficient shading in the context of stochastic rendering. We replace previous caching mechanisms by a sorting step to extract coherence, thereby ensuring that only non-occluded samples are shaded. The memory bandwidth is kept at a minimum by operating on tiles and using new buffer compression methods. Our architecture has several unique benefits not traditionally associated with deferred shading. First, shading is performed in primitive order, which enables late shading of vertex attributes and avoids the need to generate a G-buffer of pre-interpolated vertex attributes. Second, we support state changes, e.g., change of shaders and resources in the deferred shading pass, avoiding the need for a single über-shader. We perform an extensive architectural simulation to quantify the benefits of our algorithm on real workloads.

We present user-controllable and plausible defocus blur for a stochastic rasterizer. We modify circle of confusion coefficients per vertex to express more general defocus blur, and show how the method can be applied to limit the foreground blur, extend the in-focus range, simulate tilt-shift photography, and specify per-object defocus blur. Furthermore, with two simplifying assumptions, we show that existing triangle coverage tests and tile culling tests can be used with very modest modifications. Our solution is temporally stable and handles simultaneous motion blur and depth of field. © 2012 Wiley Periodicals, Inc.

The necessity to revisit water spectroscopy at 6 μm was prompted by recent work indicating that some prior measurements of H216O line strengths (ranging through seven orders of magnitude) had larger than expected systematic errors for the stronger transitions. To investigate this, linestrengths of stronger transitions were re-measured (with 14 new H2O spectra recorded with a Bruker 125 HR Fourier transform spectrometer at the Jet Propulsion Laboratory) and combined with re-analyzed prior results (obtained at higher optical densities from 32 spectra recorded with the FTS at Kitt Peak). Systematic differences for some of the older data sets were identified and corrected. In this paper, an internally-consistent sampling of 1243 selected line strengths are reported for (0 1 0)–(0 0 0) and (0 2 0)–(0 1 0) transitions between 783 and 2378 cm−1. To confirm experimental precisions, observed and calculated line strengths are compared.

This is the second of a series of articles reporting critically evaluated rotational-vibrational line positions, transition intensities, pressure dependences, and energy levels, with associated critically reviewed assignments and uncertainties, for all the main isotopologues of water. This article presents Preprint submitted to Journal of Quantitative Spectroscopy & Radiative Transfer10th June 2010 energy levels and line positions of the following singly deuterated isotopologues of water: HD16O, HD17O, and HD18O. The MARVEL (Measured Active Rotational-Vibrational Energy Levels) procedure is used to determine the levels, the lines, and their self-consistent uncertainties for the spectral regions 0–22 708, 0–1 674, and 0–12 105 cm−1 for HD16O, HD17O, and HD18O, respectively. For HD16O, 54 740 transitions were analyzed from 76 sources, the lines come from spectra recorded both at room temperature and from hot samples. These lines correspond to 36 690 distinct assignments and 8 818 energy levels. For HD17O, only 485 transitions could be analyzed from 3 sources; the lines correspond to 162 MARVEL energy levels. For HD18O, 8 729 transitions were analyzed from 11 sources and these lines correspond to 1 860 energy levels. The energy levels are checked against ones determined from accurate variational nuclear motion computations employing exact kinetic energy operators. This comparison shows that the measured transitions account for about 86 % of the anticipated absorbance of HD16O at 296 K and that the transitions predicted by the MARVEL energy levels account for essentially all the remaining absorbance. The extensive list of MARVEL lines and levels obtained are given in the Supplementary Material of this article, as well as in a distributed information system applied to water, W@DIS, where they can easily be retrieved. In addition, the transition and energy level information for H2 17O and H2 18O, given in the first paper of this series [J. Quant. Spectr. Rad. Transfer 110 (2009) 573-596], has been updated. Key words: Water vapor, transition wavenumbers, atmospheric physics, energy levels, MARVEL, information system, database, W@DIS, infrared spectra, microwave spectra, HD16O, HD17O, HD18O

Programmable shading provides artistic control over materials and geometry, but the black box nature of shaders makes some rendering optimizations difficult to apply. In many cases, it is desirable to compute bounds of shaders in order to speed up rendering. A bounding shader can be automatically derived from the original shader by a compiler using interval analysis, but creating optimized interval arithmetic code is non-trivial. A key insight in this paper is that shaders contain metadata that can be automatically extracted by the compiler using data flow analysis. We present a number of domain-specific optimizations that make the generated code faster, while computing the same bounds as before. This enables a wider use and opens up possibilities for more efficient rendering. Our results show that on average 42–44% of the shader instructions can be eliminated for a common use case: single-sided bounding shaders used in lightcuts and importance sampling.

The present analysis substantially improves the spectroscopic characterization of near infrared OCS in a window region (3850–4200 cm−1) important for atmospheric studies of Venus. Previous studies in this spectral region cataloged numerous OCS line positions, but accurate line intensities were measured for only three strong bands. In this paper, the corresponding line intensities are obtained for 41 OCS bands, including weak isotopic bands reported for the first time. The 2ν3 (0002–0000) band is analyzed for 10 OCS isotopologues (adding 16O13C34S, 17O12C32S, 16O12C36S, 18O12C34S, and 16O13C33S). In addition, observations of 0332–0330 of the main isotope, 16O12C32S, provides accurate vibration–rotation parameters for the upper state (and the lower state, 0330 of 16O12C32S). Finally, one unidentified band is seen at 3969.3 cm−1; its lower state is clearly the ground state of 16O12C32S. The line strengths of these seven previously unanalyzed bands plus 34 other bands of the OCS isotopologues, 16O12C32S, 16O12C34S, 16O13C32S, 16O12C33S, and 18O12C32S, were least-squares fitted to determine strength parameters, Sv and Herman-Wallis coefficients. Finally, the intensities of 17 additional very weak bands were estimated to provide an extensive new database of OCS line parameters to support remote sensing of Venus. The integrated intensity in cm−1/(molecule cm−2) at 296 K is 8.1×10−19 for the 3800–4200 cm−1 region.

In order to efficiently backface cull rendering primitives during computer graphics rendering, it is important to be sure that the rendering primitives to be culled are guaranteed to be backfacing even if the primitives are moving or are undergoing defocus blur. Therefore, we derive conservative tests that determine if a moving and defocused triangle is backfacing over an entire time interval and over the area of a lens. In addition, we present tests for the special cases of only motion blur and only depth of field.

To support planetary studies of the Venus atmosphere, we measured line strengths of the 2 v 3 , v 1 +2 v 2 + v 3 , and 4 v 2 + v 3 bands of the primary isotopologue of carbonyl sulfide ( 16 O 12 C 32 S), whose band centers are located at 4101.387, 3937.427, and 4141.212 cm −1 , respectively. For this, infrared absorption spectra in normal carbonyl sulfide (OCS) sample gas were recorded at an unapodized resolution of 0.0033 cm −1 at ambient room temperatures using a Bruker Fourier transform spectrometer (FTS) at the Jet Propulsion Laboratory. The FTS instrumental line shape (ILS) function was investigated, which revealed no significant instrumental line broadening or distortions. Various custom-made short cells and a multi-pass White cell were employed to achieve optical densities sufficient to observe the strong 2 v 3 and the weaker bands in the region. Gas sample impurities and the isotopic abundances were determined from mass spectrum analysis. Line strengths were retrieved spectrum by spectrum using a non-linear curve fitting algorithm adopting a standard Voigt line profile, from which Herman–Wallis factors were derived for the three bands. The band strengths of 2 v 3 , v 1 +2 v 2 + v 3 , and 4 v 2 + v 3 of 16 O 12 C 32 S (normalized at 100% of isotopologue) are observed to be 6.315(13)×10 −19 , 1.570(2)×10 −20 , and 7.949(20)×10 −21 cm −1 /molecule cm −2 , respectively, at 296 K. These results are compared with earlier measurements and the HITRAN 2004 database.

To support remote sensing of carbon dioxide in the troposphere, H2O pressure-broadened half-widths were obtained for 182 lines of CO2 in the 2250–2390 cm–1 region. For this, six spectra of CO2 were recorded at 0.003 89 cm–1 resolution using a Bruker IFS-125HR at the Jet Propulsion Laboratory. The absorption cell length was 6.14 cm, and the water pressures ranged from 20.1 to 26.5 torr (1 torr = 133.322 4 Pa) near room temperatures. Partial pressures of the species in the mixtures were determined by measuring selected line intensities in the v3 band of CO2 and the v2 band of H2O. Sample temperatures were validated by deriving rotational temperature from the v3 and v2 + v3 – v2 intensities of 12CO2 and those of CO (1–0). Positions, intensities, and half-widths were retrieved spectrum by spectrum using a nonlinear least-squares line-fitting algorithm, employing a standard Voigt line shape profile and an instrumental line shape consisting of a sinc function with aperture correction. Half-widths obtained for both the fundamental and the hot band of 12CO2 and the fundamental v3 band of 13CO2 had similar values. While half-widths of CO2 broadened by other atmospheric gases (such as N2, O2, CO2, and air) tend to decrease with increasing rotational quantum number J″, the H2O-broadened half-widths were observed to increase for intermediate J′′ (8 ≤ J′′ ≤ 42): ∼0.127 cm–1 atm–1 near J′′ = 8, and increasing to ∼0.143 cm–1 atm–1 (1 atm = 101.325 kPa) towards J″ = 42. Moreover, for 10 ≤ J″ ≤ 40, the empirical widths were within ∼2%−3% of theoretical calculations. Since water vapor could reach up to 5% of ambient atmospheric surface pressure in the tropical regions, water broadened half-widths are required to model tropospheric CO2, particularly for high-precision remote sensing, to achieve a sub-percent precision in the measurements of CO2 column averaged mixing ratio. Since little vibrational dependence in line broadening has been seen, these results at 4.3 µm can be used for other bands of CO2.

Line positions and strengths of 16O13C16O (636), 16O13C18O (638) and 16O13C17O (637), and 18O13C18O (838) bands were measured using natural and 13C- and 18O-enriched samples of CO2 at room temperature. Twenty-five near infrared (NIR) absorption spectra were recorded at 0.01–0.013 cm−1 resolution with the McMath–Pierce Fourier transform spectrometer located at the National Solar Observatory on Kitt Peak, Arizona. Absorption cells with optical path lengths ranging from 0.347 m to 385 m were used with pressures ranging between 0.5 and 147 torr. Line strengths were obtained for 17 bands of (636) between 4697 and 6797 cm−1, 13 bands of (638) between 2192 and 4954 cm−1, 4 bands of (637) between 3437 and 4981 cm−1 and 7 bands of (838) between 2182 and 4888 cm−1. Band strengths and Herman–Wallis-like F-factor coefficients were determined from least-squares fits to over 2000 measured transition intensities involving 41 different bands. The observed line positions of several bands were analyzed to obtain the upper state term values and rotational constants. Five of the 18O13C18O bands and two of the 16O13C18O bands were modeled for the first time.

A new spectroscopic database for carbon dioxide in the near infrared is presented to support remote sensing of the terrestrial planets (Mars, Venus and the Earth). The compilation contains over 28,500 transitions of 210 bands from 4300 to 7000 cm−1 and involves nine isotopologues: 16O12C16O (626), 16O13C16O (636), 16O12C18O (628), 16O12C17O (627), 16O13C18O (638), 16O13C17O (637), 18O12C18O (828), 17O12C18O (728) and 18O13C18O (838). Calculated line positions, line intensities, Lorentz half-width and pressure-induced shift coefficients for self- and air-broadening are taken from our recent measurements and are presented for the Voigt molecular line shape. The database includes line intensities for 108 bands measured using the McMath–Pierce Fourier transform spectrometer located on Kitt Peak, Arizona. The available broadening parameters (half-widths and pressure-induced shifts) of 16O12C16O are applied to all isotopologues. Broadening coefficients are computed using empirical expressions that have been fitted to the experimental data. There are limited data for the temperature dependence of widths and so no improvement has been made for those parameters. The line intensities included in the catalog vary from 4×10−30 to 1.29×10−21 cm−1/(molecule cm−2) at 296 K. The total integrated intensity for this spectral interval is 5.9559×10−20 cm−1/(molecule cm−2) at 296 K.

Previously we obtained self-broadened halfwidth and self-induced shift coefficients at room temperature for 15 near infrared CO 2 bands between 4750 and 7000 cm −1 [R.A. Toth, L.R. Brown, C.E. Miller, V.M. Devi, D.C. Benner, J. Mol. Spectrosc., 239 (2006) 243–271]. The present study expands our work on the near infrared line parameters of CO 2 to include air broadening coefficients. Here we report nearly 400 air-broadened half width and air-induced pressure shift coefficients spanning 11 different CO 2 vibrational bands in the 4750–7000 cm −1 region. Retrievals have been performed using Voigt line profiles over three distinct spectral intervals: (a) 4750–5200 cm −1 , covering the 20011 ← 00001, 20012 ← 00001, and 20013 ← 00001 Fermi Triad and three associated hot bands 21111 ← 01101, 21112 ← 01101, 21113 ← 01101; (b) 6100–7000 cm −1 , covering the 30011 ← 00001, 30012 ← 00001, 30013 ← 00001 and 30014 ← 00001 Fermi Tetrad; (c) near 6950 cm −1 for the 00031 ← 00001 overtone band. The air-broadened halfwidth and air-induced pressure shift coefficients have been modeled with empirical expressions and compared to other measurements available in the literature.

Oxygen and carbon isotope ratios in the martian CO2 are key values to study evolution of volatiles on Mars. The major problems in spectroscopic determinations of these ratios on Mars are uncertainties associated with: (1) equivalent widths of the observed absorption lines, (2) line strengths in spectroscopic databases, and (3) thermal structure of the martian atmosphere during the observation. We have made special efforts to reduce all these uncertainties. We observed Mars using the Fourier Transform Spectrometer at the Canada–France–Hawaii Telescope. While the oxygen and carbon isotope ratios on Mars were byproducts in the previous observations, our observation was specifically aimed at these isotope ratios. We covered a range of 6022 to 6308 cm−1 with the highest resolving power of ν / δ ν = 3.5 × 10 5 and a signal-to-noise ratio of 180 in the middle of the spectrum. The chosen spectral range involves 475 lines of the main isotope, 184 lines of 13CO2, 181 lines of CO18O, and 119 lines of CO17O. (Lines with strengths exceeding 10 −27 cm at 218 K are considered here.) Due to the high spectral resolution, most of the lines are not blended. Uncertainties of retrieved isotope abundances are in inverse proportion to resolving power, signal-to-noise ratio, and square root of the number of lines. Laboratory studies of the CO2 isotope spectra in the range of our observation achieved an accuracy of ∼ 1 % in the line strengths. Detailed observations of temperature profiles using MGS/TES and data on temperature variations with local time from two GCMs are used to simulate each absorption line at various heights in each part of the instrument field of view and then sum up the results. Thermal radiation of Mars' surface and atmosphere is negligible in the chosen spectral range, and this reduces errors associated with uncertainties in the thermal structure on Mars. Using a combination of all these factors, the highest accuracy has been achieved in measuring the CO2 isotope ratios: 13C/12C = 0.978 ± 0.020 and 18O/16O = 1.018 ± 0.018 times the terrestrial standards. Heavy isotopes in the atmosphere are enriched by nonthermal escape and sputtering, and depleted by fractionation with solid-phase reservoirs. The retrieved ratios show that isotope fractionation between CO2 and oxygen and carbon reservoirs in the solid phase is almost balanced by nonthermal escape and sputtering of O and C from Mars.

Line position, intensity and line shape parameters (Lorentz widths, pressure shifts, line mixing, speed dependence) are reported for transitions of the 30013 ← 00001 band of 16O12C16O (ν0 = 6227.9 cm−1). The results are determined from 26 high-resolution, high signal-to-noise ratio spectra recorded at room temperature with the McMath-Pierce Fourier transform spectrometer. To minimize the systematic errors of the retrieved parameters, we constrained the multispectrum nonlinear least squares retrieval technique to use quantum mechanical expressions for the rovibrational energies and intensities rather than retrieving the individual positions and intensities line by line. Self- and air-broadened Lorentz width and pressure-induced shift, speed dependence and line mixing (off-diagonal relaxation matrix elements) coefficients were adjusted individually. Errors were further reduced by simultaneously fitting the interfering absorptions from the weak 30012 ← 00001 band of 16O13C16O as well as the weak hot bands 31113 ← 01101, 32213 ← 02201, 40014 ← 10002 and 40013 ← 10001 of 16O12C16O in this spectral window. This study complements our previous work on line mixing and speed dependence in the 30012 ← 00001 band (ν0 = 6347.8 cm−1) [V.M. Devi, D.C. Benner, L.R. Brown, C.E. Miller, R.A. Toth, J. Mol. Spectrosc. 242 (2007) 90-117] and provides key data needed to improve atmospheric remote sensing of CO2.

Data have been collected on 55 members of the AEA Executive Committees for the years 1950–1960 (inclusive) on a variety of variables that measure the merit and non-merit characteristics of the economists. A logit is estimated in which the dependent variable is a dummy variable for whether an Executive Committee member was ever elected President of the American Economic Association (AEA). The number of publications and citations are important determinants of election. Receiving a PhD from one of the top three schools does not help and living in the South does not hurt. Economists who were older in 1956 were more likely to have eventually been elected to the AEA Presidency.

Line positions and strengths of 16O12C18O (628), 18O12C18O (828) and 17O12C18O (728) were measured between 2200 and 7000 cm−1 using 22 near infrared (NIR) absorption spectra recorded at 0.01–0.013 cm−1 resolution with the McMath–Pierce Fourier transform spectrometer located at the National Solar Observatory on Kitt Peak, Arizona. These data were obtained at room temperature using absorption cells with optical path lengths ranging from 2.4 to 385 m; the cells were filled with natural and 18O-enriched samples of CO2 at pressures ranging from 0.54 to 252 torr. The observed line positions were analyzed to obtain the upper state band centers and rotational constants for 17 bands of 16O12C18O, 19 bands of 18O12C18O and 8 bands of 17O12C18O. The majority of the 18O12C18O and 17O12C18O bands were measured for the first time. In addition, the rotational constants for the lower states 00001, 01101e and 01101f were derived for all three species using the method of combination differences in which the averaged values obtained from the line positions of two or more bands were least-squares-fitted. Rovibrational parameters were also obtained for the 02201e, 02201f, 10002 and 10001 states of 18O12C18O. The line position analysis revealed that transitions of the levels 38 ⩽ J′ ⩽ 46 of the 11111f ← 01101f band of 18O12C18O are perturbed. Perturbed transitions were also observed for the 12212 ← 02201 band and in the high-J transitions (J′ ⩾ 49) of the 20012 ← 00001 band of 18O12C18O. Band strengths and Herman–Wallis-like F-factor coefficients were determined for 21 bands of 16O12C18O, 25 bands of 18O12C18O and 8 bands of 17O12C18O from least-squares fits to more than 3700 measured transition intensities; band strengths and line positions for 34 of these bands were obtained for the first time.

Intensity and line shape parameters which predict spectral lines with absolute accuracies better than 0.3% have been determined for transitions of the 30012 ← 00001 band of 16 O 12 C 16 O centered near 6348 cm −1 from 26 high resolution, high signal-to-noise ratio spectra recorded at room temperature with the McMath–Pierce Fourier transform spectrometer. To maximize the accuracies of the retrieved parameters, the multispectrum non-linear least squares retrieval technique was modified to adjust the rovibrational constants ( G , B , D , etc.) and intensity parameters, including Herman–Wallis terms, rather than retrieving the individual positions and intensities. Speed-dependent Voigt line shapes with line mixing were required to remove systematic errors in the fit residuals. Self- and air-broadening (widths and pressure-induced shifts, speed dependence parameters) and line mixing (off-diagonal relaxation matrix elements) coefficients were thus obtained in the multispectrum fit. Remaining errors were minimized by fitting the weak 30011 ← 00001 band of 16 O 13 C 16 O as well as the weak hot bands 31112 ← 01101, 32212 ← 02201, 40012 ← 10001, and 40013 ← 10002 of 16 O 12 C 16 O that contribute interfering absorptions in this spectral window. This study presents the most extensive set of measurements to date for self- and air-broadening and self- and air-shift coefficients of a near infrared band of CO 2 . This is also the first study where line mixing parameters have been experimentally determined for any parallel CO 2 band.

Stereo rendering for 3D displays and for virtual reality headsets provide several visual cues, including convergence angle and highlight disparity. The human visual system interprets these cues to estimate surface properties of the displayed environment. Naive stereo rendering effectively doubles the computational burden of image synthesis, and thus it is desirable to reuse as many computations as possible between the stereo image pair. Computing a single radiance for a point on a surface, to be used when synthesizing both the left and right images, results in the loss of highlight disparity. Our hypothesis is that absence of highlight disparity does not impair perception of surface properties at larger distances. This is due to an ever decreasing angular difference between the surface and the two view points as distance to the surface is increased. The effect is exacerbated by the limited resolution of consumer head-mounted displays. We verify this hypothesis with a user study and provide rendering guidelines to leverage our findings.

In the previous paper, we report line strength measurements for 58 bands of 12CO2 between 4550 and 7000 cm−1 [R.A. Toth, L.R. Brown, C.E. Miller, V. Malathy Devi, D. Chris Benner, J. Mol. Spectrosc., this issue, doi:10.1016/j.jms.2006.008.001 .]. In the present study, self-broadenedwidth and self-induced pressure shift coefficients are determined in two intervals: (a) between 4750 and 5400 cm−1for bands of the Fermi triad (20011 ← 00001, 20012 ← 00001, 20013 ← 00001), three corresponding hot bands (21111 ← 01101, 21112 ← 01101, 21113 ← 01101) and the 01121← 00001 combination band; (b) between 6100 and 7000 cm−1 for the Fermi tetrad (30014 ← 00001, 30013 ← 00001, 30012 ← 00001, 30011 ← 00001), two associated hot bands (31113 ← 01101, 31112 ← 01101), as well as 00031 ← 00001 and its hot band 01131 ← 01101. Least-squares fits of the experimental width and pressure shift coefficients are modeled using empirical expressions: b 0 = exp ∑ i a ( i ) x ( i ) for widths where x ( 1 ) = 1 , x ( 2 ) = m , x ( 3 ) = m 2 , x ( 4 ) = m 3 , x ( 5 ) = m 4 , x ( 6 ) = 1 m , and d 0 = ∑ i a ( i ) x ( i ) for pressure shifts where x ( 1 ) = 1 , x ( 2 ) = 1 m , x ( 3 ) = m , x ( 4 ) = m 2 , x ( 5 ) = 1 m 2 , x ( 6 ) = 1 m 3 , x ( 7 ) = m 3 , x ( 8 ) = m m The modeled width coefficients generally agree with the experimental values with standard deviations of less than 1%, while the standard deviations of the modeled values for the pressure-induced shift coefficients range between 2.3% and 6.7%. The largest percentage error is associated with the system of the three hot bands: 21111 ← 01101, 21112 ← 01101, and 21113 ← 01101. It is observed that transitions with the same rotational quantum numbers have slightly different widths in some of the bands. As expected, pressure-induced-shift coefficients vary as a function of the band center, but there are also subtle differences from band to band for transitions with the same rotational quanta.

Line positions and strengths of 12 C 16 O 2 were measured between 4550 and 7000 cm −1 using near infrared absorption spectra recorded at 0.01–0.013 cm −1 resolution with the McMath-Pierce Fourier transform spectrometer located at the National Solar Observatory at Kitt Peak, Arizona. These were retrieved from 42 laboratory spectra obtained at room temperature with five absorption cells having various optical path lengths (from 0.1 to 409 m) filled with natural and enriched samples of CO 2 at pressures ranging from 2 to 581 Torr. In all, band strengths and Herman–Wallis-like F -factor coefficients were determined for 58 vibration–rotation bands from the least-squares fits of over 2100 unblended line strengths; strengths of 34 of these bands had not been previously reported. Band strengths in natural abundance generally ranged from 3.30 × 10 −20 to 2.8 × 10 −25 cm −1 /molecule cm −2 at 296 K. It was found that the high J transitions ( J ′ ⩾ 61) of the 20012 ← 00001 band centered at 4977.8347 cm −1 are perturbed, affecting both measured positions and strengths. Two other interacting bands, 21113e ← 01101e and 40002e ← 01101e, were also analyzed using degenerate perturbation theory. Comparisons with corresponding values from the literature indicate that absolute accuracies better than 1% and precisions of 0.5% were achieved for the strongest bands.

The temperature dependence of air-broadened halfwidths are reported for some 500 transitions in the (000)-(000) and (010)-(000) bands of H2(16)O using gas sample temperatures ranging from 241 to 388 K. These observations were obtained from infrared laboratory spectra recorded at 0.006 to 0.011 cm(exp-1) resolution with the McMath-Pierce Fourier transform spectrometer located at Kitt Peak. The experimental values of the temperature dependence exponents, eta, were grouped into eight subsets and fitted to empirical functions in a semi-global procedure. Overall, the values of eta were found to decrease with increasing rotational quantum number J. The number of measurements (over 2200) and transitions (586) involved exceeds by a large margin that of any other comparable reported study.

The space-based Orbiting Carbon Observatory (OCO) mission will achieve global measurements needed to distinguish spatial and temporal gradients in the CO2 column. Scheduled by NASA to launch in 2008, the instrument will obtain averaged dry air mole fraction ( X CO 2 ) with a precision of 1 part per million (0.3%) in order to quantify the variation of CO2 sources and sinks and to improve future climate forecasts. Retrievals of X CO 2 from ground-based measurements require even higher precisions to validate the satellite data and link them accurately and without bias to the World Meteorological Organization (WMO) standard for atmospheric CO2 observations. These retrievals will require CO2 spectroscopic parameters with unprecedented accuracy. Here we present the experimental and data analysis methods implemented in laboratory studies in order to achieve this challenging goal. To cite this article: C.E. Miller et al., C. R. Physique 6 (2005).

High-resolution spectra of H 2 O were recorded with the McMath Fourier-transform spectrometer at the National Solar Observatory located at Kitt Peak, AZ. Seventeen laboratory spectra of water plus air mixtures were recorded at a spectral resolution of 0.01 cm - 1 covering the region between 2800 and 8000 cm - 1 . Over 4000 linewidth and pressure-induced frequency shift coefficients were derived from the spectral data with sample temperatures near or at room temperature (296 K). The measurements include transitions of H2 16O, H2 17O, and H2 18O with the rare oxygen species observed in normal H 2 O gas samples. The measurements were analyzed with consideration of collision-narrowing effects. The analysis required a knowledge of the self-broadened linewidth coefficients, and the data measured and reported in the previous study were used along with earlier results for this purpose. The self-broadened and air-broadened measured linewidth coefficients obtained here as well as those reported in earlier studies covering the 600–2400 cm - 1 , were fitted to an empirical expression which contains up to 28 terms. Two fitting procedures were used: one considered the fit of families of transitions and the other was more global in scope. The computed linewidth coefficients derived from the fitted parameters are compared statistically to measurements in this work as well as other studies.

High-resolution spectra of H 2 O were recorded with a Fourier-transform spectrometer covering H 2 O transitions from 2900 to 8000 cm −1 . Over 13,000 absorptions were measured to determine line positions, strengths and self-broadened half-width coefficients. The H 2 16 O line strengths of the (0 3 0)–(0 1 0), (1 1 0)–(0 1 0), (0 1 1)–(0 1 0) and (0 3 0)–(0 0 0), (1 1 0)–(0 0 0), (0 1 1)–(0 0 0) bands were fitted to a quantum mechanical model which involves the interactions between the (0 3 0), (1 1 0), and (0 1 1) vibrational states. Also fitted were experimental strengths of the hot bands; (1 2 0)–(0 1 0) and (0 2 1)–(0 1 0). The model includes 14 dipole matrix elements for B- and A-type transitions. The measured line positions were used along with hot water emission measurements (for the (0 3 0), (0 4 0), and (0 5 0) states of H 2 16 O) in an analysis to obtain high-accuracy energy level values in the (0 3 0), (1 1 0), (0 1 1), (0 4 0), (1 2 0), (0 2 1), (2 0 0), (1 0 1), (0 0 2), and (0 5 0) vibrational states of H 2 16 O and the (1 1 0) and (0 1 1) states of H 2 17 O. Also included were measurements and analysis of self-broadened half-widths for over 4700 absorptions between 4405 and 7729 cm −1 . The results from this investigation provide new information for the noted H 2 17 O bands and present a more accurate representation of the measured H 2 16 O bands.

Line mixing has been measured for two pairs of transitions in the ν 2 band of H 2 16 O. The rotational assignments of the four lines (in terms of the rotational quantum numbers J , K a and K c ) are (a) two P branch lines: (1 0 1)←(2 1 2) mixing with (2 1 2)←(3 0 3) at 1539.061 and 1540.300 cm −1 , respectively; and (b) two R branch lines: (3 0 3)←(2 1 2) mixing with (2 1 2) ←(1 0 1) at 1652.400 and 1653.267 cm −1 , respectively. Pressure broadening coefficients of H 2 O (halfwidths, pressure shifts and off-diagonal relaxation matrix elements) are reported for seven broadeners (self, H 2 , He, CO 2 , N 2 , O 2 and air) from retrievals performed by multispectrum fitting implemented with line mixing. The effect is largest for self- and H 2 -broadening.

Measurements of line positions and strengths of D 2 18 O and HD 18 O were obtained with a Fourier transform spectrometer. The data were analyzed to obtain energy levels of the (100), (020), and (001) vibrational states of HD 18 O and the (001) and (011) states of D 2 18 O and the vibrational bands of these states connected to the ground state covered the spectral region from 2500 to 4278 cm −1 . 456 absorption lines of D 2 18 O and 856 lines of HD 18 O were assigned from the spectra. The measurements were of oxygen-18 enriched samples of deuterated water vapor and the spectra also contained features of HD 16 O, D 2 16 O, H 2 16 O, H 2 17 O and H 2 18 O of which several were used as frequency calibration standards.