Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Using sub-limb observations to measure gravity waves excited by convection

Using sub-limb observations to measure gravity waves excited by convection www.nature.com/npjmgrav ARTICLE OPEN Using sub-limb observations to measure gravity waves excited by convection 1✉ 2 2 3 Corwin J. Wright , Jörn Ungermann , Peter Preusse and Inna Polichtchouk Convective gravity waves are a major driver of atmospheric circulation, including the stratospheric and mesospheric quasi-biennial oscillation (QBO) and the Brewer–Dobson circulation. Previous work shows clear evidence that these waves can be excited by both single convective cells and by mesoscale convective complexes acting as a single unit. However, the partitioning of the generated waves and, crucially for atmospheric model development, the flux of momentum they transport between these two types of excitation process remains highly uncertain due to a fundamental lack of suitable observations at the global scale. Here, we use both theoretical calculations and sampled output from a high-resolution weather model to demonstrate that a satellite instrument using a sub-limb geometry would be well suited to characterising the short-vertical short-horizontal gravity waves these systems produce, and hence to provide the scientific knowledge needed to identify the relative wave-driving contribution of these two types of convective wave excitation. npj Microgravity (2023) 9:14 ; https://doi.org/10.1038/s41526-023-00259-2 INTRODUCTION atmospheric circulations with important implications for under- standing and better forecasting both weather and climate: the One of the most important drivers of atmospheric circulation is Brewer–Dobson circulation (BDC), the mesospheric residual the integrated effect of small-scale atmospheric gravity waves circulation (MRC), and the quasi-biennial oscillation (QBO). The (GWs). These waves are commonly generated at near-surface BDC is a key middle-atmospheric circulation which transports altitudes and then propagate upwards, transporting the momen- chemicals and trace gases from equator to pole in the strato- tum and energy needed to drive and control dynamical and sphere and the MRC is a similar circulation from pole to pole in the chemical processes in the upper troposphere, stratosphere and mesosphere, while the QBO is a pattern of alternating eastward above. They play a diverse range of roles, including acting as a 1,2 and westward winds in the tropics which repeats on an irregular major control on the speed of the jet streams , causing clear air cycle averaging 28 months in duration and which is believed to be turbulence which affects aviation , and, by coupling into the one of the sources of skill for seasonal-timescale weather forecasts charged ionosphere above ~90 km altitude, disrupting GPS and providing memory that short-term weather does not have . radio signals . Each of these circulations has important impacts on surface GWs are generated by many processes, including wind flowing weather and climate, and all three are well known to be driven by over mountains (‘orographic generation’) and geostrophic adjust- waves. The exact partitioning of this wave driving between large- ment in the atmosphere. One of the most important and hard to scale waves which are well resolved by current-generation models study, however, is GW generation by convective weather systems. and smaller-scale GWs is however poorly understood, and in These systems are by far the largest source of GWs at tropical particular, the way this division will evolve in the future is highly latitudes, and also one of the major sources at midlatitudes in all uncertain. In particular, there are indications that the QBO is seasons except winter (when orographic generation is dominant affected by climate change. While the QBO was stable from when at middle and high latitudes). it was first observed in the 1950s until 2016, since then two major Convection excites GWs via different processes and at different 11,12 unexpected disruptions have occurred . To understand and wavelengths. Past work using case studies has in general focused 5,6 predict the future of the QBO, a better understanding of its on wave excitation by single convective cells and upscaled their fundamental driving forces is essential. For the BDC and MRC findings from convective source modelling to global distribu- tions . According to these simulations, the major part of the meanwhile, while the winter BDC is primarily driven by planetary- 13 14 momentum flux transported by convective GWs is expected to be scale waves, the summer BDC and the MRC are driven mainly found at vertical wavelengths λ ≤ 10 km and λ ≪ 100 km. by GWs. As such, uncertainties in the BDC due to the interaction of z h different processes and in particular due to missing GW drag However, convection often becomes organised and forms in the tropics and subtropics mesoscale convective clusters, which act as prevent models from consistently predicting how it will evolve in 15–18 wave sources as a whole and excite GWs of much longer coming decades . horizontal wavelengths . Full-complexity very high-resolution The portion of the GW spectrum visible to a space observation global ECMWF IFS simulations point to the coexistence of both technique is governed by the observation geometry, the optical processes. This, however, is ill-constrained by observations. depth of the considered emission and the field of view of the Characterising this partitioning is important for understanding instrument. The different observational geometries suitable for the dynamic structure of the middle atmosphere. This is because this task, as described in ref. , are typically grouped into limb, GWs at low latitudes are vital to the driving of three major nadir and sub-limb sounding. Limb observations, which cover a 1 2 3 Centre for Space, Atmospheric and Oceanic Science, University of Bath, Bath, UK. Forschungszentrum Jülich, Jülich, Germany. European Centre for Medium-Range Weather Forecasts, Reading, UK. email: C.Wright@bath.ac.uk Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA 1234567890():,; C.J. Wright et al. important exception to this was the astronomical Midcourse Space Experiment (MSX), which observed 4.3 μmCO emissions and occasionally used a sub-limb mode. Out of 80 such scenes (of 20 minutes’ duration) recorded by MSX, a number contained GW signals and one was analysed in detail showing signatures of GWs of ~50 km horizontal wavelength arising from a thunder- storm. At the time of the MSX observations, such linear 1D arrays sensitive to the near-infrared were very advanced technology, but since then both detector and cryo-cooler technology have greatly advanced. This allows for the compact and relatively low-priced realisation of even more powerful approaches which would allow the measurement of both a high number of pixels across track as well as a multitude of angles at the same time. In this paper, we propose an observation concept to study this missing part of the GW spectrum using spaceborne measure- ments of mid-infrared emissions, based on the saturated emission approach in ref. . To assess the potential performance of an instrument of this type, we carry out idealised simulations and then sample a high-resolution weather forecasting model with the expected observational characteristics of such an instrument. Finally, we describe the technological underpinnings needed for such a sounder. RESULTS Observational filter We first estimate how well the envisioned instrument would capture plain sinusoidal GWs depending on their horizontal and vertical scales. To do so, we initially generate a simplified atmosphere with a climatological midlatitude temperature and CO profile, working on a horizontal grid of 1 × 1 km spacing with a vertical step of 100 m. This temperature is then perturbed with a sinusoidal monochromatic wave of temperature amplitude 10 K, and the resulting temperature perturbations are retrieved using the JURASSIC2 forward model (see “Methods”). The simulations are then repeated with systematically varying vertical and Fig. 1 Sketch of viewing geometry. From a position in orbit, the horizontal wavelengths. Using this perturbed atmosphere, we instrument views downward by an angle α below the local compute the brightness temperature the satellite would measure horizontal. Infrared thermal radiation is emitted from the saturation at varying sub-limb angles, extracting one brightness temperature layer (indicated by the bright-red ring segment above the Earth). estimate per kilometre in the horizontal. The simulated brightness temperature measurements are then wide range from short to long vertical wavelengths, have analysed with a simple least squares fit for the largest sinusoidal facilitated momentum budget studies of GWs from MCCs with 20–22 frequency, using the results of a Fast Fourier Transform as an initial wavelengths λ ≥ 100 km . Nadir observations, meanwhile, are guess. Finally, the amplitude of this GW signal is computed, and sensitive only to long vertical wavelengths (λ > 15 km). This leaves the ratio between the true amplitude of 10 K and the derived the postulated major portion of parameterised CGW momentum flux, at scales both short in the horizontal and in the vertical, value is shown in Fig. 2 across a range of wavelengths at eight uncovered. This gap in our observational coverage is a major selected sub-limb viewing angles. We find no significant bias problem since it leaves us uncertain as to how the momentum between the derived wavelength and true horizontal wavelength flux these GWs transport is partitioned between single-tower and for all simulations with sufficient sensitivity to extract such MCC sources. estimates. At a physical level, limb sounding requires optically thin Convective GWs have typically vertical wavelengths of the order emissions and the observable scales are limited by the of 6–15 km , corresponding to a phase speed range of 20–50 m/s. integration of radiance along the line of sight (LOS). Nadir From Fig. 2, we see that we are sensitive to these wavelengths at sounding, meanwhile, is based on radiances saturating around a all σ between 25 and 60°, but that this sensitivity vanishes for 80° target altitude, with the vertical resolution limited by the width and nadir. We can assess this by considering a fixed vertical wavelength, e.g., 10 km—at such a specified value, the region of of this saturation. In the sub-limb, however, while emissions also maximal sensitivity sweeps the horizontal wavelength range, saturate along the LOS, the sensitivity to GW wavelength is controlled by the alignment of the 3D GW phase fronts with the maximising at ~λ = 50 km for 25° and λ = 7 km for 60°. Thus, h h LOS. This geometry is illustrated in Fig. 1: here, the sub-limb covering the wave field with a wide range of σ, one can estimate viewing angle α and the flight altitude determine the angle σ the aspect ratio of the GWs, and from the horizontal structure which the LOS forms with the local horizontal. σ = 0 corresponds determine the horizontal wavelength. In the ideal case of to the highest tangent height and marks the limit between sub- monochromatic waves, this would also allow us to infer the limb and limb; for a flight altitude of 400 km, it is reached for α vertical wavelengths by combining the horizontal wavelength and close to 20°. the sensitivity dependence. In the past, this sub-limb geometry was mainly used by We now proceed to consider two more realistic GW cases, both microwave instruments (ref. , and references therein) where the derived from a very high-resolution global numerical weather comparably large vertical field of view limited sensitivity. An prediction model simulation. npj Microgravity (2023) 14 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA 1234567890():,; C.J. Wright et al. Fig. 2 Estimated effect of the observational filter for selected sub-limb viewing angles from 25 to 90°. a–h Colours and line contours show the fraction of the input signal amplitude recovered for GWs of this horizontal and vertical wavelength. The thick solid line shows the expected optimum visibility where the LOS angle σ aligns with the orientation of the phase fronts given by the ratio of vertical and horizontal wavelength. Model by an instrument at the height of the ISS travelling northward along each meridian. We repeat this for a range of sub-limb We use the output from a TCo7999L137 (~1.4 km horizontal, 137- observing angles, across the range 25–90°. level vertical) run of the hydrostatic ECMWF Integrated Forecast- ing System (IFS) , using a 60-s model timestep. This simulation has the highest spatial resolution ever used in a seasonal- Gravity wave analysis timescale global atmospheric model and represents an excep- The above step provides global estimates of brightness tempera- tional scientific resource for studies of this nature. In particular, it tures for a range of sub-limb observing angles and a backward- has a substantially higher spatial resolution than the global-scale looking observer flying along the meridians. We now proceed to observing system simulation experiment ‘nature runs’ often used analyse these data for GW signatures, to assess the effect of in satellite development studies . Because the IFS simulation varying the sub-limb angle on how well the GWs are recovered at used only resolves hydrostatic waves, our analysis only considers each angle. The simulated observations are analysed using the these waves. 28 two-dimensional Stockwell Transform in ref. , as modified for Compared to lower resolution versions of the IFS, this model 29,30 computational efficiency in refs. . Due to the very large volume resolves a much larger part of the GW spectrum, with equal of the data when considered at global scales, the computational contribution of small-scale GWs with λ < 100 km versus larger complexity of applying techniques of this nature to such large scale GWs with λ > 100 km to the convectively generated GW volumes of data, and the brief format of this study we focus here momentum flux . The model runs without GW or convective on two cases: a large orographic wave over the Ural mountains parameterisations, and thus all GWs in the model are explicitly (Fig. 3a) and a group of convective waves southeast of Japan generated and propagate freely on the model grid. The simulation (Fig. 4b). Other sampled cases provide results broadly consistent was initialised from the ECMWF operational analysis on the 1st of with these two examples. November 2018 and then ran freely until the sampled dates, We first interpolate the brightness temperature data onto a detailed below for each case individually. regular 1 × 1 km spatial grid centred around the GW feature of interest. Interpolation to a regular spatial grid is required as the Brightness temperature simulations S-Transform analysis is based on Fast Fourier Transform algo- We compute the simulated brightness temperature perturbations rithms. For our Urals case, this grid is centred at 55°N, 115°E and from this dataset in a linearised fashion to save computation time. extends ±25° in longitude and ±14° in latitude at mean longitude and latitude respectively, while for the Japan case, it is centred at To that end, we must first remove the temperature background 33.5°N, 146°E and extends ±12° in both longitude and latitude. We from the model temperature structures . We then configure and carry out this interpolation separately for data extracted at all run new 2D simulations using JURASSIC2. Unlike the regular spatial grid used for the observational filter calculation, we carry viewing angles from 25 to 60°, and the results from each viewing out these simulations using a coarser horizontal grid of spacing angle assessed are shown individually. 0.025° in latitude and a vertical grid of 0.5 km, approximating the We then apply the 2D ST to these data. This analysis is carried out on the regular spatial grid, which at the centre location has grid spacing used in the spectral IFS run. Cartesian x and y axes corresponding to the zonal and meridional We then compute the 2D Jacobian matrix (also called a directions. In both these directions, we use a scaling parameter c sensitivity kernel) of a single brightness temperature measure- 28,31 ment for a climatological atmosphere, to effectively linearise the of 0.25 and limit the maximum horizontal wavelength to complex forward model. By convolving this sensitivity matrix with 500 km, and we permit the analysis to find the strongest 1000 2D IFS-derived temperature perturbations individually along each frequencies present in the data. These choices are consistent with 0.025°-spaced meridian, we directly compute the (linearised) previous studies using the ST. Figs. 3b–f and 4b–f show maps of brightness temperature perturbations which would be observed the estimated wave amplitudes calculated using this approach. Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA npj Microgravity (2023) 14 C.J. Wright et al. Fig. 3 Temperature amplitudes recovered for a sample orographic wave over the Urals. a Temperature output at 37 km altitude of the TCo7999 simulations. b–g Estimated amplitude of this wave as measured using a 2D Stockwell Transform method, for brightness temperatures computed from model output sampled across a range of viewing angles. Black boxes show region averaged over to generate Fig. 5. Note that amplitude values in panel (b) have been downscaled by a factor of 3 to aid visual comparison (see text for details). Case study results cases, the measured amplitudes have been normalised to the values calculated for the 25° sub-limb viewing angle (themselves We consider first the GW case over the Urals, Fig. 3. The phase ~50% reduced from the original model fields), in order to focus on fronts of this wave are aligned north-south, consistent with the the relative amplitude reduction as a function of angle. This north-south topography of the underlying mountain range, and highlights the rapid decline in signal strength for the convective this combined with the morphology and large amplitude of the case in comparison to the orographic case. While the amplitudes wave gives significant confidence that this wave is orographic in measured for the orographic case drop never below 50% of the origin. We see peak amplitudes >1 K across all viewing angles, best-case geometry considered, the convective case rapidly falling from a peak amplitude of >3 K (note saturated colour scale) decreases, stabilising at around 30% of the original amplitude at the lowest viewing angle to ~1.25 K at 60°. As we vary the for all angles above ~50°. viewing angle of the instrument, the geographic location of the wave does not shift, and the morphology of the inferred amplitude field remains largely constant. The results seen for DISCUSSION the sub-limb viewing cases, particularly at low viewing angles, are Convective GWs have expected vertical wavelengths of 6–15 km morphologically consistent with those obtained from an equiva- and cover a huge range of horizontal scales. While nadir sounding lent S-Transform analysis applied to the raw model data, but lower instruments are almost insensitive to this wavelength range, limb in magnitude. The largest sensitivity at large viewing angles is sounders can resolve convective GWs excited by MCCs. Here the seen at 60°E, 54°N, where 3D analysis (not shown) indicates challenge is to accurately quantify momentum flux by bringing particularly low horizontal and long vertical wavelengths. 32,33 an IR limb imager into space . Convective GWs from single This case contrasts in several important ways with our Japanese towers, however, are not covered with the existing observation case, Fig. 4. In this case, the observed phase fronts are radial in techniques. morphology, and are geolocated over similar features in the Our results provide a clear example of the benefits of a sub-limb troposphere (not shown). They are therefore consistent with viewing geometry over a nadir or near-nadir geometry for convective GW generation. While again the waves remain at the observing the GWs produced by convective activity. In both the same location as the observation angle is increased, here the reduction in amplitude with increasing angle is much larger, theoretical and realistic cases considered, the amplitudes of GWs falling from a peak amplitude ~1.25 K at 25° to <0.2 K at 60°. The with the spectral properties typical of convective wave activity are suppressed significantly at high sub-limb angles (i.e., near-nadir exception, in this case, is a spot around 32°N, 140°E. There 3D geometries) and much better recovered (although still reduced) at analysis reveals northward propagating waves (phase fronts tilted to the instrument), with <50 km horizontal wavelengths and low sub-limb angles (i.e., sub-limb geometries). Due to advances in detector technology since the work different vertical wavelengths up to 20 km, i.e., the steepest waves are visible for the largest angles. described in ref. , highly sensitive infrared detector arrays for Figure 5 shows the different amplitude reductions with sub- wavelengths 4 μm with resolutions of order 1000 × 1000 pixels limb viewing angle in these two cases as comparative line plots, and frame rates ~100 frames/s are now available commercially. averaged over the black-outlined regions in Figs. 3 and 4. In both Since GW observations of the type needed for atmospheric npj Microgravity (2023) 14 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA C.J. Wright et al. Fig. 4 Temperature amplitudes recovered for a group of GWs associated with convection southeast of Japan. a Temperature output at 37 km altitude of the TCo7999 simulations. b–g Estimated amplitude of this wave as measured using a 2D Stockwell Transform method, for brightness temperatures computed from model output sampled across a range of viewing angles. Black boxes show region averaged over to generate Fig. 5. Note that amplitude values in panel (b) have been downscaled by a factor of 2 to aid visual comparison (see text for details). While the heavily reduced amplitude seen at the nadir is not a fundamental problem for wave detection in the idealised and simulatedcases considered here,itisamajorproblem forreal- world observations. Our idealised observational filter calcula- tions are inherently noise-free, and our model-based analyses lack both retrieval and instrumental noise. In the real world, however, the presence of a noise floor means that GWs with suppressed amplitude fall below this floor. As such, these waves become functionally undetectable even if, theoretically, a nadir observing geometry should allow some signal to be recovered from them. This has hampered studying of this important source of atmospheric GWs, leaving the partitioning of GW momentum flux between GWs from MCCs and single convective cells one of Fig. 5 Reduction in recovered amplitudes relative to 25° case. the large controversies in atmospheric physics. As such, a sub- Figure shows average GW amplitudes over the regions defined by limb geometry like the concept proposed here has significant black boxes in Figs. 3 (orange line) and 4 (blue line), as computed design benefits for the important scientificproblem of from brightness temperature estimates over a range of viewing angles. characterising convective atmospheric gravity waves from single convective cells. science simultaneously require a wide angle range and a small field-of-view (FOV) for each pixel, a low orbit such as that provided by the International Space Station (ISS) is highly METHODS favourable. From this orbit, the distance from the observing To simulate measurements of the envisioned instrument, we platform to the saturation altitude of the measurement is employ the Jülich Rapid Spectral Simulation Code Version 2 between 450 and 1100 km depending on the viewing angle, (JURASSIC2) forward model . This forward model uses the and an FOV width perpendicular to the instrument line of sight emissivity growth approximation to compute emissions over (LOS) 0.3–0.6 km (neglecting diffraction) would cover a swath of −1 the spectral band from 2320 to 2340 cm . JURASSIC2 is capable width 300–600 km for downward viewing angles of 25–60°, of 2D and 3D simulations and inversions, and here we employ a respectively (Fig. 1, where σ is defined as the ‘sub-limb angle’ simplified 2D forward model along the meridians, in order to hereafter). Despite the high orbital velocity of the ISS platform, reduce the computational complexity of the problem. As we wish the image recording rates needed to prevent smearing from to focus primarily on the effects of different viewing angles in the measurements with limited integration times are also easily vertical, to further simplify the analysis we do not account for FOV achievable with current-day technology. We note however that sensitivity estimates depend on the speed of the optics and are effects, i.e., each brightness temperature is computed using an thus beyond the scope of this brief study. infinitesimally thin ‘pencil’ beam. Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA npj Microgravity (2023) 14 C.J. Wright et al. Reporting summary 22. Ern, M. et al. The semiannual oscillation (SAO) in the tropical middle atmosphere and its gravity wave driving in reanalyses and satellite observations. Atmos. Further information on research design is available in the Nature Chem. Phys. 21, 13763–13795 (2021). Research Reporting Summary linked to this article. 23. Wu, D. L. & Eckermann, S. D. Global gravity wave variances from Aura MLS: characteristics and interpretation. J. Atmos. Sci. 65, 3695–3718 (2008). 24. Dewan, E. M. et al. MSX satellite observations of thunderstorm-generated gravity DATA AVAILABILITY waves in mid-wave infrared images of the upper stratosphere. Geophys. Res. Lett. The observational filter simulations presented in Fig. 2 have been generated 25, 939–942 (1998). computationally as described in the text and associated references and do not 25. Wedi, N. P. et al. A baseline for global weather and climate simulations at 1 km incorporate any underlying data. As the underlying simulation output used in resolution. J. Adv. Model. Earth Syst. 12, https://doi.org/10.1029/2020MS002192 Figs. 3–5 is extremely large (42GB per 1-h timestep) we cannot easily archive these (2020). data separately for this study, but the retrieved brightness temperatures for the 26. Holdaway, D. & Yang, Y. Study of the effect of temporal sampling frequency on regions and times used in our study have been archived at Zenodo (https://doi.org/ DSCOVR observations using the GEOS-5 Nature Run results (Part I): Earth’s 10.5281/zenodo.6637647) as Matlab-format data files. radiation budget. Remote Sens. 8, 98 (2016). 27. Strube, C., Ern, M., Preusse, P. & Riese, M. Removing spurious inertial instability signals from gravity wave temperature perturbations using spectral filtering Received: 30 May 2022; Accepted: 20 January 2023; methods. Atmos. Meas. Tech. 13, 4927–4945 (2020). 28. Hindley, N. P., Smith, N. D., Wright, C. J., Rees, D. A. S. & Mitchell, N. J. A two- dimensional Stockwell transform for gravity wave analysis of AIRS measurements. Atmos. Meas. Tech. 9, 2545–2565 (2016). 29. Wright, C. J., Hindley, N. P., Hoffmann, L., Alexander, M. J. & Mitchell, N. J. Exploring gravity wave characteristics in 3-D using a novel S-transform technique: REFERENCES AIRS/Aqua measurements over the Southern Andes and Drake Passage. Atmos. 1. Fritts, D. C. & Alexander, M. J. Gravity wave dynamics and effects in the middle Chem. Phys. 17, 8553–8575 (2017). atmosphere. Rev. Geophys. 41,1–68 (2003). 30. Hindley, N. P. et al. Gravity waves in the winter stratosphere over the Southern 2. Palmer, T. N., Shutts, G. J. & Swinbank, R. Alleviation of a systematic westerly bias Ocean: high-resolution satellite observations and 3-D spectral analysis. Atmos. in general circulation and numerical weather prediction models through an Chem. Phys. 19, 15377–15414 (2019). orographic gravity wave drag parametrization. Q. J. R. Meteorol. Soc. 112, 31. Stockwell, R., Mansinha, L. & Lowe, R. Localization of the complex spectrum: the S 1001–1039 (1986). transform. IEEE Trans. Signal Process. 44, 998–1001 (1996). 3. Williams, P. D. & Joshi, M. M. Intensification of winter transatlantic aviation tur- 32. Preusse, P. et al. New perspectives on gravity wave remote sensing by space- bulence in response to climate change. Nat. Clim. Change 3, 644–648 (2013). borne infrared limb imaging. Atmos. Meas. Tech. 2, 299–311 (2009). 4. National Research Council. Solar and Space Physics: A Science for a Technological 33. Preusse, P. et al. Characteristics of gravity waves resolved by ECMWF. Atmos. Society (The National Academies Press, 2013). Chem. Phys. 14, 10483–10508 (2014). 5. Piani, C., Durran, D., Alexander, M. J. & Holton, J. R. A numerical study of three- 34. Ungermann, J., Hoffmann, L., Preusse, P., Kaufmann, M. & Riese, M. Tomographic dimensional gravity waves triggered by deep tropical convection and their role in retrieval approach for mesoscale gravity wave observations by the PREMIER the dynamics of the QBO. J. Atmos. Sci. 57, 3689–3702 (2000). infrared limb-sounder. Atmos. Meas. Tech. 3, 339–354 (2010). 6. Choi, H.-J. & Chun, H.-Y. Momentum flux spectrum of convective gravity waves. 35. Weinreb, M. P. & Neuendorffer, A. C. Method to apply homogeneous-path Part I: an update of a parameterization using mesoscale simulations. J. Atmos. Sci. transmittance models to inhomogeneous atmospheres. J. Atmos. Sci. 30, 662–666 68, 739–759 (2011). (1973). 7. Choi, H.-J., Chun, H.-Y., Gong, J. & Wu, D. L. Comparison of gravity wave tem- perature variances from ray-based spectral parameterization of convective gravity wave drag with AIRS observations. J. Geophys. Res. Atmos. 117, 5115 (2012). 8. Trinh, Q. T. et al. Tuning of a convective gravity wave source scheme based on ACKNOWLEDGEMENTS HIRDLS observations. Atmos. Chem. Phys. 16, 7335–7356 (2016). This research used the resources of the Oak Ridge Leadership Computing Facility 9. Polichtchouk, I., Wedi, N. & Kim, Y.-H. Resolved gravity waves in the tropical (OLCF), which is a DOE Office of Science User Facility supported under Contract DE- stratosphere: impact of horizontal resolution and deep convection parametriza- AC05-00OR22725. C.J.W. is funded by Royal Society University Research Fellowship tion. Q. J. R. Meteorol. Soc. 148, 233–251 (2022). UF160545 and by NERC grants NE/S00985X/1, NE/V01837X/1 and NE/W003201/1. 10. Scaife, A. A. et al. Predictability of the quasi-biennial oscillation and its northern winter teleconnection on seasonal to decadal timescales. Geophys. Res. Lett. 41, 1752–1758 (2014). 11. Osprey, S. M. et al. An unexpected disruption of the atmospheric quasi-biennial AUTHOR CONTRIBUTIONS oscillation. Science 353, 1424–1427 (2016). C.J.W.: conceptualisation, methodology, software, formal analysis, investigation, 12. Anstey, J. A. et al. Prospect of increased disruption to the QBO in a changing writing and visualisation. J.U.: methodology, software, formal analysis, investigation, climate. Geophys. Res. Lett. 48, e2021GL093058 (2021). 13. Alexander, M. J. Gravity-wave forcing in the stratosphere: observational con- writing and data curation. P.P.: conceptualisation, methodology, investigation and straints from the Upper Atmosphere Research Satellite and implications for writing. I.P.: software, writing and data curation. parameterization in global models. J. Geophys. Res. Atmos. 108, https://doi.org/ 10.1029/2003jd003373 (2003). 14. Murgatroyd, R. J. & Singleton, F. Possible meridional circulations in the strato- COMPETING INTERESTS sphere and mesosphere. Q. J. R. Meteorol. Soc. 87, 125–135 (1961). 15. Butchart, N. The Brewer-Dobson circulation. Rev. Geophys. 52, 157–184 (2014). The authors declare no competing interests. 16. Polvani, L. M., Abalos, M., Garcia, R., Kinnison, D. & Randel, W. J. Significant weakening of Brewer-Dobson circulation trends over the 21st century as a consequence of the Montreal protocol. Geophys. Res. Lett. 45, 401–409 (2018). ADDITIONAL INFORMATION 17. Abalos, M. et al. New insights on the impact of ozone-depleting substances on the Brewer-Dobson circulation. J. Geophys. Res.: Atmos. 124, 2435–2451 (2019). Supplementary information The online version contains supplementary material 18. Richter, J. H. et al. Response of the quasi-biennial oscillation to a warming climate available at https://doi.org/10.1038/s41526-023-00259-2. in global climate models. Q. J. R. Meteorol. Soc. 148, 1490–1518 (2020). Correspondence and requests for materials should be addressed to Corwin J. Wright. 19. Wu, D. L. et al. Remote sounding of atmospheric gravity waves with satellite limb and nadir techniques. Adv. Space Res. 37, 2269–2277 (2006). Reprints and permission information is available at http://www.nature.com/ 20. Ern, M., Preusse, P., Kalisch, S., Kaufmann, M. & Riese, M. Role of gravity waves in reprints the forcing of quasi two-day waves in the mesosphere: an observational study. J. Geophys. Res.: Atmos. 118, 3467–3485 (2013). 21. Ern, M. et al. Interaction of gravity waves with the QBO: a satellite perspective. J. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims Geophys. Res.: Atmos. 119, 2329–2355 (2014). in published maps and institutional affiliations. npj Microgravity (2023) 14 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA C.J. Wright et al. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http:// creativecommons.org/licenses/by/4.0/. © The Author(s) 2023 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA npj Microgravity (2023) 14 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png npj Microgravity Springer Journals

Using sub-limb observations to measure gravity waves excited by convection

Download PDF
7 pages

Loading next page...
 
/lp/springer-journals/using-sub-limb-observations-to-measure-gravity-waves-excited-by-CNcUl9xi1O

References (37)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2023
eISSN
2373-8065
DOI
10.1038/s41526-023-00259-2
Publisher site
See Article on Publisher Site

Abstract

www.nature.com/npjmgrav ARTICLE OPEN Using sub-limb observations to measure gravity waves excited by convection 1✉ 2 2 3 Corwin J. Wright , Jörn Ungermann , Peter Preusse and Inna Polichtchouk Convective gravity waves are a major driver of atmospheric circulation, including the stratospheric and mesospheric quasi-biennial oscillation (QBO) and the Brewer–Dobson circulation. Previous work shows clear evidence that these waves can be excited by both single convective cells and by mesoscale convective complexes acting as a single unit. However, the partitioning of the generated waves and, crucially for atmospheric model development, the flux of momentum they transport between these two types of excitation process remains highly uncertain due to a fundamental lack of suitable observations at the global scale. Here, we use both theoretical calculations and sampled output from a high-resolution weather model to demonstrate that a satellite instrument using a sub-limb geometry would be well suited to characterising the short-vertical short-horizontal gravity waves these systems produce, and hence to provide the scientific knowledge needed to identify the relative wave-driving contribution of these two types of convective wave excitation. npj Microgravity (2023) 9:14 ; https://doi.org/10.1038/s41526-023-00259-2 INTRODUCTION atmospheric circulations with important implications for under- standing and better forecasting both weather and climate: the One of the most important drivers of atmospheric circulation is Brewer–Dobson circulation (BDC), the mesospheric residual the integrated effect of small-scale atmospheric gravity waves circulation (MRC), and the quasi-biennial oscillation (QBO). The (GWs). These waves are commonly generated at near-surface BDC is a key middle-atmospheric circulation which transports altitudes and then propagate upwards, transporting the momen- chemicals and trace gases from equator to pole in the strato- tum and energy needed to drive and control dynamical and sphere and the MRC is a similar circulation from pole to pole in the chemical processes in the upper troposphere, stratosphere and mesosphere, while the QBO is a pattern of alternating eastward above. They play a diverse range of roles, including acting as a 1,2 and westward winds in the tropics which repeats on an irregular major control on the speed of the jet streams , causing clear air cycle averaging 28 months in duration and which is believed to be turbulence which affects aviation , and, by coupling into the one of the sources of skill for seasonal-timescale weather forecasts charged ionosphere above ~90 km altitude, disrupting GPS and providing memory that short-term weather does not have . radio signals . Each of these circulations has important impacts on surface GWs are generated by many processes, including wind flowing weather and climate, and all three are well known to be driven by over mountains (‘orographic generation’) and geostrophic adjust- waves. The exact partitioning of this wave driving between large- ment in the atmosphere. One of the most important and hard to scale waves which are well resolved by current-generation models study, however, is GW generation by convective weather systems. and smaller-scale GWs is however poorly understood, and in These systems are by far the largest source of GWs at tropical particular, the way this division will evolve in the future is highly latitudes, and also one of the major sources at midlatitudes in all uncertain. In particular, there are indications that the QBO is seasons except winter (when orographic generation is dominant affected by climate change. While the QBO was stable from when at middle and high latitudes). it was first observed in the 1950s until 2016, since then two major Convection excites GWs via different processes and at different 11,12 unexpected disruptions have occurred . To understand and wavelengths. Past work using case studies has in general focused 5,6 predict the future of the QBO, a better understanding of its on wave excitation by single convective cells and upscaled their fundamental driving forces is essential. For the BDC and MRC findings from convective source modelling to global distribu- tions . According to these simulations, the major part of the meanwhile, while the winter BDC is primarily driven by planetary- 13 14 momentum flux transported by convective GWs is expected to be scale waves, the summer BDC and the MRC are driven mainly found at vertical wavelengths λ ≤ 10 km and λ ≪ 100 km. by GWs. As such, uncertainties in the BDC due to the interaction of z h different processes and in particular due to missing GW drag However, convection often becomes organised and forms in the tropics and subtropics mesoscale convective clusters, which act as prevent models from consistently predicting how it will evolve in 15–18 wave sources as a whole and excite GWs of much longer coming decades . horizontal wavelengths . Full-complexity very high-resolution The portion of the GW spectrum visible to a space observation global ECMWF IFS simulations point to the coexistence of both technique is governed by the observation geometry, the optical processes. This, however, is ill-constrained by observations. depth of the considered emission and the field of view of the Characterising this partitioning is important for understanding instrument. The different observational geometries suitable for the dynamic structure of the middle atmosphere. This is because this task, as described in ref. , are typically grouped into limb, GWs at low latitudes are vital to the driving of three major nadir and sub-limb sounding. Limb observations, which cover a 1 2 3 Centre for Space, Atmospheric and Oceanic Science, University of Bath, Bath, UK. Forschungszentrum Jülich, Jülich, Germany. European Centre for Medium-Range Weather Forecasts, Reading, UK. email: C.Wright@bath.ac.uk Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA 1234567890():,; C.J. Wright et al. important exception to this was the astronomical Midcourse Space Experiment (MSX), which observed 4.3 μmCO emissions and occasionally used a sub-limb mode. Out of 80 such scenes (of 20 minutes’ duration) recorded by MSX, a number contained GW signals and one was analysed in detail showing signatures of GWs of ~50 km horizontal wavelength arising from a thunder- storm. At the time of the MSX observations, such linear 1D arrays sensitive to the near-infrared were very advanced technology, but since then both detector and cryo-cooler technology have greatly advanced. This allows for the compact and relatively low-priced realisation of even more powerful approaches which would allow the measurement of both a high number of pixels across track as well as a multitude of angles at the same time. In this paper, we propose an observation concept to study this missing part of the GW spectrum using spaceborne measure- ments of mid-infrared emissions, based on the saturated emission approach in ref. . To assess the potential performance of an instrument of this type, we carry out idealised simulations and then sample a high-resolution weather forecasting model with the expected observational characteristics of such an instrument. Finally, we describe the technological underpinnings needed for such a sounder. RESULTS Observational filter We first estimate how well the envisioned instrument would capture plain sinusoidal GWs depending on their horizontal and vertical scales. To do so, we initially generate a simplified atmosphere with a climatological midlatitude temperature and CO profile, working on a horizontal grid of 1 × 1 km spacing with a vertical step of 100 m. This temperature is then perturbed with a sinusoidal monochromatic wave of temperature amplitude 10 K, and the resulting temperature perturbations are retrieved using the JURASSIC2 forward model (see “Methods”). The simulations are then repeated with systematically varying vertical and Fig. 1 Sketch of viewing geometry. From a position in orbit, the horizontal wavelengths. Using this perturbed atmosphere, we instrument views downward by an angle α below the local compute the brightness temperature the satellite would measure horizontal. Infrared thermal radiation is emitted from the saturation at varying sub-limb angles, extracting one brightness temperature layer (indicated by the bright-red ring segment above the Earth). estimate per kilometre in the horizontal. The simulated brightness temperature measurements are then wide range from short to long vertical wavelengths, have analysed with a simple least squares fit for the largest sinusoidal facilitated momentum budget studies of GWs from MCCs with 20–22 frequency, using the results of a Fast Fourier Transform as an initial wavelengths λ ≥ 100 km . Nadir observations, meanwhile, are guess. Finally, the amplitude of this GW signal is computed, and sensitive only to long vertical wavelengths (λ > 15 km). This leaves the ratio between the true amplitude of 10 K and the derived the postulated major portion of parameterised CGW momentum flux, at scales both short in the horizontal and in the vertical, value is shown in Fig. 2 across a range of wavelengths at eight uncovered. This gap in our observational coverage is a major selected sub-limb viewing angles. We find no significant bias problem since it leaves us uncertain as to how the momentum between the derived wavelength and true horizontal wavelength flux these GWs transport is partitioned between single-tower and for all simulations with sufficient sensitivity to extract such MCC sources. estimates. At a physical level, limb sounding requires optically thin Convective GWs have typically vertical wavelengths of the order emissions and the observable scales are limited by the of 6–15 km , corresponding to a phase speed range of 20–50 m/s. integration of radiance along the line of sight (LOS). Nadir From Fig. 2, we see that we are sensitive to these wavelengths at sounding, meanwhile, is based on radiances saturating around a all σ between 25 and 60°, but that this sensitivity vanishes for 80° target altitude, with the vertical resolution limited by the width and nadir. We can assess this by considering a fixed vertical wavelength, e.g., 10 km—at such a specified value, the region of of this saturation. In the sub-limb, however, while emissions also maximal sensitivity sweeps the horizontal wavelength range, saturate along the LOS, the sensitivity to GW wavelength is controlled by the alignment of the 3D GW phase fronts with the maximising at ~λ = 50 km for 25° and λ = 7 km for 60°. Thus, h h LOS. This geometry is illustrated in Fig. 1: here, the sub-limb covering the wave field with a wide range of σ, one can estimate viewing angle α and the flight altitude determine the angle σ the aspect ratio of the GWs, and from the horizontal structure which the LOS forms with the local horizontal. σ = 0 corresponds determine the horizontal wavelength. In the ideal case of to the highest tangent height and marks the limit between sub- monochromatic waves, this would also allow us to infer the limb and limb; for a flight altitude of 400 km, it is reached for α vertical wavelengths by combining the horizontal wavelength and close to 20°. the sensitivity dependence. In the past, this sub-limb geometry was mainly used by We now proceed to consider two more realistic GW cases, both microwave instruments (ref. , and references therein) where the derived from a very high-resolution global numerical weather comparably large vertical field of view limited sensitivity. An prediction model simulation. npj Microgravity (2023) 14 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA 1234567890():,; C.J. Wright et al. Fig. 2 Estimated effect of the observational filter for selected sub-limb viewing angles from 25 to 90°. a–h Colours and line contours show the fraction of the input signal amplitude recovered for GWs of this horizontal and vertical wavelength. The thick solid line shows the expected optimum visibility where the LOS angle σ aligns with the orientation of the phase fronts given by the ratio of vertical and horizontal wavelength. Model by an instrument at the height of the ISS travelling northward along each meridian. We repeat this for a range of sub-limb We use the output from a TCo7999L137 (~1.4 km horizontal, 137- observing angles, across the range 25–90°. level vertical) run of the hydrostatic ECMWF Integrated Forecast- ing System (IFS) , using a 60-s model timestep. This simulation has the highest spatial resolution ever used in a seasonal- Gravity wave analysis timescale global atmospheric model and represents an excep- The above step provides global estimates of brightness tempera- tional scientific resource for studies of this nature. In particular, it tures for a range of sub-limb observing angles and a backward- has a substantially higher spatial resolution than the global-scale looking observer flying along the meridians. We now proceed to observing system simulation experiment ‘nature runs’ often used analyse these data for GW signatures, to assess the effect of in satellite development studies . Because the IFS simulation varying the sub-limb angle on how well the GWs are recovered at used only resolves hydrostatic waves, our analysis only considers each angle. The simulated observations are analysed using the these waves. 28 two-dimensional Stockwell Transform in ref. , as modified for Compared to lower resolution versions of the IFS, this model 29,30 computational efficiency in refs. . Due to the very large volume resolves a much larger part of the GW spectrum, with equal of the data when considered at global scales, the computational contribution of small-scale GWs with λ < 100 km versus larger complexity of applying techniques of this nature to such large scale GWs with λ > 100 km to the convectively generated GW volumes of data, and the brief format of this study we focus here momentum flux . The model runs without GW or convective on two cases: a large orographic wave over the Ural mountains parameterisations, and thus all GWs in the model are explicitly (Fig. 3a) and a group of convective waves southeast of Japan generated and propagate freely on the model grid. The simulation (Fig. 4b). Other sampled cases provide results broadly consistent was initialised from the ECMWF operational analysis on the 1st of with these two examples. November 2018 and then ran freely until the sampled dates, We first interpolate the brightness temperature data onto a detailed below for each case individually. regular 1 × 1 km spatial grid centred around the GW feature of interest. Interpolation to a regular spatial grid is required as the Brightness temperature simulations S-Transform analysis is based on Fast Fourier Transform algo- We compute the simulated brightness temperature perturbations rithms. For our Urals case, this grid is centred at 55°N, 115°E and from this dataset in a linearised fashion to save computation time. extends ±25° in longitude and ±14° in latitude at mean longitude and latitude respectively, while for the Japan case, it is centred at To that end, we must first remove the temperature background 33.5°N, 146°E and extends ±12° in both longitude and latitude. We from the model temperature structures . We then configure and carry out this interpolation separately for data extracted at all run new 2D simulations using JURASSIC2. Unlike the regular spatial grid used for the observational filter calculation, we carry viewing angles from 25 to 60°, and the results from each viewing out these simulations using a coarser horizontal grid of spacing angle assessed are shown individually. 0.025° in latitude and a vertical grid of 0.5 km, approximating the We then apply the 2D ST to these data. This analysis is carried out on the regular spatial grid, which at the centre location has grid spacing used in the spectral IFS run. Cartesian x and y axes corresponding to the zonal and meridional We then compute the 2D Jacobian matrix (also called a directions. In both these directions, we use a scaling parameter c sensitivity kernel) of a single brightness temperature measure- 28,31 ment for a climatological atmosphere, to effectively linearise the of 0.25 and limit the maximum horizontal wavelength to complex forward model. By convolving this sensitivity matrix with 500 km, and we permit the analysis to find the strongest 1000 2D IFS-derived temperature perturbations individually along each frequencies present in the data. These choices are consistent with 0.025°-spaced meridian, we directly compute the (linearised) previous studies using the ST. Figs. 3b–f and 4b–f show maps of brightness temperature perturbations which would be observed the estimated wave amplitudes calculated using this approach. Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA npj Microgravity (2023) 14 C.J. Wright et al. Fig. 3 Temperature amplitudes recovered for a sample orographic wave over the Urals. a Temperature output at 37 km altitude of the TCo7999 simulations. b–g Estimated amplitude of this wave as measured using a 2D Stockwell Transform method, for brightness temperatures computed from model output sampled across a range of viewing angles. Black boxes show region averaged over to generate Fig. 5. Note that amplitude values in panel (b) have been downscaled by a factor of 3 to aid visual comparison (see text for details). Case study results cases, the measured amplitudes have been normalised to the values calculated for the 25° sub-limb viewing angle (themselves We consider first the GW case over the Urals, Fig. 3. The phase ~50% reduced from the original model fields), in order to focus on fronts of this wave are aligned north-south, consistent with the the relative amplitude reduction as a function of angle. This north-south topography of the underlying mountain range, and highlights the rapid decline in signal strength for the convective this combined with the morphology and large amplitude of the case in comparison to the orographic case. While the amplitudes wave gives significant confidence that this wave is orographic in measured for the orographic case drop never below 50% of the origin. We see peak amplitudes >1 K across all viewing angles, best-case geometry considered, the convective case rapidly falling from a peak amplitude of >3 K (note saturated colour scale) decreases, stabilising at around 30% of the original amplitude at the lowest viewing angle to ~1.25 K at 60°. As we vary the for all angles above ~50°. viewing angle of the instrument, the geographic location of the wave does not shift, and the morphology of the inferred amplitude field remains largely constant. The results seen for DISCUSSION the sub-limb viewing cases, particularly at low viewing angles, are Convective GWs have expected vertical wavelengths of 6–15 km morphologically consistent with those obtained from an equiva- and cover a huge range of horizontal scales. While nadir sounding lent S-Transform analysis applied to the raw model data, but lower instruments are almost insensitive to this wavelength range, limb in magnitude. The largest sensitivity at large viewing angles is sounders can resolve convective GWs excited by MCCs. Here the seen at 60°E, 54°N, where 3D analysis (not shown) indicates challenge is to accurately quantify momentum flux by bringing particularly low horizontal and long vertical wavelengths. 32,33 an IR limb imager into space . Convective GWs from single This case contrasts in several important ways with our Japanese towers, however, are not covered with the existing observation case, Fig. 4. In this case, the observed phase fronts are radial in techniques. morphology, and are geolocated over similar features in the Our results provide a clear example of the benefits of a sub-limb troposphere (not shown). They are therefore consistent with viewing geometry over a nadir or near-nadir geometry for convective GW generation. While again the waves remain at the observing the GWs produced by convective activity. In both the same location as the observation angle is increased, here the reduction in amplitude with increasing angle is much larger, theoretical and realistic cases considered, the amplitudes of GWs falling from a peak amplitude ~1.25 K at 25° to <0.2 K at 60°. The with the spectral properties typical of convective wave activity are suppressed significantly at high sub-limb angles (i.e., near-nadir exception, in this case, is a spot around 32°N, 140°E. There 3D geometries) and much better recovered (although still reduced) at analysis reveals northward propagating waves (phase fronts tilted to the instrument), with <50 km horizontal wavelengths and low sub-limb angles (i.e., sub-limb geometries). Due to advances in detector technology since the work different vertical wavelengths up to 20 km, i.e., the steepest waves are visible for the largest angles. described in ref. , highly sensitive infrared detector arrays for Figure 5 shows the different amplitude reductions with sub- wavelengths 4 μm with resolutions of order 1000 × 1000 pixels limb viewing angle in these two cases as comparative line plots, and frame rates ~100 frames/s are now available commercially. averaged over the black-outlined regions in Figs. 3 and 4. In both Since GW observations of the type needed for atmospheric npj Microgravity (2023) 14 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA C.J. Wright et al. Fig. 4 Temperature amplitudes recovered for a group of GWs associated with convection southeast of Japan. a Temperature output at 37 km altitude of the TCo7999 simulations. b–g Estimated amplitude of this wave as measured using a 2D Stockwell Transform method, for brightness temperatures computed from model output sampled across a range of viewing angles. Black boxes show region averaged over to generate Fig. 5. Note that amplitude values in panel (b) have been downscaled by a factor of 2 to aid visual comparison (see text for details). While the heavily reduced amplitude seen at the nadir is not a fundamental problem for wave detection in the idealised and simulatedcases considered here,itisamajorproblem forreal- world observations. Our idealised observational filter calcula- tions are inherently noise-free, and our model-based analyses lack both retrieval and instrumental noise. In the real world, however, the presence of a noise floor means that GWs with suppressed amplitude fall below this floor. As such, these waves become functionally undetectable even if, theoretically, a nadir observing geometry should allow some signal to be recovered from them. This has hampered studying of this important source of atmospheric GWs, leaving the partitioning of GW momentum flux between GWs from MCCs and single convective cells one of Fig. 5 Reduction in recovered amplitudes relative to 25° case. the large controversies in atmospheric physics. As such, a sub- Figure shows average GW amplitudes over the regions defined by limb geometry like the concept proposed here has significant black boxes in Figs. 3 (orange line) and 4 (blue line), as computed design benefits for the important scientificproblem of from brightness temperature estimates over a range of viewing angles. characterising convective atmospheric gravity waves from single convective cells. science simultaneously require a wide angle range and a small field-of-view (FOV) for each pixel, a low orbit such as that provided by the International Space Station (ISS) is highly METHODS favourable. From this orbit, the distance from the observing To simulate measurements of the envisioned instrument, we platform to the saturation altitude of the measurement is employ the Jülich Rapid Spectral Simulation Code Version 2 between 450 and 1100 km depending on the viewing angle, (JURASSIC2) forward model . This forward model uses the and an FOV width perpendicular to the instrument line of sight emissivity growth approximation to compute emissions over (LOS) 0.3–0.6 km (neglecting diffraction) would cover a swath of −1 the spectral band from 2320 to 2340 cm . JURASSIC2 is capable width 300–600 km for downward viewing angles of 25–60°, of 2D and 3D simulations and inversions, and here we employ a respectively (Fig. 1, where σ is defined as the ‘sub-limb angle’ simplified 2D forward model along the meridians, in order to hereafter). Despite the high orbital velocity of the ISS platform, reduce the computational complexity of the problem. As we wish the image recording rates needed to prevent smearing from to focus primarily on the effects of different viewing angles in the measurements with limited integration times are also easily vertical, to further simplify the analysis we do not account for FOV achievable with current-day technology. We note however that sensitivity estimates depend on the speed of the optics and are effects, i.e., each brightness temperature is computed using an thus beyond the scope of this brief study. infinitesimally thin ‘pencil’ beam. Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA npj Microgravity (2023) 14 C.J. Wright et al. Reporting summary 22. Ern, M. et al. The semiannual oscillation (SAO) in the tropical middle atmosphere and its gravity wave driving in reanalyses and satellite observations. Atmos. Further information on research design is available in the Nature Chem. Phys. 21, 13763–13795 (2021). Research Reporting Summary linked to this article. 23. Wu, D. L. & Eckermann, S. D. Global gravity wave variances from Aura MLS: characteristics and interpretation. J. Atmos. Sci. 65, 3695–3718 (2008). 24. Dewan, E. M. et al. MSX satellite observations of thunderstorm-generated gravity DATA AVAILABILITY waves in mid-wave infrared images of the upper stratosphere. Geophys. Res. Lett. The observational filter simulations presented in Fig. 2 have been generated 25, 939–942 (1998). computationally as described in the text and associated references and do not 25. Wedi, N. P. et al. A baseline for global weather and climate simulations at 1 km incorporate any underlying data. As the underlying simulation output used in resolution. J. Adv. Model. Earth Syst. 12, https://doi.org/10.1029/2020MS002192 Figs. 3–5 is extremely large (42GB per 1-h timestep) we cannot easily archive these (2020). data separately for this study, but the retrieved brightness temperatures for the 26. Holdaway, D. & Yang, Y. Study of the effect of temporal sampling frequency on regions and times used in our study have been archived at Zenodo (https://doi.org/ DSCOVR observations using the GEOS-5 Nature Run results (Part I): Earth’s 10.5281/zenodo.6637647) as Matlab-format data files. radiation budget. Remote Sens. 8, 98 (2016). 27. Strube, C., Ern, M., Preusse, P. & Riese, M. Removing spurious inertial instability signals from gravity wave temperature perturbations using spectral filtering Received: 30 May 2022; Accepted: 20 January 2023; methods. Atmos. Meas. Tech. 13, 4927–4945 (2020). 28. Hindley, N. P., Smith, N. D., Wright, C. J., Rees, D. A. S. & Mitchell, N. J. A two- dimensional Stockwell transform for gravity wave analysis of AIRS measurements. Atmos. Meas. Tech. 9, 2545–2565 (2016). 29. Wright, C. J., Hindley, N. P., Hoffmann, L., Alexander, M. J. & Mitchell, N. J. Exploring gravity wave characteristics in 3-D using a novel S-transform technique: REFERENCES AIRS/Aqua measurements over the Southern Andes and Drake Passage. Atmos. 1. Fritts, D. C. & Alexander, M. J. Gravity wave dynamics and effects in the middle Chem. Phys. 17, 8553–8575 (2017). atmosphere. Rev. Geophys. 41,1–68 (2003). 30. Hindley, N. P. et al. Gravity waves in the winter stratosphere over the Southern 2. Palmer, T. N., Shutts, G. J. & Swinbank, R. Alleviation of a systematic westerly bias Ocean: high-resolution satellite observations and 3-D spectral analysis. Atmos. in general circulation and numerical weather prediction models through an Chem. Phys. 19, 15377–15414 (2019). orographic gravity wave drag parametrization. Q. J. R. Meteorol. Soc. 112, 31. Stockwell, R., Mansinha, L. & Lowe, R. Localization of the complex spectrum: the S 1001–1039 (1986). transform. IEEE Trans. Signal Process. 44, 998–1001 (1996). 3. Williams, P. D. & Joshi, M. M. Intensification of winter transatlantic aviation tur- 32. Preusse, P. et al. New perspectives on gravity wave remote sensing by space- bulence in response to climate change. Nat. Clim. Change 3, 644–648 (2013). borne infrared limb imaging. Atmos. Meas. Tech. 2, 299–311 (2009). 4. National Research Council. Solar and Space Physics: A Science for a Technological 33. Preusse, P. et al. Characteristics of gravity waves resolved by ECMWF. Atmos. Society (The National Academies Press, 2013). Chem. Phys. 14, 10483–10508 (2014). 5. Piani, C., Durran, D., Alexander, M. J. & Holton, J. R. A numerical study of three- 34. Ungermann, J., Hoffmann, L., Preusse, P., Kaufmann, M. & Riese, M. Tomographic dimensional gravity waves triggered by deep tropical convection and their role in retrieval approach for mesoscale gravity wave observations by the PREMIER the dynamics of the QBO. J. Atmos. Sci. 57, 3689–3702 (2000). infrared limb-sounder. Atmos. Meas. Tech. 3, 339–354 (2010). 6. Choi, H.-J. & Chun, H.-Y. Momentum flux spectrum of convective gravity waves. 35. Weinreb, M. P. & Neuendorffer, A. C. Method to apply homogeneous-path Part I: an update of a parameterization using mesoscale simulations. J. Atmos. Sci. transmittance models to inhomogeneous atmospheres. J. Atmos. Sci. 30, 662–666 68, 739–759 (2011). (1973). 7. Choi, H.-J., Chun, H.-Y., Gong, J. & Wu, D. L. Comparison of gravity wave tem- perature variances from ray-based spectral parameterization of convective gravity wave drag with AIRS observations. J. Geophys. Res. Atmos. 117, 5115 (2012). 8. Trinh, Q. T. et al. Tuning of a convective gravity wave source scheme based on ACKNOWLEDGEMENTS HIRDLS observations. Atmos. Chem. Phys. 16, 7335–7356 (2016). This research used the resources of the Oak Ridge Leadership Computing Facility 9. Polichtchouk, I., Wedi, N. & Kim, Y.-H. Resolved gravity waves in the tropical (OLCF), which is a DOE Office of Science User Facility supported under Contract DE- stratosphere: impact of horizontal resolution and deep convection parametriza- AC05-00OR22725. C.J.W. is funded by Royal Society University Research Fellowship tion. Q. J. R. Meteorol. Soc. 148, 233–251 (2022). UF160545 and by NERC grants NE/S00985X/1, NE/V01837X/1 and NE/W003201/1. 10. Scaife, A. A. et al. Predictability of the quasi-biennial oscillation and its northern winter teleconnection on seasonal to decadal timescales. Geophys. Res. Lett. 41, 1752–1758 (2014). 11. Osprey, S. M. et al. An unexpected disruption of the atmospheric quasi-biennial AUTHOR CONTRIBUTIONS oscillation. Science 353, 1424–1427 (2016). C.J.W.: conceptualisation, methodology, software, formal analysis, investigation, 12. Anstey, J. A. et al. Prospect of increased disruption to the QBO in a changing writing and visualisation. J.U.: methodology, software, formal analysis, investigation, climate. Geophys. Res. Lett. 48, e2021GL093058 (2021). 13. Alexander, M. J. Gravity-wave forcing in the stratosphere: observational con- writing and data curation. P.P.: conceptualisation, methodology, investigation and straints from the Upper Atmosphere Research Satellite and implications for writing. I.P.: software, writing and data curation. parameterization in global models. J. Geophys. Res. Atmos. 108, https://doi.org/ 10.1029/2003jd003373 (2003). 14. Murgatroyd, R. J. & Singleton, F. Possible meridional circulations in the strato- COMPETING INTERESTS sphere and mesosphere. Q. J. R. Meteorol. Soc. 87, 125–135 (1961). 15. Butchart, N. The Brewer-Dobson circulation. Rev. Geophys. 52, 157–184 (2014). The authors declare no competing interests. 16. Polvani, L. M., Abalos, M., Garcia, R., Kinnison, D. & Randel, W. J. Significant weakening of Brewer-Dobson circulation trends over the 21st century as a consequence of the Montreal protocol. Geophys. Res. Lett. 45, 401–409 (2018). ADDITIONAL INFORMATION 17. Abalos, M. et al. New insights on the impact of ozone-depleting substances on the Brewer-Dobson circulation. J. Geophys. Res.: Atmos. 124, 2435–2451 (2019). Supplementary information The online version contains supplementary material 18. Richter, J. H. et al. Response of the quasi-biennial oscillation to a warming climate available at https://doi.org/10.1038/s41526-023-00259-2. in global climate models. Q. J. R. Meteorol. Soc. 148, 1490–1518 (2020). Correspondence and requests for materials should be addressed to Corwin J. Wright. 19. Wu, D. L. et al. Remote sounding of atmospheric gravity waves with satellite limb and nadir techniques. Adv. Space Res. 37, 2269–2277 (2006). Reprints and permission information is available at http://www.nature.com/ 20. Ern, M., Preusse, P., Kalisch, S., Kaufmann, M. & Riese, M. Role of gravity waves in reprints the forcing of quasi two-day waves in the mesosphere: an observational study. J. Geophys. Res.: Atmos. 118, 3467–3485 (2013). 21. Ern, M. et al. Interaction of gravity waves with the QBO: a satellite perspective. J. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims Geophys. Res.: Atmos. 119, 2329–2355 (2014). in published maps and institutional affiliations. npj Microgravity (2023) 14 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA C.J. Wright et al. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http:// creativecommons.org/licenses/by/4.0/. © The Author(s) 2023 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA npj Microgravity (2023) 14

Journal

npj Microgravity – Springer Journals

Published: Feb 8, 2023

There are no references for this article.