Decadal changes in atmospheric circulation detected in cloud motion vectors

Decadal changes in atmospheric circulation detected in cloud motion vectors | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Physical Sciences - Article Decadal changes in atmospheric circulation detected in cloud motion vectors Larry Di Girolamo, Guangyu Zhao, Gan Zhang, Zhuo Wang, Jesse Loveridge, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5296185/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 09 Jul, 2025 Read the published version in Nature → Version 1 posted You are reading this latest preprint version Abstract Changing atmospheric circulations shift global weather patterns and their extremes, with profound effects on human societies and ecosystems. Studies using atmospheric reanalysis and climate model data 1-9 indicate a variety of circulation changes in recent decades but show discrepancies in magnitude and even direction. Therefore, validation with independent, climate-quality measurements is urgently needed 3 . Here we use the satellite-observed, height-resolved cloud motion vectors from the Multi-angle Imaging SpectroRadiometer (MISR) 10,11 to analyze tropospheric circulation changes during 2000–2020. We find significant changes in tropospheric circulations, with upper tropospheric cloud motion speeds in midlatitudes increasing by up to ~4 m s⁻¹ decade⁻¹, primarily due to the strengthening of meridional flow that could indicate increased poleward trajectories or intensification of extratropical cyclones. Furthermore, the northern and southern hemisphere tropics shifted poleward at a rate of 0.42±0.22 and 0.02±0.14 °latitude decade⁻¹ (95% CI), respectively, while the corresponding polar front shifted at 0.37±0.31 and 0.31±0.21 °latitude decade⁻¹. Comparison with the widely used ERA5 12 reanalysis winds subsampled to MISR show good agreement with MISR’s climatological values and trends but indicate likely ERA5 biases in the upper troposphere. These MISR-based observations provide critical benchmarks for refining reanalysis and climate models to advance our understanding of climate change impacts on cloud and atmospheric circulations. Earth and environmental sciences/Climate sciences/Climate change/Climate-change impacts Earth and environmental sciences/Climate sciences/Atmospheric science/Atmospheric dynamics Figures Figure 1 Figure 2 Main Earth’s atmospheric general circulation regulates the climate and extremes that profoundly impact human societies and ecosystems. The circulation embodies a variety of components, including the meridional overturning circulation that occupies the tropics (i.e., Hadley circulation) and the turbulence-laden westerlies that prevail the midlatitude troposphere (i.e., midlatitude jets), each having their own covarying cloud and wind features. Understanding natural variations and human-driven trends in the atmospheric general circulation has long been a focus of climate research. For example, climate models and theoretical arguments have projected a slowdown of the overturning circulation 13 , a poleward expansion of the Hadley circulation 3 , and a poleward shift of midlatitude westerlies 1,2,14,15 . However, the quantitative projections are highly uncertain. The Coupled Model Intercomparison Project Phase 6 (CMIP6) models, for instance, show a mean expansion rate in the Hadley cell of both hemispheres of ~0.2 °latitude decade -1 between 1979-2008, but with a large spread of ~0.5 °latitude decade -1 amongst the models 8 . Evidence for such changes exists in satellite observations and atmospheric reanalysis datasets, but with mixed results regarding the magnitude of the expansion rate – some reaching as high as 3 °latitude/decade 16 . These mixed results arise partly from the diverse metrics used to describe features of the general circulation and the fundamental limitations in the datasets being used. For example, the “edge” of the Hadley cell has been defined with upper- or lower-tropospheric quantities. Definitions based on upper atmospheric quantities (e.g., outgoing longwave radiation, total ozone amount, gradients in tropopause height) are easily measurable by satellite instruments but have recently been deemed as unreliable for quantifying dynamical changes of the general ciculation 3,8 . Consequently, recent assessments of tropical expansion have focused on metrics from lower tropospheric data, which are considered more dynamically consistent. These metrics include the latitude of zero-crossing of the meridional stream function at 500 hPa and the latitude of zero-crossing of the near-surface zonal wind speed that divides the tropical easterlies from the midlatitude westerlies (hereafter Lat_U0 ) 3,8 . Such lower tropospheric metrics are easily derived from reanalysis datasets, thereby making these datasets the primary source for verifying climate model predictions 1,3,8,9 . However, this approach has important limitations. Both climate models and atmospheric reanalyses use dynamical models, which are subject to common uncertainties related to parameterized physics within the models 17-19 . Furthermore, reanalyses also assimilate observations 20 from a changing mix of satellite instruments and sparsely distributed surface and upper-air stations. Data discontinuities, calibration drifts, and the evolving nature of the global observing system contribute to time-varying uncertainties in the reanalysis data that are particularly problematic for trend detection. Thus, it is critical to validate atmospheric circulation changes using independent, global observations 3 . Satellite-based cloud motion vectors (CMVs) offer a valuable dataset for evaluating changes in the atmospheric circulation. Here, we provide the first analysis of climatological means and trends in height-resolved CMVs, binned to 1 km vertical bins, collected over the past two decades by MISR 10,11 onboard the Terra satellite. The MISR CMVs offer unique advantages in studying decadal trends compared to other satellite-based CMVs. Unlike other satellite Earth science records, Terra maintained a stable equator-crossing time (ECT) for >20 years. This stability and longevity eliminates the aliasing of diurnal variability into the record of longer-term variability, and avoids artificial discontinuities from stitching data records of multiple short-lived satellites 21 . Critically, MISR CMVs are derived stereoscopically 10,21 – thus they are independent of long-term drift in radiometric calibration and able to provide an extraordinary stable record for change detection (Methods). The height and speed of MISR CMVs are highly accurate and have an error budget that is self-contained (i.e., no dependence on ancillary meteorological data), fully traceable, and extensively validated 23-26 . Moreover, since MISR CMVs are not assimilated into any atmospheric reanalyses, they can serve as an independent validation dataset for climate model projections and reanalysis-based estimates of atmospheric circulation change. Decadal Trends in CMVs Our analysis proceeds with the understanding that CMVs are not representative of all winds at all altitudes at all times. Unlike reanalysis datasets, which do provide a continuous estimate of wind in space and time, CMVs are contingent on cloud being present, providing winds at the cloud-top-altitude of a detectable cloud layer at the time and location of MISR observations. Hence, the cloud-dependence of the CMV sampling results in a lack of sampling of clear sky conditions and conditions below cloud top. Still, clouds do occur frequently in the planetary boundary layer, even in fair weather, and only a few small boundary layer cumuli need to be present for MISR to retrieve a CMV at a resolution of 17.6x17.6 km 2 . Cloud tops (hence CMV samples) in the mid to upper troposphere are often associated with meso- to synoptic-scale weather disturbances, such as tropical and extratropical cyclones. The climatology of CMVs is also linked to climatic features such as the Intertropical Convergence Zone (ITCZ) and polar fronts 27,28 . As such, while conditionally sampled, CMVs capture a broad range of components in the atmospheric circulation. Interpreting trends in MISR CMVs as changes in cloud-top-conditioned circulations requires careful consideration of several potential confounding factors, as described and analyzed in Extended Data Analyses and Discussion (EDAD). Our EDAD shows that the impacts of these confounding factors on the observed trends are small, supporting the conclusion that much of the observed MISR CMV changes (Figure 1) are due to changes in the cloud-top-conditioned atmospheric circulation. Figure 1 shows the 21-year mean and statistically significant trend of the tropospheric zonal-mean CMVs. This figure is broken down by season in EDAD. The mean of the CMVs shows key climatological features in the circulation including the midlatitude westerly jets (Figure 1d) and the meridional overturning circulation (Figure 1g). The wind speed trends are largest in the midlatitude upper troposphere, where the CMVs show statistically significant increases of up to ~2 m s -1 decade -1 in the northern hemisphere (NH) and ~4 m s -1 decade -1 in the southern hemisphere (SH) (Fig. 1a). Figure 1d,g show both poleward and westerly flow components strengthening for the NH upper troposphere. But in the SH, the poleward flow component strengthens while the westerly flow component weakens. The SH zonal wind weakening between 30-50°S is accompanied by an increase in the mid-troposphere zonal wind (U) of ~ 0.5 – 1.5 m s -1 decade -1 near 60°S. These changes near the SH polar front suggests a poleward shift of the SH westerly flow and midlatitude storm track, which is consistent with climate model projections and recent findings from atmospheric reanalyses 14,15,29 . Interestingly, the overall wind speed increase mainly arises from the meridional wind (V). The strengthening of poleward flow indicated by CMV may be associated with warm conveyor belts or atmospheric rivers embedded in extratropical cyclones, which are key processes involved in midlatitude extreme precipitation. Together with the overall increase in water vapor content in a warmer atmosphere, the strengthening of poleward wind may contribute to the observed and simulated increases of extreme precipitation 15,30,31 . In the equatorial regions, CMV trends reveal a pattern potentially indicative of asymmetric Hadley cell changes. The mean CMVs in the equatorial upper troposphere show easterlies (Fig. 1d) with a strong southward component (Fig. 1g), owing to the climatological ascent associated with the ITCZ being ~6°N 32 . The trends in CMVs show a moderate increase in the upper-level easterlies in the SH deep tropics. In comparison, the V-component of CMVs show stronger and more widespread trends in the deep tropics of both NH and SH, with the southward component in the upper troposphere strengthening at a rate of up to ~4 m s -1 decade -1 and northward component in the lower troposphere strengthening at a rate of up to 2 m s -1 decade -1 . The strengthening trend of the upper tropospheric southward flow is consistent across seasons (Extended Data Figs. 5-8). This pattern in the CMV trends suggests the intensification of the cross-equatorial circulation that contributes to stronger low-tropospheric convergence and upper tropospheric divergence near 6°N. This would be consistent with the faster warming of the NH 33 and the associated circulation response required by the global energetic constraint 34 . However, internal climate variability 35 , such as the Atlantic Multi-decadal Oscillation, within our record could also be a contributing factor. Consistency with Atmospheric Reanalyses To assess the consistency of our findings with atmospheric reanalysis data, we compared the MISR CMV climatology and trends with those from the widely used ERA5 12 . We sample the ERA5 winds at the times and locations (latitude, longitude and altitude) of MISR CMVs to ensure identical sampling (see Methods). We refer to these samples as ERA5_MS (MISR Sampled). Of course, ERA5 may not be properly representing observed clouds; however, this misrepresentation was assessed in EDAD and found to have a negligible impact on our findings presented below. We also repeat the analysis using all wind data available (ERA5_AW), i.e. irrespective of the presence of cloud, in the ERA5 dataset sampled at 10:30 AM local time. ERA5_AW is a means to analyze how cloud-top-conditioned winds differ from all winds, and to better connect our analysis to other studies that use reanalysis datasets, albeit using diurnal means 1-9 . Where good agreement is found between ERA_MS and MISR CMV, we cannot reject the circulation and trends in ERA5_AW using MISR, thereby increasing our confidence in circulation and trends computed using ERA over the MISR record. Figure 1 shows good agreement between the MISR CMVs and the ERA5_MS winds, both for climatological means and trends between 2000-2020. However, there are some notable differences. In the upper troposphere of the deep tropics, the climatological means of the ERA5_MS speeds are smaller than the MISR CMV by ~2 to 8 m s -1 . In midlatitudes, the upper tropospheric wind speed of ERA5_MS is greater than MISR CMV by ~1 to 3 m s -1 in the NH and 1 to 2 m/s smaller in the SH. These differences are driven primarily by the V-component. Since these differences are much larger than the uncertainty in MISR CMVs (Methods and EDAD), these findings likely indicate that further improvements to ERA5 are needed to reduce these upper tropospheric biases – a finding supported by other evidence (see EDAD). The means of ERA5_MS and ERA5_AW also differ since ERA5_MS are non-random samples (equal to MISR CMV sampling). Compared to ERA5_AW, the climatological MISR CMVs and ERA5_MS shows stronger V-component and weaker U-component in the midlatitudes, corresponding to cloudy, poleward air streams embedded in cyclones. Focusing on decadal trends, the upper tropospheric midlatitude jets show similar patterns between the independent MISR CMV and ERA5_MS datasets. They show an increase in cloud motion speed in both hemispheres, a strengthening trend of poleward flow in the midlatitude upper troposphere, especially in the SH midlatitude, and an increase in U along the polar front in the SH. But there are also important, statistically significant differences (Extended Data Fig. 1). CMV speed trends are larger by up to ~2 m s -1 decade -1 compared to ERA5_MS in the upper troposphere of the SH and tropics. This is true for the V components as well, except for the striking difference between ~20°S and 40°S, where the ERA5_MS V-component displays trends that are ~4 m s -1 decade -1 weaker than MISR. Near this region, the ERA5_MS U-component trends are up to 2.5 m s -1 decade -1 larger than MISR, with ERA5_MS showing a trend of up to +2.5 m s -1 decade -1 just north of ~30°S and MISR CMV showing a trend of -2.5 m s -1 decade -1 just south of ~30°S. Since MISR CMV retrievals are agnostic to location, and since these trend differences are much larger than the uncertainty and stability in MISR CMVs (Methods), we suspect an issue with ERA5 – perhaps erroneous trends in the input data assimilated into ERA5 that would impact ERA5_MS trends in the upper troposphere of the SH and tropics. When we also consider ERA5_AW, the increase in U along the SH polar front is the only common dominant change in the mid-latitudes amongst the three datasets and has been observed in other reanalysis datasets as well 2 . The weak trends that appear in the ERA5_AW V-component in midlatitudes, where MISR CMV and ERA_MS show strong poleward trends, suggests a potential compensating trend in the equatorward meridional flow in clear-sky conditions. In the tropical upper troposphere, the trend agreement between ERA5_AW and ERA5_MS V-components may be due to the suspected erroneous trends in the inputs to ERA5 noted above. Expansion and Migration Rates To assess the expansion rate of the Hadley cells, we apply the Lat_U0 metric to all three datasets using winds in the 0-1 km altitude bin (see Methods). Figure 2a,b shows the time series of deseasonalized monthly anomalies in the latitudinal position of Lat_U0 for NH and SH, respectively. The correlation coefficient, r , in the monthly anomalies amongst these datasets is strong, particularly between the MISR CMV and ERA5_MS in the NH ( r = 0.87) and SH ( r = 0.95). The mean monthly values of the Lat_U0 (the insets in Fig. 2a,b) indicate agreement in the seasonal cycle of the edge of the Hadley cells amongst the three datasets, but with absolute differences of up to ~2° in latitude. All datasets show a poleward expansion rate of ~0.3 – 0.5 ± 0.2 (95% CI) °latitude decade -1 in the NH and ~0 – 0.2 ± 0.1 (95% CI) °latitude decade -1 in the SH, with only a trend in the NH appearing at a reasonable confidence level based on their low P -values (see methods). These estimates of poleward expansion rate between 2000 – 2020 are in line with multi-model mean values in climate model and other reanalysis model estimates for earlier periods that use the Lat_U0 metric 3,8,16 , though the potential impact of internal climate variability on individual model simulations and relatively short observational record should be kept in mind. A common lower-tropospheric metric to examine the zonal mean position of the polar front jet is the latitude of the maximum value in monthly mean U ( Lat_U max ) at the 850 mb pressure level 5,16 . Here we use the 1 – 2 km altitude bin, where the 850 mb pressure level typically resides (see Methods). Fig. 2c,d shows the deseasonalized monthly anomalies in the latitudinal position of Lat_U max . The correlation coefficient in the monthly anomalies amongst these datasets is excellent, particularly between the MISR CMV and ERA5_MS in the NH ( r = 0.98) and SH ( r = 0.99). The mean monthly values of Lat_U max (the insets of Fig. 2c,d) also indicate agreement in the seasonal cycle in the position of the polar front jet, particularly between the MISR CMV and ERA5_MS datasets. From these datasets, we see a poleward migration rate of ~0.3 – 0.4 ± 0.3 (95% CI) °latitude decade -1 in the NH and ~0.3 – 0.4 ± 0.2 (95% CI) °latitude decade -1 in the SH for all three datasets, with trends at only moderate to low confidence levels based on their P -values. This low to moderate confidence is consistent with the confidence reported by the IPCC 1 . Discussion and Conclusion Unique to this study is the use of stable and accurate height-resolved cloud motion vectors (CMVs) from MISR to assess climatological values of cloud motion and their variability between 2000-2020 (Fig. 1). We show that CMVs have significantly (95% CI) increased in speed in the upper troposphere during this period in both NH and SH by up to 2 and 4 m s -1 decade -1 , respectively. This speed increase occurs with an increase in meridional flow towards the poles in both hemispheres, but with westerly flows strengthening in the NH and weakening in the SH. This poleward increase in the meridional component of the CMV could indicate an intensification, an increase in moisture transport in the warm sector, a change in the overall structure, or a poleward shift in the tracks of extratropical cyclones. High confidence (P = 0.06) is placed in a poleward expansion of the Hadley circulation in the NH at a rate of 0.42 ± 0.22 (95% CI) °latitude/decade using the Lat_U0 metric applied to CMVs. No significant expansion was observed in the SH. The Lat_Umax metric applied to CMVs suggests that the clouds associated the polar jets are migrating poleward at similar rates to each other in both hemispheres, but only with low to moderate confidence levels. When considering CMV changes at all altitudes throughout the tropics, the observations suggest weakening of the Hadley circulation in the NH and strengthening in the southern hemisphere; the degree to which cannot be quantified with CMVs as they do not represent mass stream functions that are commonly used in assessing circulation strength 9 . We provided the first evaluation of systematic errors in ERA5 mean winds and their trends throughout the troposphere against independent, height-resolved MISR CMVs. This evaluation is important because the suitability of reanalysis data for trend detection in atmospheric circulation, in part for validating climate model projections, has raised concerns 3 . To facilitate a direct comparison, the ERA5 winds have been sampled (ERA5_MS) to match the time and location of MISR CMVs. We show very good agreement between MISR CMVs and ERA5_MS winds. Nonetheless, small, but significant differences between MISR CMV and ERA5_MS are present in the upper troposphere. These differences are much larger than the uncertainties in MISR CMVs, suggesting that the ERA5 data likely suffer systematic biases in the upper troposphere, specifically in the SH and the tropics. Our ERA5_MS cloud-conditional analysis presented in EDAD also points to potential problems in ERA5’s ability in representing clouds. However, this had little impact on our results, which we suspect may be due to ERA5 winds being strongly constrained by the assimilation of observational data (e.g., rawinsondes) that are independent of the presence of cloud. This raises concern in using ERA5 for studying one of the key science questions in climate science, namely “How do clouds and circulation interact?” 36 Despite some upper-troposphere differences, we show excellent agreement between MISR CMV, ERA5_MS and ERA_AW in the Hadley cell expansion rates and the polar front migration rates for the 2000-2020 period using the Lat_U0 and Lat_Umax metrics. Excellent agreement amongst these datasets is also observed in the seasonality of these metrics. Therefore our findings support the use of MISR CMVs and ERA5 to monitor changes using the Lat_U0 and Lat_Umax metrics that are favored by the community. However, caution is still recommended when extending this finding beyond this period, particularly moving backwards in time because of the reduced capabilities in our global observing system that are assimilated into ERA5. The higher level of confidence in MISR wind speed trends in the upper troposphere relative to lower altitudes may suggest that signals of warming-induced circulation changes may first emerge in the upper troposphere, at least in cloudy conditions. ERA5_MS also picks up these trends, but the trends in ERA5_AW (i.e., all clear and cloudy winds) are much weaker. This may imply that there is a compensating trend in the meridional flow towards the equator in clear conditions. Together, they may suggest an intensification in the variability of tropical-extratropical transports that make the poleward transport of moisture and heat more extreme. This has important implications for human societies sensitive to precipitation and temperature extremes. Changes to atmospheric circulations are a critical component of climate change that is already impacting modern society 37 . Our assessment of these changes in the 2000-2020 period using MISR CMVs provides much-needed benchmarks for reanalysis and climate model datasets. Still, while MISR is our longest climate-quality record from satellites for height-resolved CMVs, it is still short in light of internal climate variability. Therefore any discussion in our analysis of change is specific to changes over the past two decades only, which likely contain natural climate variability and human-induced changes. Here, we focused on zonal mean CMVs, ignoring the finer regional details important to understanding and quantifying regional impacts 2 . Given the larger regional internal variability of CMVs, a record longer than the MISR dataset would help detect critically important changes in regional circulations and should be pursued. Methods Datasets This work uses the MISR Cloud Motion Vector (CMV) product (version F02_0002) 10 over the period of March 2000 to December 2020. The product is not a usual gridded, monthly mean product normally used in climatological studies. Instead, the product contains a simple list of all CMV retrievals for a given month, with each retrieval tagged by latitude, longitude and time. The height-resolved CMVs are obtained through stereoscopic means by tracking the progression of features in the MISR 275-m resolution red-band imagery (380-km swath) over a 3.5-minute period between the initial 70° forward view and the nadir view, and again for the 3.5-minute period between the nadir view and 70° aft view 10 . The resolution of the MISR CMV product is 17.6km x 17.6km. Our analysis uses only the daytime descending node of the MISR orbit to keep local time consistent within high latitude grids. The latest version of MISR cloud top heights and cloud motion vectors have been extensively validated 24-26 . The near global validation of cloud top height has been validated against a space-based lidar 26 , showing a bias ± precision of -280m ± 370m. The precision in cloud motion speed is 3.7 m/s, with biases in U = 0.0 m/s and V = 0.3 m/s relative to static ground targets, and with biases in U and V relative to geostationary (for cloud top heights where they have moderate agreement) derived cloud motion vectors < ±0.5 m/s and possibly up to -1.5 m/s for the V-component, depending on the method of assessment 24,25 . The stability of the product is also relevant for trend analysis. While MISR geometric telemetry needed for stereoscopic retrievals indicate no trends over the operation of the mission (Veljko Jovanovic, personal communication), we nonetheless perform here the first analysis to quantify its stability. The MISR stereographic retrievals are agnostic to the texture being observed, be it from cloud or land surfaces, receiving no prior. Therefore, we use the surface as a stable target for measuring the stability of the MISR TC_Cloud_F0_0001 product 21 , which is the main input to the CMV product. We analyzed cloud top height and wind retrievals from data flagged as “high-confidence near-surface” by the Stereoscopically-Derived Cloud Mask in the TC_Cloud product, which typically indicate clear sky or the occasional near-surface cloud. Our analysis encompasses 20 years of global land data between 50°N and 50°S as in Mitra et al. 26 . We conducted a trend analysis on the modes (rather than mean to avoid any possible trend in near-surface clouds) of annual histograms of these retrievals. For the surface heights, the trend is small at 0.54 ± 2.5 m/decade (95% CI) per decade and insignificant (p-value = 0.94). Near-surface wind retrievals also exhibit negligible trends: the U-component shows a trend of 0.00 ± 0.01 m/s/decade (p-value = 0.94) and the V-component indicates a trend of 0.02 ± 0.05 m/s/decade (p-value = 0.51). These results confirm the long-term stability and reliability of MISR stereo measurements for climate research. For the reanalysis model dataset, this study uses hourly data of the fifth-generation European Centre for Medium-Range Weather Forecasts (ECMWF) Atmospheric Reanalysis (ERA5) 12 . We use the U and V components of ERA5 wind, as well as the geopotential, on all of the available 37 pressure levels ranging from 1000 hpa to 1 hpa. These hourly data are downloaded at a global 0.25° × 0.25° latitude-longitude grid, the highest spatial and temporal resolutions available for the ERA5 archive. While the quality of ERA5 winds is evaluated against MISR in Main, we provide additional discussion on ERA5 wind evaluation against other independent datasets below in Extended Data Analysis and Discussion, showing excellent agreement with MISR in the very limited regions that the other datasets report on. Sampled ERA5 data for each MISR CMV data record (ERA5_MS) Since the ERA5 data has a spatial-temporal resolution that is comparable to MISR CMV, we adopt a nearest-neighbor approach to sample the ERA5 U and V at the time, location, and altitude of each CMV retrieval. The time, latitude and longitude for each MISR CMV retrieval at a 17.6km x 17.6km resolution are used to find the closest hour and the nearest grid point of the ERA5 data. For a CMV retrieval at a specific height, we locate the nearest ERA5 data point using the geopotential information at pressure levels. Specifically, geopotential heights are calculated by dividing the geopotential values by the Earth’s gravitational acceleration, given by 9.80665 m/s 2 (constant). Hence, the sampled ERA5 data (ERA5_MS) have the exact same record length as the MISR CMV data. Wind speed is calculated from the U and V components for each record of CMV and ERA5_MS. Trend analysis Before trend analyses, the U, V and wind speed data of MISR CMV and ERA5_MS are first aggerated into monthly, 0.25° × 0.25° latitude-longitude grid boxes. The aggregated data is then sorted into 20 height bins ranging from 0 to 20 km with a bin width of 1km (with closed left side and opened right side). The mean of all the 17.6 km retrievals in each grid box and height bin is calculated and stored into an intermediate file along with the number of the retrievals for each bin. Hence, in one monthly intermediate file, U and V are stored into 720 (latitude) × 1440 (longitude) x 20 (altitude) bins. For the zonal analysis (Fig. 1), the total number of the bins is further reduced to 720 (latitude) x 20 (altitude) bins by averaging the data along the longitudinal dimension excluding the bins with no valid retrievals (e.g., due to high altitude terrain lying above say the 0-1 km altitude bin). The zonal map of the total number of CMV retrievals is given in Extended Data Fig. 2. To ensure a large sample size, only bins that have a total of > 5000 CMV retrievals over the 2000-2020 period are used in the zonal analyses. This effectively removed the low-sample observations of the stratospheric clouds and the associated wind speed, thus keeping the focus of our discussion to the troposphere. As a reference for readers, the mean tropopause heights are plotted in Figure 1. The mean tropopause heights were derived for the period 2000 and 2020 using the tavgM_2d_slv_Nx product of the Modern-Era Retrospective analysis for Research and Applications, Version 2 (MERRA-2) 38 , as tropopause heights are not directly available in ERA5. The deseasonalized monthly anomalies for each bin are calculated as the deviation from the monthly means averaged over the 2000-2020. Trends analysis of the deseasonalized anomalies is conducted by initially applying the nonparametric Mann-Kendall test for the trend and the nonparametric Sen’s method for the magnitude of the trend using the python package “pyMannkKendall” 39 . In all analyses and figures involving trend analysis, after performing local significance tests, we applied a tightened False Discovery Rate (FDR) correction, following the same procedure described by Ventura et al 40 . We aimed to control the FDR at a nominal level of 5%; however, the significance threshold was adjusted based on the estimated proportion (50%) of true null hypotheses to maintain this control, as recommended by Ventura et al 40 . This adjustment often resulted in a higher nominal significance threshold, increasing the power to detect true effects while ensuring that the expected proportion of false discoveries remained at or below 5%. Grid points with adjusted p-values below the adjusted FDR threshold were considered statistically significant. Calculation of Lat_U0 and Lat_Umax This study evaluates the tropical width and the position of the polar jets using the zonal averages of the U data in the CMV, ERA5_MS and ERA5_AW. We use two metrics from the Tropical-width Diagnostics (TropD) software package 41 . Using TropD allows for consistency with other studies. Lat_U0 is the latitude at which the zonal-mean U-component of the wind in the 0-1km altitude bin equals zero after linear interpolation between two neighboring latitude bins. This marks the latitude in the subtropics where U switches sign from negative (easterly) to positive (westerly). It is calculated using the TropD_Metric_UAS module in TropD with default settings. Lat_Umax is the latitude of maximum zonal-mean U in the 1-2km altitude bin. It is calculated using the TropD_Metric_EDJ module in TropD. We use the “peak” method (weak smoothing) with the smoothing parameter of n = 6 as recommended in other studies 42,43 . The other parameters in the module are set with default settings. Extended Data Analyses and Discussion Examination of Confounding Factors There are potential confounding factors at play when interpreting trends in MISR CMV as trends in speed, namely trends in the number of MISR CMV samples, their within-bin heights in the presence of within-bin vertical gradients in wind speed, and their within-bin longitudes in the presence of within-bin horizontal gradients in wind speed. Extended Data Fig. 2 shows the number of CMV samples and their trends in terms of the within-bin percentage change. We see that the observed statistically significant trends in CMV samples is very small, mostly ranging from -0.6 to +0.2 %/decade. There is a decreasing fraction of CMV samples in the upper troposphere and an increasing fraction in the lower troposphere. These should not be compared to cloud cover changes because a CMV retrieval is not sensitive to the underlying cloud fraction (i.e., whether the 17.6 km x 17.6 km area is 100% cloudy or 5% cloudy, we still get a CMV sample). Moreover, the positive trend in the lower troposphere is confounded by the decreasing trend in the upper troposphere, since less clouds above leads to more opportunity to retrieve clouds below. The decreasing trend in the upper atmosphere may be related to decreases in the frequency of occurrence of optically thin cirrus that reside near the detectability threshold of MISR stereo 26,44 . Note the spatial patterns in the small trends shown in Extended Data Fig. 2b do not match the spatial patterns we see in the trends in Fig. 1 a,d,g, which does not support the notion that sample trends alone can explain the trends seen in Fig. 1. Moreover, these small trends in sample numbers would have no impact on trends in cloud-top-conditioned winds without a corresponding shift in the CMV height and longitudinal distributions within the 1-km bin, which we examine next. The relative change in the CMV height distribution within a 1-km altitude bin and how the heights and winds covary within an altitude-bin can produce confounding effects in interpreting MISR CMV trends reported in Fig. 1 as trends in speed. Extended Data Fig. 3 shows the within-bin mean CMV height trend. We see some statistically significant trends that are small, mostly in the 0 to ±40 m/decade range. If we consider a moderately large gradient in wind speed with altitude of 5 m/s/km in the free troposphere (cf. Fig. 1 mean values), then we estimate a 5/1000 m/s/m x ±40 m/decade = ±0.2 m/s/decade as an extreme influence of this effect on CMV trends. This is small relative to the wind speed trends discussed with reference to Figure 1. Moreover, the CMV heights and winds within a 1-km bin are not well correlated, with correlation coefficients < |0.2| for all bins (figure not shown). The poor correlation is as expected since (1) the uncertainty in MISR heights is only about twice as small as the bin-width and (2) the uncertainty in the MISR winds is about the same value as we would expect in wind speed changes over a 1-km depth. These two facts were the primary motivators for choosing the 1 km vertical bin width to begin with for our analyses. In addition, the spatial patterns in Extended Data Fig. 3 do not match the spatial patterns we see in the trends in Fig. 1 a,d,g. Therefore, there is no support that the large MISR CMV trends in Fig. 1 are significantly impacted by the confounding effects of changing cloud heighs and their co-variability with wind within a 1-km altitude bin. Finally, a longitudinal shift of the MISR CMV samples to a region of different large-scale circulation (e.g., a shift from the jet entrance toward the jet core) may also be a confounding factor, even if the large-scale atmospheric circulation does not have a significant trend. We examined whether there are any significant trends in the centroid of the longitudinal distributions of the CMV samples for each latitude/altitude bin, and it was found that few regions have significant trends (Extended Data Fig. 4), and where they did these regions do not completely overlap with those shown in Fig. 1. Hence, the trends shown in Fig. 1 cannot be simply attributed to longitudinal shifts in CMV samples. In summary, the confounding factors discussed above are small or cannot be used to explain the CMV changes in Figure 1 a,d,g. Therefore we cannot reject the notion that the observed MISR CMV changes are mostly attributed to changes in the cloud-top-conditioned atmospheric circulation. An ERA5 Cloud Conditional Analysis A non-random sample of the true wind field and its comparison to the same samples reported in ERA5 is sufficient to indicate uncertainty in ERA5 winds, but not a full characterization of the ERA wind uncertainty since the samples are non-random. This statement is true regardless of the conditioning (e.g. true cloud-tops only) placed on these non-random samples. These samples could be further examined to help diagnose problems within ERA5 (e.g., did ERA5 place a cloud top in the right spot?). Similarly, MISR CMVs are non-random samples conditioned to observed cloud tops. Differences between MISR CMV and ERA5_MS winds (i.e., Fig. 1) would indicate uncertainty in ERA5_MS winds in regions where differences are significantly larger than the uncertainty in MISR CMVs, as quantified in Methods. This is true regardless of the cloud-conditioned nature of MISR CMV samples. As discussed in Main, such significant differences were only observed in certain regions of the upper troposphere. As a diagnostic, the reader may be curious as to whether these ERA5_MS samples are also ERA5 samples of cloud-top. We extract the ERA5 Fraction of Cloud Cover parameter associated with each ERA_MS wind sample. In a sample-by-sample comparison, we find that 71.2% of the total ERA5_MS samples have a cloud cover > 0 at the altitude of the ERA5_MS sample; the remaining 28.8% are clear (i.e., cloud cover = 0). We also use more strict criteria for the ERA5_MS to contain a cloud-top: (1) ERA5_MS cloud cover > 0 at the altitude of the ERA5_MS sample, and (2) there are no ERA5 clouds above this altitude. Using these criteria we find that only 10.5% of the total ERA5_MS samples have a cloud top at the same altitude as the MISR CMV. This is stricter than it needs to be since MISR stereo can see through optically thin clouds to retrieve a lower cloud without any degradation in the quality of the retrieval 26 . Still, the difference between 10.5% and 71.2% is much more than can be explained by the frequency of observed thin high cloud over thicker lower cloud 45 . Regardless, when we recreated the Figure 1 ERA5_MS analysis separately using the 10.5% cloud-top, 89.5% non-cloud-top, 71.2% cloud, and 28.8% clear ERA_MS samples, we found that their differences are not statistically different (95% CI) between each other or against Figure 1 b,e,h. These results provide strong evidence that MISR CMVs can be used to evaluate ERA5 winds at the times and locations of MISR CMV sampling, regardless of whether ERA5 says there’s a cloud (or cloud-top) there or not. The results are symptomatic of a large uncertainty in the ERA5 parameterization of cloud physics, particularly in how it relates to the coupling of clouds and circulation. It’s small impact on ERA5 winds may be due to the assimilation of vast amounts of data (e.g., rawinsondes) that are independent of the presence of cloud. If so, this may make ERA5 data problematic for studying one of the key science questions in climate science, namely “How do clouds and circulation interact?” 36 Comparison to other works It is instructive to compare differences in ERA5_MS wind and MISR CMV reported here to differences in ERA winds against other observations reported in other studies. This is done to gain confidence in our analyses and those reported in other studies. In one study 46 satellite altimeter and scatterometer data were used validate ERA5 surface winds (10 m) over the Atlantic between 60°N and 60°S. Over this region, they show that ERA5 have zonal surface wind speed relative biases that vary latitudinally between 0 to 0.8 m/s. Other studies 47,48 have compared ERA5 surface winds to land surface station data, the vast majority of which were equatorward of 60° latitude. These land station comparisons indicated mean absolute difference with ERA5 surface winds < 0.4 m/s. These results are in line with ERA5_MS biases relative to MISR CMV for the lowest 1 km bin, with results varying latitudinally (and averaged over all longitudes) within the range of -0.2 m/s to + 0.8 m/s between 60°S to 60°N. If we restrict ourselves to 35°N to 60°N, where we have a dense network of land surface stations 48 , then the latitudinally varying surface wind speed relative biases in this latitude band between MISR CMV and ERA5_MS range from -0.1 m/s to +0.1 m/s. This improvement is expected given (1) the dense global network of station data that is assimilated in ERA5 over land within this latitude range, and (2) the high accuracy of the MISR CMV product. Over ocean, however, few surface stations data are assimilated into ERA5, so the larger ERA5 wind biases relative to altimeter, scatterometer, and MISR data makes sense. For winds above the surface, this study is the first validation of ERA5 tropospheric winds (cloud-top-conditioned or otherwise) over the globe based on independent observations. However, one study 49 using Aeolus 50 data over one rawinsonde station in Singapore also evaluated the ERA5 winds. Aeolus is a Doppler wind lidar, capable of deriving vertically-resolved, zonal winds (i.e., U). Using data between 2019 and 2021, they show the height-resolved, mean zonal winds measured by Aeolus is within ±1.5 m/s of ERA5 between the surface and 14 km. Above 14 km, ERA5 reaches a maximum bias relative to Aeolus of +3.5 m/s at an altitude of 16.5 km (i.e., near the tropopause). We extracted 20 years of MISR CMV U-component over Singapore and it showed very similar results, despite being cloud-conditional: within ± 1.0 m/s of ERA5_MS between 0 and 14 km, with a maximum relative bias of +3.2 m/s also at 16.5 km. The similarities are remarkable, which speaks to the very high quality of both Aeolus Doppler winds and MISR CMVs, as well as to the high quality of ERA5 winds at altitudes in the lower to middle troposphere, at least at this tropical location. That study was able to attribute the large relative bias near the tropical tropopause to the poor representation of Kelvin wave dynamics in ERA5, where reanalyses are known to struggle 51 . The positive impact that the assimilation of global Aeolus winds had on NWP model forecasts, including ECMWF 52,53 , is further evidence that modeled winds still have room for improvements, particularly in the upper troposphere (i.e., where the mean MISR CMV show the largest disagreement with ERA5_MS in Figure 1). Comparison of the time series of ERA5 winds against independent satellite data does not yet exist – the results here with MISR are a first. A time series analysis with satellite scatterometers is trickier because of the different instruments with different orbit (and orbit drifts) that need to be stitched together. In one study 54 that used a blended method with other data to help with some shortcomings in the satellite data, they show trends of surface winds over ocean between 60°N and 60°S between 1992-2012. Their results show latitudinal variability in zonal mean trends ranging between -0.2 m/s/decade to +0.2 m/s/decade. In the case of MISR CMVs, ERA5_MS and ERA5_AW, few latitude bins show statistically significant trends in the surface (0 – 1 km) bin, and where they do the trends range between -0.2 to +0.2 m/s/decade (Figure 1). This is similar to the scatterometer study, recognizing the caveat in the comparison due to differences in time periods and ocean only. Based on the above comparisons with other studies, we find similar relative biases with ERA5 winds as those reported using MISR for the very limited regions of the troposphere that these studies cover. These comparisons, along with extensive validation of MISR CMVs that show a highly accurate and stable dataset (see Methods), supports the conclusion that ERA5_MS winds are insignificantly different than MISR CMV in the lower to middle troposphere, and have small, but significant differences in the upper troposphere as described in Main. Seasonal Variability of MISR CMV & ERA5 Winds Trends Modeling studies have shown that the rates and drivers of tropical expansion have some seasonality. Hence, we have compared the decadal trends of seasonal means in height-resolved winds (Extended Data Fig. 5-8) against the trends in deseasonalized monthly anomalies (Fig. 1 in Main). There is general agreement between the two patterns of trends, except that the level of significance is reduced in seasonal trends as each seasonal plot has only a quarter of the total data used in Fig. 1. There are only two notable exceptions to this broad agreement. The first exception is in the strengthening of the U-component of the winds along the polar front in the SH seen in Fig. 1 – this feature largely disappears in boreal winter (DJF) for all three datasets. During DJF, stratospheric ozone depletion over the Antarctic regions has been attributed as a mechanism for enhanced poleward shifts in the eddy-driven jet and the SH Hadley cell edge in climate models. This enhanced poleward movement would result in a more meridional flow of wind than zonal in the polar jet and could likely explain the lack of strengthening in the U-component over these months. The second is in the presence of substantial strengthening of the U-component in the subtropical jet of the SH seen in the ERA5_MS data but not in the CMV – it is largely absent in the ERA5_MS in the boreal winter (DJF), weakened in boreal summer (JJA), and very strong in the boreal spring and fall seasons (MAM and SON, respectively). Apart from these two exceptions, the lack of strong seasonality in the trends in Fig. 1 implies that whatever is driving the trends is doing so regardless of seasonal forcing. Declarations Data availability MISR CMV data are publicly available at the NASA Langley Atmospheric Science Data Center (https://asdc.larc.nasa.gov/project/MISR/MI3MCMVN_2). ERA5 hourly data are publicly available from the European Centre for Medium-Range Weather Forecasts and Copernicus Climate Change Service Climate Data Store (https://cds.climate.copernicus.eu/). Code availability Python code used for data processing and statistical analyses will be made available upon request from the corresponding author. Acknowledgements L.D., G.Zhao, J.L. and A.M. acknowledge the support from the MISR project through the Jet Propulsion Laboratory of the California Institute of Technology (contract no. 1474871). G. Zhang is supported by the US National Science Foundation award (2327959). We thank Dr. Yulan Hong for providing the tropopause height data. We also thank Dr. Ad Stoffelen and an anonymous reviewer for their constructive comments. Author contributions L.D. conceived and designed this study. G.Zhao carried out the data analysis. L.D. and G.Zhao. drafted the manuscript with input from G.Zhang, Z.W., L.J. and A.M. Competing interests The authors declare no competing interests. References IPCC: Summary for Policymakers. In Climate Change 2021: The Physical Science Basis (eds Masson-Delmotte, V. et al.) Cambridge Univ. Press, (2021). Manney, G.L. & Hegglin, M.I. Seasonal and regional variations of long-term changes in upper-tropospheric jets from reanalyses. J. Clim . 31 , 423–448 (2018). Staten, P.W., Lu., J., Grise, K.M., Davis, S.M. & Birner, T. Re-examining tropical expansion. Nature Climate Change 8 , 768-775 (2018). Lee, S.H., Williams, P.D. & Frame, T.H.A. Increased shear in the North Atlantic upper-level jet stream over the past four decades. Nature 572 , 639-642 (2019). Gillett, Z.E., Hendon, H.H., J.M., Arblaster, J.M. & Lim, E.-P. Tropical and extratropical influences on the variability of the southern hemisphere wintertime subtropical jet. J. Clim. 34 , 4009-4022 (2021). Martin, J.E. Recent trends in the waviness of the northern hemisphere wintertime polar and subtropical jets. J. Geophys. Res. Atmos. 126 , https://doi.org/10.1029/2020JD033668 (2021). O’Connor, G.K., Steig, E.J. & Hakim, G.J. Strengthening Southern Hemisphere westerlies and Amundsen Sea Low deepening over the 20th century revealed by proxy-data assimilation. Geophysical Research Letters 48 , e2021GL095999 (2021). Grise, K.M. & Davis, S.M. Hadley cell expansion in CMIP6 models. Atmos. Chem. Phys. 20 , 5249-5268 (2020). Chemke, R. & Polvani, L.M. Opposite tropical circulation trends in climate models and in reanalyses. Nat. Geosci. 12, 528–532 https://doi.org/10.1038/s41561-019-0383-x (2019). Mueller, K., M.J. Garay, L. Di Girolamo, V. Jovanovic, and C. Moroney, 2012: MISR Cloud Motion Vector Product Algorithm Theoretical Basis , JPL D-64973, Pasadena, CA, 28 pp. [Archived online at https://eospso.nasa.gov/sites/default/files/atbd/MISR_L3_CMV_ATBD.pdf] Diner, D. J., et al. , Multi-angle Imaging Spectro- Radiometer (MISR) description and experiment overview. IEEE Trans. Geosci. Remote Sens., 36, 1072–1087 (1998). Hersbach, H., et al. The ERA5 global reanalysis. Q. J. Royal Meteorol. Soc . 146 , 1999-2049 (2020). Held, I. M., & Soden, B. J. Robust Responses of the Hydrological Cycle to Global Warming. J. Climate , 19 , 5686–5699, https://doi.org/10.1175/JCLI3990.1 (2006) Simpson, I. R., Shaw, T. A. & Seager, R. A diagnosis of the seasonally and longitudinally varying midlatitude circulation response to global warming. J. Atmos. Sci. 71, 2489–2515 (2014). Held, I. M., & Soden, B. J. Robust Responses of the Hydrological Cycle to Global Warming. J. Climate , 19 , 5686–5699, https://doi.org/10.1175/JCLI3990.1 (2006). Lorenz, D. J. & DeWeaver, E. T. Tropopause height and zonal wind response to global warming in the IPCC scenario integrations. J. Geophys. Res. 112, D10119 (2007). Staten et al. Tropical Widening: from global variations to regional impacts. Bull. American Meteorol. Soc. 101 , 897-904 (2020). Janjić, T. et al. Waller, J.A., & Weston, P. On the representation error in data assimilation, Q J R Meteorol Soc . 144, 1257– 1278. https://doi.org/10.1002/qj.3130 (2018). Morrison, H. et al. Confronting the challenge of modeling cloud and precipitation microphysics. Journal of Advances in Modeling Earth Systems , 12 , e2019MS001689. https://doi.org/10.1029/2019MS001689 (2020). Ferguson, C. R. & Villarini, G. Detecting inhomogeneities in the Twentieth Century Reanalysis over the central United States, J. Geophys. Res. , 117 , D05123, doi:10.1029/2011JD016988 (2012). Carrassi, A, Bocquet, M, Bertino, L, & Evensen, G. Data assimilation in the geosciences: An overview of methods, issues, and perspectives. WIREs Clim Change . 9, e535, https://doi.org/10.1002/wcc.535 (2018). Mueller, K. J., Moroney, C. P., Jovanovic, V., Muller, J.-P., Di Girolamo, L., & Davies, R., 2013: MISR Level 2 Cloud Product Algorithm Theoretical Basis . JPL Tech. Doc. JPL D-73327 , 51 pp. [Archived online at https://eospso.gsfc.nasa.gov/sites/default/files/atbd/MISR_L2_CLOUD_ATBD-1.pdf.] Norris, J. R., & Evan A. T. Empirical Removal of Artifacts from the ISCCP and PATMOS-x Satellite Cloud Records. J. Atmos. Oceanic Technol. , 32 , 691–702, https://doi.org/10.1175/JTECH-D-14-00058.1 (2015). Marchand, R. T., Ackerman, T. P., & Moroney, C. An assessment of Multiangle Imaging Spectroradiometer (MISR) stereo-derived cloud top heights and cloud top winds using ground-based radar, lidar, and microwave radiometers, J. Geophys. Res. , 112 , D06204, doi:10.1029/2006JD007091 (2007). Horváth, Á. Improvements to MISR stereo motion vectors. J. Geophys. Res. Atmos. , 118 , 5600–5620 (2013). Mueller, K. J. et al . Assessment of MISR cloud motion vectors (CMVs) relative to GOES and MODIS atmospheric motion vectors (AMVs). J. Applied Meteorol. Climatol. , 56 , 555–572 (2017). Mitra, A., Di Girolamo, L., Hong, Y., Zhan, Y., & Mueller, K. J. Assessment and error analysis of Terra- MODIS and MISR cloud-top heights through comparison with ISS-CATS lidar. J. Geophys. Res. Atmos. , 126 , e2020JD034281 (2021). Stewart, R.E., Szeto, K.K., Reinking, R.F., Cloudh, S.A., & Ballard, S.P. Midlatitude cyclonic cloud systems and their features affecting large scales and climate. Rev. Geophys., 36 , 245-273 (1998). Eckhardt, S. et al. A 15-year climatology of warm conveyor belts. J. Clim ., 17 , 218-237 (2004). Woollings, T., Drouard, M., O’Reilly, C.H. et al. Trends in the atmospheric jet streams are emerging in observations and could be linked to tropical warming. Commun Earth Environ ., 4 , 125 (2023). Asadieh, B. & Krakauer, N. Y. Global trends in extreme precipitation: climate models versus observations, Hydrol. Earth Syst. Sci ., 19 , 877–891 (2015). Myhre, G., Alterskjær, K., Stjern, C.W. et al. Frequency of extreme precipitation increases extensively with event rareness under global warming. Sci. Rep ., 9 , 16063 (2019) Berry, G., & Reeder, M. J. Objective identification of the intertropical convergence zone: Climatology and trends from the ERA-Interim. J.Climate , 27 , 1894-1909, (2014). Cowtan, K. & Way, R.G., Coverage bias in the HadCRUT4 temperature series and its impact on recent temperature trends. Q.J.R. Meteorol. Soc ., 140 , 1935-1944. https://doi.org/10.1002/qj.2297 (2014). Kang, S. M., D. M. W. Frierson, & Held, I. M. The Tropical Response to Extratropical Thermal Forcing in an Idealized GCM: The Importance of Radiative Feedbacks and Convective Parameterization. J. Atmos. Sci. , 66 , 2812–2827, https://doi.org/10.1175/2009JAS2924.1 (2009). Deser, C. et al. Uncertainty in climate change projections: The role of internal variability. Climate Dyn ., 38 , 527-546 (2012). Bony, S. et al. Clouds, circulation and climate sensitivity. Nature Geosci., 8, 261-268, https://doi.org/10.1038/ngeo2398 (2015) USGCRP, 2018: Impacts, Risks, and Adaptationin the United States: Fourth National Climate Assessment, Volume II [Reidmiller, D.R., C.W. Avery, D.R. Easterling, K.E. Kunkel, K.L.M. Lewis, T.K. Maycock, and B.C. Stewart (eds.)]. U.S. Global Change Research Program, Washington, DC, USA, 1515 pp. doi: 10.7930/NCA4.2018 (2018). Gelaro, R. et al. The Modern-Era Retrospective Analysis for Research and Applications, Version 2 (MERRA-2). J. Clim., 30 , 5419–5454 DOI: 10.1175/JCLI-D-16-0758.1 (2017). Hussain et al. pyMannKendall: a python package for non-parametric Mann Kendall family of trend tests. J. Open Source Software , 4 , 1556 (2019). Ventura, V., Paciorek, C. J. & Risbey, J. S. Controlling the proportion of falsely rejected hypotheses when conducting multiple tests with climatological data. J. Clim. 17 , 4343–4356 (2004). Adam, O. et al. The TropD software package (v1): standardized methods for calculating tropical-width diagnostics. Geoscientific Model Development, 11 , 4339–4357. https://doi.org/10.5194/gmd-11-4339-2018 (2018). Liu, X., Grise, K. M., Schmidt, D.F., & Davis, R. E. Regional characteristics of variability in the Northern Hemisphere wintertime polar front jet and subtropical jet in observations and CMIP6 models. J. Geophys. Res. Atmos. , 126 , e2021JD034876 (2021). Grise, K.M. et al. Regional and seasonal characteristics of the recent expansion of the tropics. J. Clim., 31 , 6839-6856 (2018). Mitra, A., 2023: Validation, trend analysis & bias reduction through satellite fusion of the cloud-top height record from Terra MODIS & MISR. Ph.D. dissertation, University of Illinois at Urbana-Champaign, Urbana, IL, 166 pp. Oreopoulos, L., Cho, N., & Lee, D., New insights about cloud vertical structure from CloudSat and CALIPSO observations. J. Geophys. Res. Atmos ., 122 , 9280-9300 (2017). Campos, R.M., Gramcianinov, C.B., de Camargo, R., & da Silva Dias, P. Assessment and calibration of ERA5 severe winds in the Atlantic ocean using satellite data. Remote Sens. , 14 , 4918. https://doi.org/10.3390/rs14194918 (2022). Ramon, J., Lledo, L., Torralba, V., Soret, A., & Doblas-Reyes, F.J. What global reanalysis best represents surface winds? Q. J. R. Meteorol. Soc ., 145 , 3236-3251 (2019). Molina, M., Gutierrez, C., & Sanchez, E. Comparison of ERA5 surface wind speed climatologies over Europe with observations from the HadISD dataset. Int. J. Climatol ., 41 , 4864-4878 (2021). Banyard, T.P., et al . Aeolus wind lidar observations of the 2019/2020 quasi-biennial oscillation disruption with comparison to radiosondes and reanalysis. Atmos. Chem. Phys ., 24 , 2465-2490 (2024). Stoffelen et al . The atmospheric dynamics mission for global wind field measurement. B. Am. Meteorol. Soc ., 86 , 73-88 (2005). Flannaghan, T.J. & Fueglistaler, S. Vertical mixing and the temperature and wind structure of the tropical tropopause layer. J. Atmos. Sci ., 71 , 1609-1622 (2014). Rennie, M.P., et al . The impact of Aeolus wind retrievals on ECMWF global weather forecasts. Q. J. R. Meteorol. Soc ., 147 , 3555-3586 (2021). Zuo, H., Stoffelen, Rennie, M. & Hasager, C.B. The contribution of Aeolus wind observations to ECMWF sea surface wind forecasts. J. Geophys. Res. Atmos ., 129 , e2023JD039555 (2024). Desbiolles, F., et al. Two decades [1992-2012] of surface wind analysis based on satellite scatterometer observations. J. Marine Sci. , 168 , 38-56 (2017). Additional Declarations There is NO Competing Interest. Supplementary Files ExtendedDataFigures.docx Cite Share Download PDF Status: Published Journal Publication published 09 Jul, 2025 Read the published version in Nature → Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5296185","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Physical Sciences - Article","associatedPublications":[],"authors":[{"id":438641639,"identity":"386bd27c-49ce-43c6-8c8e-50736b4462ce","order_by":0,"name":"Larry Di Girolamo","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAtklEQVRIiWNgGAWjYFCCxMYHCQUWPAwMjA1AHjNRWpoNEgwkSNKSwCbBYCAB4xGhRd49ua3igYGEjDl7c/MHhgrrxAZCWgzPPGy7AXKYZc/BNgmGM+lEaJmRCNFicCOxjYGx7TBxWgqgWpo/MP4jQou8BNBwqJYGCcYGIrQY8DxslgBrOQP0S8KxdGPCtrSnP/z4o8LG3uB4++MPH2qsZQnbcgCZl0BIOdgWgoaOglEwCkbBKAAAke8+YULGEvcAAAAASUVORK5CYII=","orcid":"","institution":"University of Illinois Urbana-Champaign","correspondingAuthor":true,"prefix":"","firstName":"Larry","middleName":"Di","lastName":"Girolamo","suffix":""},{"id":438641640,"identity":"dafcf2b2-4eea-4c3d-b3c0-375708605de9","order_by":1,"name":"Guangyu Zhao","email":"","orcid":"","institution":"University of Illinois at Urbana-Champaign","correspondingAuthor":false,"prefix":"","firstName":"Guangyu","middleName":"","lastName":"Zhao","suffix":""},{"id":438641641,"identity":"020f9d30-b171-4468-9ba3-13fa83c2d3ff","order_by":2,"name":"Gan Zhang","email":"","orcid":"https://orcid.org/0000-0002-7323-3409","institution":"University of Illinois at Urbana-Champaign","correspondingAuthor":false,"prefix":"","firstName":"Gan","middleName":"","lastName":"Zhang","suffix":""},{"id":438641642,"identity":"a92f9417-0900-4954-a300-c39a51bdb55f","order_by":3,"name":"Zhuo Wang","email":"","orcid":"","institution":"University of Illinois Urbana-Champaign","correspondingAuthor":false,"prefix":"","firstName":"Zhuo","middleName":"","lastName":"Wang","suffix":""},{"id":438641643,"identity":"c7a8ef6c-fcf4-4ee7-8749-913d080c04e1","order_by":4,"name":"Jesse Loveridge","email":"","orcid":"https://orcid.org/0000-0002-7127-6907","institution":"Colorado State University","correspondingAuthor":false,"prefix":"","firstName":"Jesse","middleName":"","lastName":"Loveridge","suffix":""},{"id":438641644,"identity":"f50b47e4-92cb-4005-9ac8-e6c1daa230b7","order_by":5,"name":"Arka Mitra","email":"","orcid":"","institution":"Argonne National Laboratory","correspondingAuthor":false,"prefix":"","firstName":"Arka","middleName":"","lastName":"Mitra","suffix":""}],"badges":[],"createdAt":"2024-10-19 21:50:06","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5296185/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5296185/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41586-025-09242-1","type":"published","date":"2025-07-09T04:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":80818892,"identity":"5d60ed1d-6120-46a7-8a1d-a0c461db3ff8","added_by":"auto","created_at":"2025-04-17 11:51:09","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":4737457,"visible":true,"origin":"","legend":"\u003cp\u003eThe mean and trend values of wind speed and their zonal (U; positive eastward) and meridional (V, positive northward) components for CMV \u003cstrong\u003e(a\u003c/strong\u003e,\u003cstrong\u003ed\u003c/strong\u003e,\u003cstrong\u003eg)\u003c/strong\u003e, ERA5_MS \u003cstrong\u003e(b\u003c/strong\u003e,\u003cstrong\u003ee\u003c/strong\u003e,\u003cstrong\u003eh)\u003c/strong\u003e, and ERA5_AW \u003cstrong\u003e(c\u003c/strong\u003e,\u003cstrong\u003ef\u003c/strong\u003e,\u003cstrong\u003ei) \u003c/strong\u003eduring 2000-2020. The black contour lines represent the mean values in m s\u003csup\u003e-1\u003c/sup\u003e averaged over the 21-year MISR record. The trends in m s\u003csup\u003e-1\u003c/sup\u003e decade\u003csup\u003e-1\u003c/sup\u003e are colored and use a different color scale for each variable. The trends are calculated with deseasonalized monthly CMV anomalies\u0026nbsp; that have passed the tightened False Discovery Rate (FDR) correction at the 5% level. The grey lines represent the mean tropopause heights averaged between 2000 and 2020. See Methods for details.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-5296185/v1/174d9f6ca277885aedbc2952.png"},{"id":80819808,"identity":"1ced75ab-fd47-41cf-b5a3-34aee199747a","added_by":"auto","created_at":"2025-04-17 11:59:09","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":3124388,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eThe time series of the deseasonalized monthly anomalies in the latitudinal positions of Lat_U0 for NH and SH \u003c/em\u003e\u003cem\u003e\u003cstrong\u003e(a\u003c/strong\u003e\u003c/em\u003e\u003cem\u003e, \u003c/em\u003e\u003cem\u003e\u003cstrong\u003eb)\u003c/strong\u003e\u003c/em\u003e\u003cem\u003e and Lat_U\u003c/em\u003e\u003csub\u003e\u003cem\u003emax\u003c/em\u003e\u003c/sub\u003e\u003cem\u003e for NH and SH (\u003c/em\u003e\u003cem\u003e\u003cstrong\u003ec\u003c/strong\u003e\u003c/em\u003e\u003cem\u003e, \u003c/em\u003e\u003cem\u003e\u003cstrong\u003ed)\u003c/strong\u003e\u003c/em\u003e\u003cem\u003e. The solid lines across the data series are the least square regression lines with their color matching their corresponding data series, although they are largely overlapping. The slope and its uncertainty (95% CI) for each fitting line are given in the legends, as well as its P value. The insets of \u003c/em\u003e\u003cem\u003e\u003cstrong\u003ea\u003c/strong\u003e\u003c/em\u003e\u003cem\u003e-\u003c/em\u003e\u003cem\u003e\u003cstrong\u003ed\u003c/strong\u003e\u003c/em\u003e\u003cem\u003e show the mean monthly values of the latitudinal positions averaged over the 2000-2020 period for each of the three datasets.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-5296185/v1/039789e49abeb97f9463e347.png"},{"id":87467378,"identity":"0c167aa8-4e26-4415-8213-7a056dbe140e","added_by":"auto","created_at":"2025-07-24 08:08:26","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":8256407,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5296185/v1/a0e0ea03-673a-47fc-9ab4-2755a2e706fc.pdf"},{"id":80818890,"identity":"fb9487f0-bdcb-4d24-8eeb-e4adb267d529","added_by":"auto","created_at":"2025-04-17 11:51:09","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":6760532,"visible":true,"origin":"","legend":"","description":"","filename":"ExtendedDataFigures.docx","url":"https://assets-eu.researchsquare.com/files/rs-5296185/v1/3b4d7535daf8c86ba555d02e.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Decadal changes in atmospheric circulation detected in cloud motion vectors","fulltext":[{"header":"Main","content":"\u003cp\u003eEarth\u0026rsquo;s atmospheric general circulation regulates the climate and extremes that profoundly impact human societies and ecosystems. The circulation embodies a variety of components, including the meridional overturning circulation that occupies the tropics (i.e., Hadley circulation) and the turbulence-laden westerlies that prevail the midlatitude troposphere (i.e., midlatitude jets), each having their own covarying cloud and wind features. Understanding natural variations and human-driven trends in the atmospheric general circulation has long been a focus of climate research. For example, climate models and theoretical arguments have projected a slowdown of the overturning circulation\u003csup\u003e13\u003c/sup\u003e, a poleward expansion of the Hadley circulation\u003csup\u003e3\u003c/sup\u003e, and a poleward shift of midlatitude westerlies\u003csup\u003e1,2,14,15\u003c/sup\u003e. However, the quantitative projections are highly uncertain. The Coupled Model Intercomparison Project Phase 6 (CMIP6) models, for instance, show a mean expansion rate in the Hadley cell of both hemispheres of ~0.2 \u0026deg;latitude decade\u003csup\u003e-1\u003c/sup\u003e between 1979-2008, but with a large spread of ~0.5 \u0026deg;latitude decade\u003csup\u003e-1\u003c/sup\u003e amongst the models\u003csup\u003e8\u003c/sup\u003e. Evidence for such changes exists in satellite observations and atmospheric reanalysis datasets, but with mixed results regarding the magnitude of the expansion rate \u0026ndash; some reaching as high as 3 \u0026deg;latitude/decade\u003csup\u003e16\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eThese mixed results arise partly from the diverse metrics used to describe features of the general circulation and the fundamental limitations in the datasets being used. For example, the \u0026ldquo;edge\u0026rdquo; of the Hadley cell has been defined with upper- or lower-tropospheric quantities. Definitions based on upper atmospheric quantities (e.g., outgoing longwave radiation, total ozone amount, gradients in tropopause height) are easily measurable by satellite instruments but have recently been deemed as unreliable for quantifying dynamical changes of the general ciculation\u003csup\u003e3,8\u003c/sup\u003e. Consequently, recent assessments of tropical expansion have focused on metrics from lower tropospheric data, which are considered more dynamically consistent. These metrics include the latitude of zero-crossing of the meridional stream function at 500 hPa and the latitude of zero-crossing of the near-surface zonal wind speed that divides the tropical easterlies from the midlatitude westerlies (hereafter \u003cem\u003eLat_U0\u003c/em\u003e)\u003csup\u003e3,8\u003c/sup\u003e. Such lower tropospheric metrics are easily derived from reanalysis datasets, thereby making these datasets the primary source for verifying climate model predictions\u003csup\u003e1,3,8,9\u003c/sup\u003e. However, this approach has important limitations. Both climate models and atmospheric reanalyses use dynamical models, which are subject to common uncertainties related to parameterized physics within the models\u003csup\u003e17-19\u003c/sup\u003e. Furthermore, reanalyses also assimilate observations\u003csup\u003e20\u003c/sup\u003e from a changing mix of satellite instruments and sparsely distributed surface and upper-air stations. Data discontinuities, calibration drifts, and the evolving nature of the global observing system contribute to time-varying uncertainties in the reanalysis data that are particularly problematic for trend detection. Thus, it is critical to validate atmospheric circulation changes using independent, global observations\u003csup\u003e3\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eSatellite-based cloud motion vectors (CMVs) offer a valuable dataset for evaluating changes in the atmospheric circulation. Here, we provide the first analysis of climatological means and trends in height-resolved CMVs, binned to 1 km vertical bins, collected over the past two decades by MISR\u003csup\u003e10,11\u003c/sup\u003e onboard the \u003cem\u003eTerra\u0026nbsp;\u003c/em\u003esatellite. \u0026nbsp;The MISR CMVs offer unique advantages in studying decadal trends compared to other satellite-based CMVs. Unlike other satellite Earth science records, \u003cem\u003eTerra\u003c/em\u003e maintained a stable equator-crossing time (ECT) for \u0026gt;20 years. This stability and longevity eliminates the aliasing of diurnal variability into the record of longer-term variability, and avoids artificial discontinuities from stitching data records of multiple short-lived satellites\u003csup\u003e21\u003c/sup\u003e. Critically, MISR CMVs are derived stereoscopically\u003csup\u003e10,21\u003c/sup\u003e \u0026ndash; thus they are independent of long-term drift in radiometric calibration and able to provide an extraordinary stable record for change detection (Methods). The height and speed of MISR CMVs are highly accurate and have an error budget that is self-contained (i.e., no dependence on ancillary meteorological data), fully traceable, and extensively validated\u003csup\u003e23-26\u003c/sup\u003e. Moreover, since MISR CMVs are not assimilated into any atmospheric reanalyses, they can serve as an independent validation dataset for climate model projections and reanalysis-based estimates of atmospheric circulation change.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDecadal Trends in CMVs\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eOur analysis proceeds with the understanding that CMVs are not representative of all winds at all altitudes at all times. Unlike reanalysis datasets, which do provide a continuous estimate of wind in space and time, CMVs are contingent on cloud being present, providing winds at the cloud-top-altitude of a detectable cloud layer at the time and location of MISR observations. Hence, the cloud-dependence of the CMV sampling results in a lack of sampling of clear sky conditions and conditions below cloud top. Still, clouds do occur frequently in the planetary boundary layer, even in fair weather, and only a few small boundary layer cumuli need to be present for MISR to retrieve a CMV at a resolution of 17.6x17.6 km\u003csup\u003e2\u003c/sup\u003e. Cloud tops (hence CMV samples) in the mid to upper troposphere are often associated with meso- to synoptic-scale weather disturbances, such as tropical and extratropical cyclones. The climatology of CMVs is also linked to climatic features such as the Intertropical Convergence Zone (ITCZ) and polar fronts\u003csup\u003e27,28\u003c/sup\u003e. As such, while conditionally sampled, CMVs capture a broad range of components in the atmospheric circulation.\u003c/p\u003e\n\u003cp\u003eInterpreting trends in MISR CMVs as changes in cloud-top-conditioned circulations requires careful consideration of \u0026nbsp;several potential confounding factors, as described and analyzed in Extended Data Analyses and Discussion (EDAD). Our EDAD shows that the impacts of these confounding factors on the observed trends are small, supporting the conclusion that much of the observed MISR CMV changes (Figure 1) are due to changes in the cloud-top-conditioned atmospheric circulation. Figure 1 shows the 21-year mean and statistically significant trend of the tropospheric zonal-mean CMVs. This figure is broken down by season in EDAD. The mean of the CMVs shows key climatological features in the circulation including the midlatitude westerly jets (Figure 1d) and the meridional overturning circulation (Figure 1g). The wind speed trends are largest in the midlatitude upper troposphere, where the CMVs show statistically significant increases of up to ~2 m s\u003csup\u003e-1\u003c/sup\u003e decade\u003csup\u003e-1\u003c/sup\u003e in the northern hemisphere (NH) and ~4 m s\u003csup\u003e-1\u003c/sup\u003e decade\u003csup\u003e-1\u003c/sup\u003e in the southern hemisphere (SH) (Fig. 1a). Figure 1d,g show both poleward and westerly flow components strengthening for the NH upper troposphere. But in the SH, the poleward flow component strengthens while the westerly flow component weakens. The SH zonal wind weakening between 30-50\u0026deg;S is accompanied by an increase in the mid-troposphere zonal wind (U) of ~ 0.5 \u0026ndash; 1.5 m s\u003csup\u003e-1\u003c/sup\u003e decade\u003csup\u003e-1\u003c/sup\u003e near 60\u0026deg;S. These changes near the SH polar front suggests a poleward shift of the SH westerly flow and midlatitude storm track, which is consistent with climate model projections and recent findings from atmospheric reanalyses\u003csup\u003e14,15,29\u003c/sup\u003e. Interestingly, the overall wind speed increase mainly arises from the meridional wind (V).\u0026nbsp;The strengthening of poleward flow indicated by CMV may be associated with warm conveyor belts or atmospheric rivers embedded in extratropical cyclones, which are key processes involved in midlatitude extreme precipitation. Together with the overall increase in water vapor content in a warmer atmosphere, the strengthening of poleward wind may contribute to the observed and simulated increases of extreme precipitation\u003csup\u003e15,30,31\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eIn the equatorial regions, CMV trends reveal a pattern potentially indicative of asymmetric Hadley cell changes.\u0026nbsp;The mean CMVs in the equatorial upper troposphere show easterlies (Fig. 1d) with a strong southward component (Fig. 1g), owing to the climatological ascent associated with the ITCZ being ~6\u0026deg;N\u003csup\u003e32\u003c/sup\u003e. The trends in CMVs show a moderate increase in the upper-level easterlies in the SH deep tropics. In comparison, the V-component of CMVs show stronger and more widespread trends in the deep tropics of both NH and SH, with the southward component in the upper troposphere strengthening at a rate of up to ~4 m s\u003csup\u003e-1\u003c/sup\u003e decade\u003csup\u003e-1\u003c/sup\u003e and northward component in the lower troposphere strengthening at a rate of up to 2 m s\u003csup\u003e-1\u003c/sup\u003e decade\u003csup\u003e-1\u003c/sup\u003e. The strengthening trend of the upper tropospheric southward flow is consistent across seasons (Extended Data Figs. 5-8). This pattern in the CMV trends suggests the intensification of the cross-equatorial circulation that contributes to stronger low-tropospheric convergence and upper tropospheric divergence near 6\u0026deg;N. \u0026nbsp;This would be consistent with the faster warming of the NH\u003csup\u003e33\u003c/sup\u003e and the associated circulation response required by the global energetic constraint\u003csup\u003e34\u003c/sup\u003e. However, internal climate variability\u003csup\u003e35\u003c/sup\u003e, such as the Atlantic Multi-decadal Oscillation, within our record could also be a contributing factor.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eConsistency with Atmospheric Reanalyses\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTo assess the consistency of our findings with atmospheric reanalysis data, we compared the MISR CMV climatology and trends with those from the widely used ERA5\u003csup\u003e12\u003c/sup\u003e. We sample the ERA5 winds at the times and locations (latitude, longitude and altitude) of MISR CMVs to ensure identical sampling (see Methods). \u0026nbsp;We refer to these samples as ERA5_MS (MISR Sampled). Of course, ERA5 may not be properly representing observed clouds; however, this misrepresentation was assessed in EDAD and found to have a negligible impact on our findings presented below. We also repeat the analysis using all wind data available (ERA5_AW), i.e. irrespective of the presence of cloud, in the ERA5 dataset sampled at 10:30 AM local time. ERA5_AW is a means to analyze how cloud-top-conditioned winds differ from all winds, and to better connect our analysis to other studies that use reanalysis datasets, albeit using diurnal means\u003csup\u003e1-9\u003c/sup\u003e. Where good agreement is found between ERA_MS and MISR CMV, we cannot reject the circulation and trends in ERA5_AW using MISR, thereby increasing our confidence in circulation and trends computed using ERA over the MISR record.\u003c/p\u003e\n\u003cp\u003eFigure 1 shows good agreement between the MISR CMVs and the ERA5_MS winds, both for climatological means and trends between 2000-2020. However, there are some notable differences. In the upper troposphere of the deep tropics, the climatological means of the ERA5_MS speeds are smaller than the MISR CMV by ~2 to 8 m s\u003csup\u003e-1\u003c/sup\u003e. In midlatitudes, the upper tropospheric wind speed of ERA5_MS is greater than MISR CMV by ~1 to 3 m s\u003csup\u003e-1\u003c/sup\u003e in the NH and \u0026nbsp;1 to 2 m/s smaller in the SH. \u0026nbsp;These differences are driven primarily by the V-component. Since these differences are much larger than the uncertainty in MISR CMVs (Methods and EDAD), these findings likely indicate that further improvements to ERA5 are needed to reduce these upper tropospheric biases \u0026ndash; a finding supported by other evidence (see EDAD). \u0026nbsp;The means of ERA5_MS and ERA5_AW also differ since ERA5_MS are non-random samples (equal to MISR CMV sampling). Compared to ERA5_AW, the climatological MISR CMVs and ERA5_MS shows stronger V-component and weaker U-component in the midlatitudes, corresponding to cloudy, poleward air streams embedded in cyclones.\u003c/p\u003e\n\u003cp\u003eFocusing on decadal trends, the upper tropospheric midlatitude jets show similar patterns between the independent MISR CMV and ERA5_MS datasets. They show an increase in cloud motion speed in both hemispheres, a strengthening trend of poleward flow in the midlatitude upper troposphere, especially in the SH midlatitude, and an increase in U along the polar front in the SH. But there are also important, statistically significant differences (Extended Data Fig. 1). CMV speed trends are larger by up to ~2 m s\u003csup\u003e-1\u003c/sup\u003e decade\u003csup\u003e-1\u003c/sup\u003e compared to ERA5_MS in the upper troposphere of the SH and tropics. This is true for the V components as well, except for the striking difference between ~20\u0026deg;S and 40\u0026deg;S, where the ERA5_MS V-component displays trends that are ~4 m s\u003csup\u003e-1\u003c/sup\u003e decade\u003csup\u003e-1\u003c/sup\u003e weaker than MISR. \u0026nbsp;Near this region, the ERA5_MS U-component trends are up to 2.5 m s\u003csup\u003e-1\u003c/sup\u003e decade\u003csup\u003e-1\u003c/sup\u003e larger than MISR, with ERA5_MS showing a trend of up to +2.5 m s\u003csup\u003e-1\u003c/sup\u003e decade\u003csup\u003e-1\u003c/sup\u003e just north of ~30\u0026deg;S and MISR CMV showing a trend of -2.5 m s\u003csup\u003e-1\u003c/sup\u003e decade\u003csup\u003e-1\u003c/sup\u003e just south of ~30\u0026deg;S. Since MISR CMV retrievals are agnostic to location, and since these trend differences are much larger than the uncertainty and stability in MISR CMVs (Methods), we suspect an issue with ERA5 \u0026ndash; perhaps erroneous trends in the input data assimilated into ERA5 that would impact ERA5_MS trends in the upper troposphere of the SH and tropics. \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWhen we also consider ERA5_AW, the increase in U along the SH polar front is the only common dominant change in the mid-latitudes amongst the three datasets and has been observed in other reanalysis datasets as well\u003csup\u003e2\u003c/sup\u003e. \u0026nbsp;The weak trends that appear in the ERA5_AW V-component in midlatitudes, where MISR CMV and ERA_MS show strong poleward trends, suggests a potential compensating trend in the equatorward meridional flow in clear-sky conditions. In the tropical upper troposphere, the trend agreement between ERA5_AW and ERA5_MS V-components may be due to the suspected erroneous trends in the inputs to ERA5 noted above.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eExpansion and Migration Rates\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTo assess the expansion rate of the Hadley cells, we apply the \u003cem\u003eLat_U0\u003c/em\u003e metric to all three datasets using winds in the 0-1 km altitude bin (see Methods). Figure 2a,b shows the time series of deseasonalized monthly anomalies in the latitudinal position of \u003cem\u003eLat_U0\u003c/em\u003e for NH and SH, respectively. The correlation coefficient, \u003cem\u003er\u003c/em\u003e, in the monthly anomalies amongst these datasets is strong, particularly between the MISR CMV and ERA5_MS in the NH (\u003cem\u003er\u003c/em\u003e = 0.87) and SH (\u003cem\u003er\u003c/em\u003e = 0.95). The mean monthly values of the \u003cem\u003eLat_U0\u003c/em\u003e (the insets in Fig. 2a,b) indicate agreement in the seasonal cycle of the edge of the Hadley cells amongst the three datasets, but with absolute differences of up to ~2\u0026deg; in latitude. All datasets show a poleward expansion rate of ~0.3 \u0026ndash; 0.5 \u0026plusmn; 0.2 (95% CI) \u0026deg;latitude decade\u003csup\u003e-1\u003c/sup\u003e in the NH and ~0 \u0026ndash; 0.2 \u0026plusmn; 0.1 (95% CI) \u0026deg;latitude decade\u003csup\u003e-1\u003c/sup\u003e in the SH, with only a trend in the NH appearing at a reasonable confidence level based on their low \u003cem\u003eP\u003c/em\u003e-values (see methods). These estimates of poleward expansion rate between 2000 \u0026ndash; 2020 are in line with multi-model mean values in climate model and other reanalysis model estimates for earlier periods that use the \u003cem\u003eLat_U0\u003c/em\u003e metric\u003csup\u003e3,8,16\u003c/sup\u003e, though the potential impact of internal climate variability on individual model simulations and relatively short observational record should be kept in mind.\u003c/p\u003e\n\u003cp\u003eA common lower-tropospheric metric to examine the zonal mean position of the polar front jet is the latitude of the maximum value in monthly \u0026nbsp;mean U (\u003cem\u003eLat_U\u003csub\u003emax\u003c/sub\u003e\u003c/em\u003e) at the 850 mb pressure level\u003csup\u003e5,16\u003c/sup\u003e. Here we use the 1 \u0026ndash; 2 km altitude bin, where the 850 mb pressure level typically resides (see Methods). Fig. 2c,d shows the deseasonalized monthly anomalies in the latitudinal position of \u003cem\u003eLat_U\u003csub\u003emax\u003c/sub\u003e\u003c/em\u003e. The correlation coefficient in the monthly anomalies amongst these datasets is excellent, particularly between the MISR CMV and ERA5_MS in the NH (\u003cem\u003er\u003c/em\u003e = 0.98) and SH (\u003cem\u003er\u003c/em\u003e = 0.99). The mean monthly values of \u003cem\u003eLat_U\u003csub\u003emax\u003c/sub\u003e\u003c/em\u003e (the insets of Fig. 2c,d) also indicate agreement in the seasonal cycle in the position of the polar front jet, particularly between the MISR CMV and ERA5_MS datasets. From these datasets, we see a poleward migration rate of ~0.3 \u0026ndash; 0.4 \u0026plusmn; 0.3 (95% CI) \u0026deg;latitude decade\u003csup\u003e-1\u003c/sup\u003e in the NH and ~0.3 \u0026ndash; 0.4 \u0026plusmn; 0.2 (95% CI) \u0026deg;latitude decade\u003csup\u003e-1\u003c/sup\u003e in the SH for all three datasets, with trends at only moderate to low confidence levels based on their \u003cem\u003eP\u003c/em\u003e-values. This low to moderate confidence is consistent with the confidence reported by the IPCC\u003csup\u003e1\u003c/sup\u003e.\u003c/p\u003e"},{"header":"Discussion and Conclusion","content":"\u003cp\u003eUnique to this study is the use of stable and accurate height-resolved cloud motion vectors (CMVs) from MISR to assess climatological values of cloud motion and their variability between 2000-2020 (Fig. 1). We show that CMVs have significantly (95% CI) increased in speed in the upper troposphere during this period in both NH and SH by up to 2 and 4 m s\u003csup\u003e-1\u0026nbsp;\u003c/sup\u003edecade\u003csup\u003e-1\u003c/sup\u003e, respectively. This speed increase occurs with an increase in meridional flow towards the poles in both hemispheres, but with westerly flows strengthening in the NH and weakening in the SH. This poleward increase in the meridional component of the CMV could indicate an intensification, an increase in moisture transport in the warm sector, a change in the overall structure, or a poleward shift in the tracks of extratropical cyclones. High confidence (P = 0.06) is placed in a poleward expansion of the Hadley circulation in the NH at a rate of 0.42 ± 0.22 (95% CI) °latitude/decade using the \u003cem\u003eLat_U0\u003c/em\u003e metric applied to CMVs. No significant expansion was observed in the SH. The \u003cem\u003eLat_Umax\u003c/em\u003e metric applied to CMVs suggests that the clouds associated the polar jets are migrating poleward at similar rates to each other in both hemispheres, but only with low to moderate confidence levels. When considering CMV changes at all altitudes throughout the tropics, the observations suggest weakening of the Hadley circulation in the NH and strengthening in the southern hemisphere; the degree to which cannot be quantified with CMVs as they do not represent mass stream functions that are commonly used in assessing circulation strength\u003csup\u003e9\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eWe provided the first evaluation of systematic errors in ERA5 mean winds and their trends throughout the troposphere against independent, height-resolved MISR CMVs. This evaluation is important because the suitability of reanalysis data for trend detection in atmospheric circulation, in part for validating climate model projections, has raised concerns\u003csup\u003e3\u003c/sup\u003e. To facilitate a direct comparison, the ERA5 winds have been sampled (ERA5_MS) to match the time and location of MISR CMVs. We show very good agreement between MISR CMVs and ERA5_MS winds. Nonetheless, small, but significant differences \u0026nbsp;between MISR CMV and ERA5_MS are present in the upper troposphere. These differences are much larger than the uncertainties in MISR CMVs, suggesting that the ERA5 data likely suffer systematic biases in the upper troposphere, specifically in the SH and the tropics. Our ERA5_MS cloud-conditional analysis presented in EDAD also points to potential problems in ERA5’s ability in representing clouds. However, this had little impact on our results, which we suspect may be due to ERA5 winds being strongly constrained by the assimilation of observational data (e.g., rawinsondes) that are independent of the presence of cloud. This raises concern in using ERA5 for studying one of the key science questions in climate science, namely “How do clouds and circulation interact?”\u003csup\u003e36\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003eDespite some upper-troposphere differences, we show excellent agreement between MISR CMV, ERA5_MS and ERA_AW in the Hadley cell expansion rates and the polar front migration rates for the 2000-2020 period using the \u003cem\u003eLat_U0\u003c/em\u003e and \u003cem\u003eLat_Umax\u003c/em\u003e metrics. Excellent agreement amongst these datasets is also observed in the seasonality of these metrics. \u0026nbsp;Therefore our findings support the use of MISR CMVs and ERA5 to monitor changes using the \u003cem\u003eLat_U0\u003c/em\u003e and \u003cem\u003eLat_Umax\u003c/em\u003e metrics that are favored by the community. \u0026nbsp;However, caution is still recommended when extending this finding beyond this period, particularly moving backwards in time because of the reduced capabilities in our global observing system that are assimilated into ERA5.\u003c/p\u003e\n\u003cp\u003eThe higher level of confidence in MISR wind speed trends in the upper troposphere relative to lower altitudes may suggest that signals of warming-induced circulation changes may first emerge in the upper troposphere, at least in cloudy conditions. ERA5_MS also picks up these trends, but the trends in ERA5_AW (i.e., all clear and cloudy winds) are much weaker. This may imply that there is a compensating trend in the meridional flow towards the equator in clear conditions. Together, they may suggest an intensification in the variability of tropical-extratropical transports that make the poleward transport of moisture and heat more extreme. This has important implications for human societies sensitive to precipitation and temperature extremes.\u003c/p\u003e\n\u003cp\u003eChanges to atmospheric circulations are a critical component of climate change that is already impacting modern society\u003csup\u003e37\u003c/sup\u003e. Our assessment of these changes in the 2000-2020 period using MISR CMVs provides much-needed benchmarks for reanalysis and climate model datasets. Still, while MISR is our longest climate-quality record from satellites for height-resolved CMVs, it is still short in light of internal climate variability. Therefore any discussion in our analysis of change is specific to changes over the past two decades only, which likely contain natural climate variability and human-induced changes. Here, we focused on zonal mean CMVs, ignoring the finer regional details important to understanding and quantifying regional impacts\u003csup\u003e2\u003c/sup\u003e. Given the larger regional internal variability of CMVs, a record longer than the MISR dataset would help detect critically important changes in regional circulations and should be pursued.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cem\u003eDatasets\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThis work uses the MISR Cloud Motion Vector (CMV) product (version F02_0002)\u003csup\u003e10\u0026nbsp;\u003c/sup\u003eover the period of March 2000 to December 2020. The product is not a usual gridded, monthly mean product normally used in climatological studies. Instead, the product contains a simple list of all CMV retrievals for a given month, with each retrieval tagged by latitude, longitude and time. The height-resolved CMVs are obtained through stereoscopic means by tracking the progression of features in the MISR 275-m resolution red-band imagery (380-km swath) over a 3.5-minute period between the initial 70° forward view and the nadir view, and again for the 3.5-minute period between the nadir view and 70° aft view\u003csup\u003e10\u003c/sup\u003e. The resolution of the MISR CMV product is 17.6km x 17.6km. Our analysis uses only the daytime descending node of the MISR orbit to keep local time consistent within high latitude grids. The latest version of MISR cloud top heights and cloud motion vectors have been extensively validated\u003csup\u003e24-26\u003c/sup\u003e. The near global validation of cloud top height has been validated against a space-based lidar\u003csup\u003e26\u003c/sup\u003e, showing a bias ± precision of -280m ± 370m. The precision in cloud motion speed is 3.7 m/s, with biases in U = 0.0 m/s and V = 0.3 m/s relative to static ground targets, and with biases in U and V relative to geostationary (for cloud top heights where they have moderate agreement) derived cloud motion vectors \u0026lt; ±0.5 m/s and possibly up to -1.5 m/s for the V-component, depending on the method of assessment\u003csup\u003e24,25\u003c/sup\u003e. \u0026nbsp;The stability of the product is also relevant for trend analysis. While MISR geometric telemetry needed for stereoscopic retrievals indicate no trends over the operation of the mission (Veljko Jovanovic, personal communication), we nonetheless perform here the first analysis to quantify its stability. The MISR stereographic retrievals are agnostic to the texture being observed, be it from cloud or land surfaces, receiving no prior. Therefore, we use the surface as a stable target for measuring the stability of the MISR TC_Cloud_F0_0001 product\u003csup\u003e21\u003c/sup\u003e, which is the main input to the CMV product. We analyzed cloud top height and wind retrievals from data flagged as “high-confidence near-surface”\u0026nbsp;by the Stereoscopically-Derived Cloud Mask in the TC_Cloud product, which typically indicate clear sky or the occasional near-surface cloud. Our analysis encompasses 20 years of global land data between 50°N and 50°S as in Mitra et al.\u003csup\u003e26\u003c/sup\u003e. We conducted a trend analysis on the modes (rather than mean to avoid any possible trend in near-surface clouds) of annual histograms of these retrievals. For the surface heights, the trend is small at 0.54 ± 2.5 m/decade (95% CI) per decade and insignificant (p-value = 0.94). Near-surface wind retrievals also exhibit negligible trends: the U-component shows a trend of 0.00 ± 0.01 m/s/decade (p-value = 0.94) and the V-component indicates a trend of 0.02 ± 0.05 m/s/decade (p-value = 0.51). These results confirm the long-term stability and reliability of MISR stereo measurements for climate research.\u003c/p\u003e\n\u003cp\u003eFor the reanalysis model dataset, this study uses hourly data of the fifth-generation European Centre for Medium-Range Weather Forecasts (ECMWF) Atmospheric Reanalysis (ERA5)\u003csup\u003e12\u003c/sup\u003e. We use the U and V components of ERA5 wind, as well as the geopotential, on all of the available 37 pressure levels ranging from 1000 hpa to 1 hpa. These hourly data are downloaded at a global 0.25° × 0.25° latitude-longitude grid, the highest spatial and temporal resolutions available for the ERA5 archive. While the quality of ERA5 winds is evaluated against MISR in Main, we provide additional discussion on ERA5 wind evaluation against other independent datasets below in Extended Data Analysis and Discussion, showing excellent agreement with MISR in the very limited regions that the other datasets report on.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eSampled ERA5 data for each MISR CMV data record (ERA5_MS)\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eSince the ERA5 data has a spatial-temporal resolution that is comparable to MISR CMV, we adopt a nearest-neighbor approach to sample the ERA5 U and V at the time, location, and altitude of each CMV retrieval. The time, latitude and longitude for each MISR CMV retrieval at a 17.6km x 17.6km resolution are used to find the closest hour and the nearest grid point of the ERA5 data. \u0026nbsp;For a CMV retrieval at a specific height, we locate the nearest ERA5 data point using the geopotential information at pressure levels. Specifically, geopotential heights are calculated by dividing the geopotential values by the Earth’s gravitational acceleration, given by 9.80665 m/s\u003csup\u003e2\u0026nbsp;\u003c/sup\u003e(constant). Hence, the sampled ERA5 data (ERA5_MS) have the exact same record length as the MISR CMV data. Wind speed is calculated from the U and V components for each record of CMV and ERA5_MS.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eTrend analysis\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eBefore trend analyses, the U, V and wind speed data of MISR CMV and ERA5_MS are first aggerated into monthly, 0.25° × 0.25° latitude-longitude grid boxes. The aggregated data is then sorted into 20 height bins ranging from 0 to 20 km with a bin width of 1km (with closed left side and opened right side). The mean of all the 17.6 km retrievals in each grid box and height bin is calculated and stored into an intermediate file along with the number of the retrievals for each bin. Hence, in one monthly intermediate file, U and V are stored into 720 (latitude) × 1440 (longitude) x 20 (altitude) bins. For the zonal analysis (Fig. 1), the total number of the bins is further reduced to 720 (latitude) x 20 (altitude) bins by averaging the data along the longitudinal dimension excluding the bins with no valid retrievals (e.g., due to high altitude terrain lying above say the 0-1 km altitude bin).\u0026nbsp;The zonal map of the total number of CMV retrievals is given in Extended Data Fig. 2.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo ensure a large sample size, only bins that have a total of \u0026gt; 5000 CMV retrievals over the 2000-2020 period are used in the zonal analyses. This effectively removed the low-sample \u0026nbsp;observations of the stratospheric clouds and the associated wind speed, thus keeping the focus of our discussion to the troposphere. \u0026nbsp;As a reference\u0026nbsp;for readers,\u0026nbsp;the mean tropopause heights are plotted in Figure 1.\u0026nbsp;The\u0026nbsp;mean tropopause heights were derived for the period 2000 and 2020 using the \u0026nbsp;tavgM_2d_slv_Nx product of the Modern-Era Retrospective analysis for Research and Applications, Version 2 (MERRA-2)\u003csup\u003e38\u003c/sup\u003e, as tropopause heights are not directly available in ERA5.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe deseasonalized monthly anomalies for each bin are calculated as the deviation from the monthly means averaged over the 2000-2020. Trends analysis of the deseasonalized anomalies is conducted by initially applying the nonparametric Mann-Kendall test for the trend and the nonparametric Sen’s method for the magnitude of the trend using the python package “pyMannkKendall”\u003csup\u003e39\u003c/sup\u003e. In all analyses and figures involving trend analysis, after performing local significance tests, we applied a tightened False Discovery Rate (FDR) correction, following the same procedure described by Ventura et al\u003csup\u003e40\u003c/sup\u003e. \u0026nbsp;We aimed to control the FDR at a nominal level of 5%; however, the significance threshold was adjusted based on the estimated proportion (50%) of true null hypotheses to maintain this control, as recommended by Ventura et al\u003csup\u003e40\u003c/sup\u003e. This adjustment often resulted in a higher nominal significance threshold, increasing the power to detect true effects while ensuring that the expected proportion of false discoveries remained at or below 5%. Grid points with adjusted p-values below the adjusted FDR threshold were considered statistically significant.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eCalculation of Lat_U0 and Lat_Umax\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThis study evaluates the tropical width and the position of the polar jets using the zonal averages of the U data in the CMV, ERA5_MS and ERA5_AW. We use two metrics from the Tropical-width Diagnostics (TropD) software package\u003csup\u003e41\u003c/sup\u003e. Using TropD allows for consistency with other studies. \u0026nbsp;\u003cem\u003eLat_U0\u0026nbsp;\u003c/em\u003eis the latitude at which the zonal-mean U-component of the wind in the 0-1km altitude bin equals zero after linear interpolation between two neighboring latitude bins. This marks the latitude in the subtropics where U switches sign from negative (easterly) to positive (westerly).\u0026nbsp;\u0026nbsp;It is calculated using the TropD_Metric_UAS module in TropD with default settings.\u0026nbsp;\u003cem\u003eLat_Umax\u003c/em\u003e is the latitude of maximum zonal-mean U in the 1-2km altitude bin. It is calculated using the TropD_Metric_EDJ module in TropD. We use the “peak” method (weak smoothing) with the smoothing parameter of \u003cem\u003en\u003c/em\u003e = 6 as recommended in other studies\u003csup\u003e42,43\u003c/sup\u003e. The other parameters in the module are set with default settings. \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n"},{"header":"Extended Data Analyses and Discussion","content":"\u003cp\u003e\u003cem\u003eExamination of Confounding Factors\u003c/em\u003e\u003c/p\u003e\u003cp\u003eThere are potential confounding factors at play when interpreting trends in MISR CMV as trends in speed, namely trends in the number of MISR CMV samples, their within-bin heights in the presence of within-bin vertical gradients in wind speed, and their within-bin longitudes in the presence of within-bin horizontal gradients in wind speed.\u003c/p\u003e\u003cp\u003eExtended Data Fig. 2 shows the number of CMV samples and their trends in terms of the within-bin percentage change. We see that the observed statistically significant trends in CMV samples is very small, mostly ranging from -0.6 to +0.2 %/decade.\u0026nbsp;There is a decreasing fraction of CMV samples in the upper troposphere and an increasing fraction in the lower troposphere. These should not be compared to cloud cover changes because a CMV retrieval is not sensitive to the underlying cloud fraction (i.e., whether the 17.6 km x 17.6 km area is 100% cloudy or 5% cloudy, we still get a CMV sample). Moreover, the positive trend in the lower troposphere is confounded by the decreasing trend in the upper troposphere, since less clouds above leads to more opportunity to retrieve clouds below. The decreasing trend in the upper atmosphere may be related to decreases in the frequency of occurrence of optically thin cirrus that reside near the detectability threshold of MISR stereo\u003csup\u003e26,44\u003c/sup\u003e. Note the spatial patterns in the small trends shown in Extended Data Fig. 2b do not match the spatial patterns we see in the trends in Fig. 1 a,d,g, which does not support the notion that sample trends alone can explain the trends seen in Fig. 1. Moreover, these small trends in sample numbers would have no impact on trends in cloud-top-conditioned winds without a corresponding shift in the CMV height and longitudinal distributions within the 1-km bin, which we examine next.\u003c/p\u003e\u003cp\u003eThe relative change in the CMV height distribution within a 1-km altitude bin and how the heights and winds covary within an altitude-bin can produce confounding effects in interpreting MISR CMV trends reported in Fig. 1 as trends in speed. Extended Data Fig. 3 shows the within-bin mean CMV height trend. We\u0026nbsp;see some statistically significant trends that are small, mostly in the 0 to ±40 m/decade range. If we consider a moderately large gradient in wind speed with altitude of 5 m/s/km in the free troposphere (cf. Fig. 1 mean values), then we estimate a 5/1000 m/s/m x ±40 m/decade = ±0.2 m/s/decade as an extreme influence of this effect on CMV trends. This is small relative to the wind speed trends discussed with reference to Figure 1. Moreover, the CMV heights and winds within a 1-km bin are not well correlated, with correlation coefficients \u0026lt; |0.2| for all bins (figure not shown). The poor correlation is as expected since (1) the uncertainty in MISR heights is only about twice as small as the bin-width and (2) the uncertainty in the MISR winds is about the same value as we would expect in wind speed changes over a 1-km depth. These two facts were the primary motivators for choosing the 1 km vertical bin width to begin with for our analyses. \u0026nbsp;In addition, the spatial patterns in Extended Data Fig. 3 do not match the spatial patterns we see in the trends in Fig. 1 a,d,g. Therefore, there is no support that the large MISR CMV trends in Fig. 1 are significantly impacted by the confounding effects of changing cloud heighs and their co-variability with wind within a 1-km altitude bin.\u003c/p\u003e\u003cp\u003eFinally, a longitudinal shift of the MISR CMV samples to a region of different large-scale circulation (e.g., a shift from the jet entrance toward the jet core) may also be a confounding factor, even if the large-scale atmospheric circulation does not have a significant trend. We examined whether there are any significant trends in the centroid of the longitudinal distributions of the CMV samples for each latitude/altitude bin, and it was found that few regions have significant trends (Extended Data Fig. 4), and where they did these regions do not completely overlap with\u0026nbsp;those\u0026nbsp;shown in Fig. 1. Hence, the trends shown in Fig. 1 cannot be simply attributed to longitudinal shifts in CMV samples.\u003c/p\u003e\u003cp\u003eIn summary, the confounding factors discussed above are small or cannot be used to explain the CMV changes in Figure 1 a,d,g. Therefore we cannot reject the notion that the observed MISR CMV changes are mostly attributed to changes in the cloud-top-conditioned atmospheric circulation.\u003c/p\u003e\u003cp\u003e\u003cem\u003eAn ERA5 Cloud Conditional Analysis\u003c/em\u003e\u003c/p\u003e\u003cp\u003eA non-random sample of the true wind field and its comparison to the same samples reported in ERA5 is sufficient to indicate uncertainty in ERA5 winds, but not a full characterization of the ERA wind uncertainty since the samples are non-random. This statement is true regardless of the conditioning (e.g. true cloud-tops only) placed on these non-random samples. These samples could be further examined to help diagnose problems within ERA5 (e.g., did ERA5 place a cloud top in the right spot?). Similarly, MISR CMVs are non-random samples conditioned to observed cloud tops. Differences between MISR CMV and ERA5_MS winds (i.e., Fig. 1) would indicate uncertainty in ERA5_MS winds in regions where differences are significantly larger than the uncertainty in MISR CMVs, as quantified in Methods. This is true regardless of the cloud-conditioned nature of MISR CMV samples. As discussed in Main, such significant differences were only observed in certain regions of the upper troposphere.\u003c/p\u003e\u003cp\u003eAs a diagnostic, the reader may be curious as to whether these ERA5_MS samples are also ERA5 samples of cloud-top. We extract the ERA5 Fraction of Cloud Cover parameter associated with each ERA_MS wind sample. In a sample-by-sample comparison, we find that 71.2% of the total ERA5_MS samples have a cloud cover \u0026gt; 0 at the altitude of the ERA5_MS sample; the remaining 28.8% are clear (i.e., cloud cover = 0). We also use more strict criteria for the ERA5_MS to contain a cloud-top: (1) ERA5_MS cloud cover \u0026gt; 0 at the altitude of the ERA5_MS sample, and (2) there are no ERA5 clouds above this altitude. Using these criteria we find that only 10.5% of the total ERA5_MS samples have a cloud top at the same altitude as the MISR CMV. This is stricter than it needs to be since MISR stereo can see through optically thin clouds to retrieve a lower cloud without any degradation in the quality of the retrieval\u003csup\u003e26\u003c/sup\u003e. Still, the difference between 10.5% and 71.2% is much more than can be explained by the frequency of observed thin high cloud over thicker lower cloud\u003csup\u003e45\u003c/sup\u003e. Regardless, when we recreated the Figure 1 ERA5_MS analysis separately using the 10.5% cloud-top, 89.5% non-cloud-top, 71.2% cloud, and 28.8% clear ERA_MS samples, we found that their differences are not statistically different (95% CI) between each other or against Figure 1 b,e,h.\u003c/p\u003e\u003cp\u003eThese results provide strong evidence that MISR CMVs can be used to evaluate ERA5 winds at the times and locations of MISR CMV sampling, regardless of whether ERA5 says there’s a cloud (or cloud-top) there or not. The results are symptomatic of a large uncertainty in the ERA5 parameterization of cloud physics, particularly in how it relates to the coupling of clouds and circulation. It’s small impact on ERA5 winds may be due to the assimilation of vast amounts of data (e.g., rawinsondes) that are independent of the presence of cloud. If so, this may make ERA5 data problematic for studying one of the key science questions in climate science, namely “How do clouds and circulation interact?”\u003csup\u003e36\u003c/sup\u003e\u0026nbsp;\u003c/p\u003e\u003cp\u003e\u003cem\u003eComparison to other works\u003c/em\u003e\u003c/p\u003e\u003cp\u003eIt is instructive to compare differences in ERA5_MS wind and MISR CMV reported here to differences in ERA winds against other observations reported in other studies. This is done to gain confidence in our analyses and those reported in other studies.\u003c/p\u003e\u003cp\u003eIn one study\u003csup\u003e46\u003c/sup\u003e satellite altimeter and scatterometer data were used validate ERA5 surface winds (10 m) over the Atlantic between 60°N and 60°S. \u0026nbsp;Over this region, they show that ERA5 have zonal surface wind speed relative biases that vary latitudinally between 0 to 0.8 m/s. \u0026nbsp; Other studies\u003csup\u003e47,48\u003c/sup\u003e have compared ERA5 surface winds to land surface station data, the vast majority of which were equatorward of 60° latitude. These land station comparisons indicated mean absolute difference with ERA5 surface winds \u0026lt; 0.4 m/s. These results are in line with ERA5_MS biases relative to MISR CMV for the lowest 1 km bin, with results varying latitudinally (and averaged over all longitudes) within the range of -0.2 m/s to + 0.8 m/s between 60°S to 60°N. If we restrict ourselves to 35°N to 60°N, where we have a dense network of land surface stations\u003csup\u003e48\u003c/sup\u003e, then the latitudinally varying surface wind speed relative biases in this latitude band between MISR CMV and ERA5_MS range from -0.1 m/s to +0.1 m/s. This improvement is expected given (1) the dense global network of station data that is assimilated in ERA5 over land within this latitude range, and (2) the high accuracy of the MISR CMV product. Over ocean, however, few surface stations data are assimilated into ERA5, so the larger ERA5 wind biases relative to altimeter, scatterometer, and MISR data makes sense.\u003c/p\u003e\u003cp\u003eFor winds above the surface, this study is the first validation of ERA5 tropospheric winds (cloud-top-conditioned or otherwise) over the globe based on independent observations. However, one study\u003csup\u003e49\u003c/sup\u003e using Aeolus\u003csup\u003e50\u003c/sup\u003e data over one rawinsonde station in Singapore also evaluated the ERA5 winds. Aeolus is a Doppler wind lidar, capable of deriving vertically-resolved, zonal winds (i.e., U). Using data between 2019 and 2021, they show the height-resolved, mean zonal winds measured by Aeolus is within ±1.5 m/s of ERA5 between the surface and 14 km. Above 14 km, ERA5 reaches a maximum bias relative to Aeolus of +3.5 m/s at an altitude of 16.5 km (i.e., near the tropopause). \u0026nbsp; We extracted 20 years of MISR CMV U-component over Singapore and it showed very similar results, despite being cloud-conditional: within ± 1.0 m/s of ERA5_MS \u0026nbsp;between 0 and 14 km, with a maximum relative bias of +3.2 m/s also at 16.5 km. The similarities are remarkable, which speaks to the very high quality of both Aeolus Doppler winds and MISR CMVs, as well as to the high quality of ERA5 winds at altitudes in the lower to middle troposphere, at least at this tropical location. That study was able to attribute the large relative bias near the tropical tropopause to the poor representation of Kelvin wave dynamics in ERA5, where reanalyses are known to struggle\u003csup\u003e51\u003c/sup\u003e. The positive impact that the assimilation of global Aeolus winds had on NWP model forecasts, including ECMWF\u003csup\u003e52,53\u003c/sup\u003e, is further evidence that modeled winds still have room for improvements, particularly in the upper troposphere (i.e., where the mean MISR CMV show the largest disagreement with ERA5_MS in Figure 1). \u0026nbsp;\u003c/p\u003e\u003cp\u003eComparison of the time series of ERA5 winds against independent satellite data does not yet exist – the results here with MISR are a first. \u0026nbsp;A time series analysis with satellite scatterometers is trickier because of the different instruments with different orbit (and orbit drifts) that need to be stitched together. In one study\u003csup\u003e54\u003c/sup\u003e that used a blended method with other data to help with some shortcomings in the satellite data, they show trends of surface winds over ocean between 60°N and 60°S between 1992-2012. Their results show latitudinal variability in zonal mean trends ranging between -0.2 m/s/decade to +0.2 m/s/decade. In the case of MISR CMVs, ERA5_MS and ERA5_AW, few latitude bins show statistically significant trends in the surface (0 – 1 km) bin, and where they do the trends range between -0.2 to +0.2 m/s/decade (Figure 1). This is similar to the scatterometer study, recognizing the caveat in the comparison due to differences in time periods and ocean only.\u003c/p\u003e\u003cp\u003eBased on the above comparisons with other studies, we find similar relative biases with ERA5 winds as those reported using MISR for the very limited regions of the troposphere that these studies cover. These comparisons, along with extensive validation of MISR CMVs that show a highly accurate and stable dataset (see Methods), supports the conclusion that ERA5_MS winds are insignificantly different than MISR CMV in the lower to middle troposphere, and have small, but significant differences in the upper troposphere as described in Main. \u0026nbsp;\u0026nbsp;\u003c/p\u003e\u003cp\u003e\u003cem\u003eSeasonal Variability of MISR CMV \u0026amp; ERA5 Winds Trends\u003c/em\u003e\u003c/p\u003e\u003cp\u003eModeling studies have shown that the rates and drivers of tropical expansion have some seasonality. Hence, we have compared the decadal trends of seasonal means in height-resolved winds (Extended Data Fig. 5-8) against the trends in deseasonalized monthly anomalies (Fig. 1 in Main). There is general agreement between the two patterns of trends, except that the level of significance is reduced in seasonal trends as each seasonal plot has only a quarter of the total data used in Fig. 1. There are only two notable exceptions to this broad agreement.\u003c/p\u003e\u003cp\u003eThe first exception is in the strengthening of the U-component of the winds along the polar front in the SH seen in Fig. 1 – this feature largely disappears in boreal winter (DJF) for all three datasets. During DJF, stratospheric ozone depletion over the Antarctic regions has been attributed as a mechanism for enhanced poleward shifts in the eddy-driven jet and the SH Hadley cell edge in climate models. This enhanced poleward movement would result in a more meridional flow of wind than zonal in the polar jet and could likely explain the lack of strengthening in the U-component over these months.\u003c/p\u003e\u003cp\u003eThe second is in the presence of substantial strengthening of the U-component in the subtropical jet of the SH seen in the ERA5_MS data but not in the CMV – it is largely absent in the ERA5_MS in the boreal winter (DJF), weakened in boreal summer (JJA), and very strong in the boreal spring and fall seasons (MAM and SON, respectively). Apart from these two exceptions, the lack of strong seasonality in the trends in Fig. 1 implies that whatever is driving the trends is doing so regardless of seasonal forcing.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMISR CMV data are publicly available at the NASA Langley Atmospheric Science Data Center (https://asdc.larc.nasa.gov/project/MISR/MI3MCMVN_2). ERA5 hourly data are publicly available from the European Centre for Medium-Range Weather Forecasts and Copernicus Climate Change Service Climate Data Store (https://cds.climate.copernicus.eu/). \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCode availability\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePython code used for data processing and statistical analyses will be made available upon request from the corresponding author.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eL.D., G.Zhao, J.L. and A.M. acknowledge the support from the MISR project through the Jet Propulsion Laboratory of the California Institute of Technology (contract no. 1474871). G. Zhang is supported by the US National Science Foundation award (2327959). We thank Dr. Yulan Hong for providing the tropopause height data. We also thank Dr. Ad Stoffelen and an anonymous reviewer for their constructive comments.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eL.D. conceived and designed this study. G.Zhao carried out the data analysis. L.D. and G.Zhao. drafted the manuscript with input from G.Zhang, Z.W., L.J. and A.M.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eIPCC: Summary for Policymakers. In Climate Change 2021: The Physical Science Basis (eds Masson-Delmotte, V. et al.) Cambridge Univ. Press, (2021).\u003c/li\u003e\n \u003cli\u003eManney, G.L. \u0026amp; Hegglin, M.I. Seasonal and regional variations of long-term changes in upper-tropospheric jets from reanalyses. \u003cem\u003eJ. Clim\u003c/em\u003e. \u003cstrong\u003e31\u003c/strong\u003e, 423\u0026ndash;448 (2018).\u003c/li\u003e\n \u003cli\u003eStaten, P.W., Lu., J., Grise, K.M., Davis, S.M. \u0026amp; Birner, T. Re-examining tropical expansion. \u003cem\u003eNature Climate Change\u003c/em\u003e\u003cstrong\u003e8\u003c/strong\u003e, 768-775 (2018).\u003c/li\u003e\n \u003cli\u003eLee, S.H., Williams, P.D. \u0026amp; Frame, T.H.A. Increased shear in the North Atlantic upper-level jet stream over the past four decades. \u003cem\u003eNature\u003c/em\u003e\u003cstrong\u003e572\u003c/strong\u003e, 639-642 (2019).\u003c/li\u003e\n \u003cli\u003eGillett, Z.E., Hendon, H.H., J.M., Arblaster, J.M. \u0026amp; Lim, E.-P. Tropical and extratropical influences on the variability of the southern hemisphere wintertime subtropical jet. \u003cem\u003eJ. Clim.\u003c/em\u003e\u003cstrong\u003e34\u003c/strong\u003e, 4009-4022 (2021).\u003c/li\u003e\n \u003cli\u003eMartin, J.E. Recent trends in the waviness of the northern hemisphere wintertime polar and subtropical jets. \u003cem\u003eJ. Geophys. Res. Atmos.\u003c/em\u003e\u003cstrong\u003e126\u003c/strong\u003e, https://doi.org/10.1029/2020JD033668 (2021).\u003c/li\u003e\n \u003cli\u003eO\u0026rsquo;Connor, G.K., Steig, E.J. \u0026amp; Hakim, G.J. Strengthening Southern Hemisphere westerlies and Amundsen Sea Low deepening over the 20th century revealed by proxy-data assimilation. \u003cem\u003eGeophysical Research Letters\u003c/em\u003e\u003cstrong\u003e48\u003c/strong\u003e, e2021GL095999 (2021).\u003c/li\u003e\n \u003cli\u003eGrise, K.M. \u0026amp; Davis, S.M. Hadley cell expansion in CMIP6 models. \u003cem\u003eAtmos. Chem. Phys.\u003c/em\u003e\u003cstrong\u003e20\u003c/strong\u003e, 5249-5268 (2020).\u003c/li\u003e\n \u003cli\u003eChemke, R. \u0026amp; Polvani, L.M. Opposite tropical circulation trends in climate models and in reanalyses. \u003cem\u003eNat. Geosci.\u003c/em\u003e 12, 528\u0026ndash;532 https://doi.org/10.1038/s41561-019-0383-x (2019).\u003c/li\u003e\n \u003cli\u003eMueller, K., M.J. Garay, L. Di Girolamo, V. Jovanovic, and C. Moroney, 2012: \u003cem\u003eMISR Cloud Motion Vector Product Algorithm Theoretical Basis\u003c/em\u003e, JPL D-64973, Pasadena, CA, 28 pp. [Archived online at https://eospso.nasa.gov/sites/default/files/atbd/MISR_L3_CMV_ATBD.pdf]\u003c/li\u003e\n \u003cli\u003eDiner, D. J., et al. , Multi-angle Imaging Spectro- Radiometer (MISR) description and experiment overview. \u003cem\u003eIEEE Trans. Geosci. Remote Sens.,\u0026nbsp;\u003c/em\u003e\u003cstrong\u003e36,\u0026nbsp;\u003c/strong\u003e1072\u0026ndash;1087 (1998).\u003c/li\u003e\n \u003cli\u003eHersbach, H., et al. The ERA5 global reanalysis. \u003cem\u003eQ. J. Royal Meteorol. Soc\u003c/em\u003e. \u003cstrong\u003e146\u003c/strong\u003e, 1999-2049 (2020).\u003c/li\u003e\n \u003cli\u003eHeld, I. M., \u0026amp; Soden, B. J. Robust Responses of the Hydrological Cycle to Global Warming. \u003cem\u003eJ. Climate\u003c/em\u003e, \u003cstrong\u003e19\u003c/strong\u003e, 5686\u0026ndash;5699, https://doi.org/10.1175/JCLI3990.1 (2006)\u003c/li\u003e\n \u003cli\u003eSimpson, I. R., Shaw, T. A. \u0026amp; Seager, R. A diagnosis of the seasonally and longitudinally varying midlatitude circulation response to global warming. J. Atmos. Sci. 71, 2489\u0026ndash;2515 (2014). Held, I. M., \u0026amp; Soden, B. J. Robust Responses of the Hydrological Cycle to Global Warming. \u003cem\u003eJ. Climate\u003c/em\u003e, \u003cstrong\u003e19\u003c/strong\u003e, 5686\u0026ndash;5699, https://doi.org/10.1175/JCLI3990.1 (2006).\u003c/li\u003e\n \u003cli\u003eLorenz, D. J. \u0026amp; DeWeaver, E. T. Tropopause height and zonal wind response to global warming in the IPCC scenario integrations. J. Geophys. Res. 112, D10119 (2007).\u003c/li\u003e\n \u003cli\u003eStaten et al. Tropical Widening: from global variations to regional impacts. \u003cem\u003eBull. American Meteorol. Soc.\u003c/em\u003e\u003cstrong\u003e101\u003c/strong\u003e, 897-904 (2020).\u003c/li\u003e\n \u003cli\u003eJanjić, T. et al. Waller, J.A., \u0026amp; Weston, P. On the representation error in data assimilation, \u003cem\u003eQ J R Meteorol Soc\u003c/em\u003e. \u003cstrong\u003e144,\u003c/strong\u003e 1257\u0026ndash; 1278. https://doi.org/10.1002/qj.3130 (2018).\u003c/li\u003e\n \u003cli\u003eMorrison, H. et al. Confronting the challenge of modeling cloud and precipitation microphysics. \u003cem\u003eJournal of Advances in Modeling Earth Systems\u003c/em\u003e, \u003cstrong\u003e12\u003c/strong\u003e, e2019MS001689. https://doi.org/10.1029/2019MS001689 (2020).\u003c/li\u003e\n \u003cli\u003eFerguson, C. R. \u0026amp; Villarini, G. Detecting inhomogeneities in the Twentieth Century Reanalysis over the central United States, \u003cem\u003eJ. Geophys. Res.\u003c/em\u003e, \u003cstrong\u003e117\u003c/strong\u003e, D05123, doi:10.1029/2011JD016988 (2012).\u003c/li\u003e\n \u003cli\u003eCarrassi, A, Bocquet, M, Bertino, L, \u0026amp; Evensen, G. Data assimilation in the geosciences: An overview of methods, issues, and perspectives. \u003cem\u003eWIREs Clim Change\u003c/em\u003e. \u003cstrong\u003e9,\u0026nbsp;\u003c/strong\u003ee535, https://doi.org/10.1002/wcc.535 (2018).\u003c/li\u003e\n \u003cli\u003eMueller, K. J., Moroney, C. P., Jovanovic, V., Muller, J.-P., Di Girolamo, L., \u0026amp; Davies, R., 2013: \u003cem\u003eMISR Level 2 Cloud Product Algorithm Theoretical Basis\u003c/em\u003e. JPL Tech. Doc. \u003cstrong\u003eJPL D-73327\u003c/strong\u003e, 51 pp. [Archived online at https://eospso.gsfc.nasa.gov/sites/default/files/atbd/MISR_L2_CLOUD_ATBD-1.pdf.]\u003c/li\u003e\n \u003cli\u003eNorris, J. R., \u0026amp; Evan A. T. Empirical Removal of Artifacts from the ISCCP and PATMOS-x Satellite Cloud Records. \u003cem\u003eJ. Atmos. Oceanic Technol.\u003c/em\u003e, \u003cstrong\u003e32\u003c/strong\u003e, 691\u0026ndash;702, https://doi.org/10.1175/JTECH-D-14-00058.1 (2015).\u003c/li\u003e\n \u003cli\u003eMarchand, R. T., Ackerman, T. P., \u0026amp; Moroney, C. An assessment of Multiangle Imaging Spectroradiometer (MISR) stereo-derived cloud top heights and cloud top winds using ground-based radar, lidar, and microwave radiometers, \u003cem\u003eJ. Geophys. Res.\u003c/em\u003e, \u003cstrong\u003e112\u003c/strong\u003e, D06204, doi:10.1029/2006JD007091 (2007).\u003c/li\u003e\n \u003cli\u003eHorváth, Á. Improvements to MISR stereo motion vectors. \u003cem\u003eJ. Geophys. Res. Atmos.\u003c/em\u003e, \u003cstrong\u003e118\u003c/strong\u003e, 5600\u0026ndash;5620 (2013).\u003c/li\u003e\n \u003cli\u003eMueller, K. J. \u003cem\u003eet al\u003c/em\u003e. Assessment of MISR cloud motion vectors (CMVs) relative to GOES and MODIS atmospheric motion vectors (AMVs). \u003cem\u003eJ. Applied Meteorol. Climatol.\u003c/em\u003e, \u003cstrong\u003e56\u003c/strong\u003e, 555\u0026ndash;572 (2017).\u003c/li\u003e\n \u003cli\u003eMitra, A., Di Girolamo, L., Hong, Y., Zhan, Y., \u0026amp; Mueller, K. J. Assessment and error analysis of Terra- MODIS and MISR cloud-top heights through comparison with ISS-CATS lidar. \u003cem\u003eJ. Geophys. Res. Atmos.\u003c/em\u003e, \u003cstrong\u003e126\u003c/strong\u003e, e2020JD034281 (2021).\u003c/li\u003e\n \u003cli\u003eStewart, R.E., Szeto, K.K., Reinking, R.F., Cloudh, S.A., \u0026amp; Ballard, S.P. Midlatitude cyclonic cloud systems and their features affecting large scales and climate. \u003cem\u003eRev. Geophys.,\u003c/em\u003e\u003cstrong\u003e36\u003c/strong\u003e, 245-273 (1998).\u003c/li\u003e\n \u003cli\u003eEckhardt, S. et al. A 15-year climatology of warm conveyor belts. \u003cem\u003eJ. Clim\u003c/em\u003e., \u003cstrong\u003e17\u003c/strong\u003e, 218-237 (2004).\u003c/li\u003e\n \u003cli\u003eWoollings, T., Drouard, M., O\u0026rsquo;Reilly, C.H. et al. Trends in the atmospheric jet streams are emerging in observations and could be linked to tropical warming. \u003cem\u003eCommun Earth Environ\u003c/em\u003e., \u003cstrong\u003e4\u003c/strong\u003e, 125 (2023).\u003c/li\u003e\n \u003cli\u003eAsadieh, B. \u0026amp; Krakauer, N. Y. Global trends in extreme precipitation: climate models versus observations, \u003cem\u003eHydrol. Earth Syst. Sci\u003c/em\u003e., \u003cstrong\u003e19\u003c/strong\u003e, 877\u0026ndash;891 (2015).\u003c/li\u003e\n \u003cli\u003eMyhre, G., Alterskj\u0026aelig;r, K., Stjern, C.W. et al. Frequency of extreme precipitation increases extensively with event rareness under global warming. \u003cem\u003eSci. Rep\u003c/em\u003e., \u003cstrong\u003e9\u003c/strong\u003e, 16063 (2019)\u003c/li\u003e\n \u003cli\u003eBerry, G., \u0026amp; Reeder, M. J. Objective identification of the intertropical convergence zone: Climatology and trends from the ERA-Interim. \u003cem\u003eJ.Climate\u003c/em\u003e, \u003cstrong\u003e27\u003c/strong\u003e, 1894-1909, (2014).\u003c/li\u003e\n \u003cli\u003eCowtan, K. \u0026amp; Way, R.G., Coverage bias in the HadCRUT4 temperature series and its impact on recent temperature trends. \u003cem\u003eQ.J.R. Meteorol. Soc\u003c/em\u003e., \u003cstrong\u003e140\u003c/strong\u003e, 1935-1944. https://doi.org/10.1002/qj.2297 (2014).\u003c/li\u003e\n \u003cli\u003eKang, S. M., D. M. W. Frierson, \u0026amp; Held, I. M. The Tropical Response to Extratropical Thermal Forcing in an Idealized GCM: The Importance of Radiative Feedbacks and Convective Parameterization. \u003cem\u003eJ. Atmos. Sci.\u003c/em\u003e, \u003cstrong\u003e66\u003c/strong\u003e, 2812\u0026ndash;2827, https://doi.org/10.1175/2009JAS2924.1 (2009).\u003c/li\u003e\n \u003cli\u003eDeser, C. et al. Uncertainty in climate change projections: The role of internal variability. \u003cem\u003eClimate Dyn\u003c/em\u003e., \u003cstrong\u003e38\u003c/strong\u003e, 527-546 (2012).\u003c/li\u003e\n \u003cli\u003eBony, S. et al. Clouds, circulation and climate sensitivity. Nature Geosci., 8, 261-268, https://doi.org/10.1038/ngeo2398 (2015)\u003c/li\u003e\n \u003cli\u003eUSGCRP, 2018: \u003cem\u003eImpacts, Risks, and Adaptationin the United States: Fourth National Climate Assessment, Volume II\u003c/em\u003e [Reidmiller, D.R., C.W. Avery, D.R. Easterling, K.E. Kunkel, K.L.M. Lewis, T.K. Maycock, and B.C. Stewart (eds.)]. U.S. Global Change Research Program, Washington, DC, USA, 1515 pp. doi: 10.7930/NCA4.2018 (2018).\u003c/li\u003e\n \u003cli\u003eGelaro, R. et al. The Modern-Era Retrospective Analysis for Research and Applications, Version 2 (MERRA-2). \u003cem\u003eJ. Clim.,\u003c/em\u003e\u003cstrong\u003e30\u003c/strong\u003e, 5419\u0026ndash;5454 DOI: 10.1175/JCLI-D-16-0758.1 (2017).\u003c/li\u003e\n \u003cli\u003eHussain et al. pyMannKendall: a python package for non-parametric Mann Kendall family of trend tests. \u003cem\u003eJ. Open Source Software\u003c/em\u003e, \u003cstrong\u003e4\u003c/strong\u003e, 1556 (2019).\u003c/li\u003e\n \u003cli\u003eVentura, V., Paciorek, C. J. \u0026amp; Risbey, J. S. Controlling the proportion of falsely rejected hypotheses when conducting multiple tests with climatological data. \u003cem\u003eJ. Clim.\u003c/em\u003e\u003cstrong\u003e17\u003c/strong\u003e, 4343\u0026ndash;4356 (2004).\u003c/li\u003e\n \u003cli\u003eAdam, O. et al. The TropD software package (v1): standardized methods for calculating tropical-width diagnostics. Geoscientific Model Development, \u003cstrong\u003e11\u003c/strong\u003e, 4339\u0026ndash;4357. https://doi.org/10.5194/gmd-11-4339-2018 (2018).\u003c/li\u003e\n \u003cli\u003eLiu, X., Grise, K. M., Schmidt, D.F., \u0026amp; Davis, R. E. Regional characteristics of variability in the Northern Hemisphere wintertime polar front jet and subtropical jet in observations and CMIP6 models. \u003cem\u003eJ. Geophys. Res. Atmos.\u003c/em\u003e, \u003cstrong\u003e126\u003c/strong\u003e, e2021JD034876 (2021).\u003c/li\u003e\n \u003cli\u003eGrise, K.M. et al. Regional and seasonal characteristics of the recent expansion of the tropics. \u003cem\u003eJ. Clim.,\u003c/em\u003e\u003cstrong\u003e31\u003c/strong\u003e, 6839-6856 (2018).\u003c/li\u003e\n \u003cli\u003eMitra, A., 2023: \u003cem\u003eValidation, trend analysis \u0026amp; bias reduction through satellite fusion of the cloud-top height record from Terra MODIS \u0026amp; MISR.\u003c/em\u003e Ph.D. dissertation, University of Illinois at Urbana-Champaign, Urbana, IL, 166 pp.\u003c/li\u003e\n \u003cli\u003eOreopoulos, L., Cho, N., \u0026amp; Lee, D., New insights about cloud vertical structure from CloudSat and CALIPSO observations. \u003cem\u003eJ. Geophys. Res. Atmos\u003c/em\u003e., \u003cstrong\u003e122\u003c/strong\u003e, 9280-9300 (2017).\u003c/li\u003e\n \u003cli\u003eCampos, R.M., Gramcianinov, C.B., de Camargo, R., \u0026amp; da Silva Dias, P. Assessment and calibration of ERA5 severe winds in the Atlantic ocean using satellite data. \u003cem\u003eRemote Sens.\u003c/em\u003e, \u003cstrong\u003e14\u003c/strong\u003e, 4918. https://doi.org/10.3390/rs14194918 (2022).\u003c/li\u003e\n \u003cli\u003eRamon, J., Lledo, L., Torralba, V., Soret, A., \u0026amp; Doblas-Reyes, F.J. What global reanalysis best represents surface winds? \u003cem\u003eQ. J. R. Meteorol. Soc\u003c/em\u003e., \u003cstrong\u003e145\u003c/strong\u003e, 3236-3251 (2019).\u003c/li\u003e\n \u003cli\u003eMolina, M., Gutierrez, C., \u0026amp; Sanchez, E. Comparison of ERA5 surface wind speed climatologies over Europe with observations from the HadISD dataset. \u003cem\u003eInt. J. Climatol\u003c/em\u003e., \u003cstrong\u003e41\u003c/strong\u003e, 4864-4878 (2021).\u003c/li\u003e\n \u003cli\u003eBanyard, T.P., \u003cem\u003eet al\u003c/em\u003e. Aeolus wind lidar observations of the 2019/2020 quasi-biennial oscillation disruption with comparison to radiosondes and reanalysis. \u003cem\u003eAtmos. Chem. Phys\u003c/em\u003e., \u003cstrong\u003e24\u003c/strong\u003e, 2465-2490 (2024).\u003c/li\u003e\n \u003cli\u003eStoffelen \u003cem\u003eet al\u003c/em\u003e. The atmospheric dynamics mission for global wind field measurement. \u003cem\u003eB. Am. Meteorol. Soc\u003c/em\u003e., \u003cstrong\u003e86\u003c/strong\u003e, 73-88 (2005).\u003c/li\u003e\n \u003cli\u003eFlannaghan, T.J. \u0026amp; Fueglistaler, S. Vertical mixing and the temperature and wind structure of the tropical tropopause layer. \u003cem\u003eJ. Atmos. Sci\u003c/em\u003e., \u003cstrong\u003e71\u003c/strong\u003e, 1609-1622 (2014).\u003c/li\u003e\n \u003cli\u003eRennie, M.P., \u003cem\u003eet al\u003c/em\u003e. The impact of Aeolus wind retrievals on ECMWF global weather forecasts. \u003cem\u003eQ. J. R. Meteorol. Soc\u003c/em\u003e., \u003cstrong\u003e147\u003c/strong\u003e, 3555-3586 (2021).\u003c/li\u003e\n \u003cli\u003eZuo, H., Stoffelen, Rennie, M. \u0026amp; Hasager, C.B. The contribution of Aeolus wind observations to ECMWF sea surface wind forecasts. \u003cem\u003eJ. Geophys. Res. Atmos\u003c/em\u003e., \u003cstrong\u003e129\u003c/strong\u003e, e2023JD039555 (2024).\u003c/li\u003e\n \u003cli\u003eDesbiolles, F., et al. Two decades [1992-2012] of surface wind analysis based on satellite scatterometer observations. \u003cem\u003eJ. Marine Sci.\u003c/em\u003e, \u003cstrong\u003e168\u003c/strong\u003e, 38-56 (2017).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-5296185/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5296185/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eChanging atmospheric circulations shift global weather patterns and their extremes, with profound effects on human societies and ecosystems. Studies using atmospheric reanalysis and climate model data\u003csup\u003e1-9\u003c/sup\u003e indicate a variety of circulation changes in recent decades but show discrepancies in magnitude and even direction. Therefore, validation with independent, climate-quality measurements is urgently needed\u003csup\u003e3\u003c/sup\u003e. Here we use the satellite-observed, height-resolved cloud motion vectors from the Multi-angle Imaging SpectroRadiometer (MISR)\u003csup\u003e10,11\u003c/sup\u003e to analyze tropospheric circulation changes during 2000–2020. We find significant changes in tropospheric circulations, with upper tropospheric cloud motion speeds in midlatitudes increasing by up to ~4 m s⁻¹ decade⁻¹, primarily due to the strengthening of meridional flow that could indicate increased poleward trajectories or intensification of extratropical cyclones. Furthermore, the northern and southern hemisphere tropics shifted poleward at a rate of 0.42±0.22 and 0.02±0.14 °latitude decade⁻¹ (95% CI), respectively, while the corresponding polar front shifted at 0.37±0.31 and 0.31±0.21 °latitude decade⁻¹. Comparison with the widely used ERA5\u003csup\u003e12\u003c/sup\u003e reanalysis winds subsampled to MISR show good agreement with MISR’s climatological values and trends but indicate likely ERA5 biases in the upper troposphere. These MISR-based observations provide critical benchmarks for refining reanalysis and climate models to advance our understanding of climate change impacts on cloud and atmospheric circulations.\u003c/p\u003e","manuscriptTitle":"Decadal changes in atmospheric circulation detected in cloud motion vectors","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-17 11:51:05","doi":"10.21203/rs.3.rs-5296185/v1","editorialEvents":[],"status":"published","journal":{"display":false,"email":"[email protected]","identity":"nature","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"nature","sideBox":"Learn more about [Nature](http://www.nature.com/nature/)","snPcode":"","submissionUrl":"","title":"Nature","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"f4388904-7a0f-4e86-a243-95fdeb94c6d2","owner":[],"postedDate":"April 17th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":46711627,"name":"Earth and environmental sciences/Climate sciences/Climate change/Climate-change impacts"},{"id":46711628,"name":"Earth and environmental sciences/Climate sciences/Atmospheric science/Atmospheric dynamics"}],"tags":[],"updatedAt":"2025-07-24T07:35:08+00:00","versionOfRecord":{"articleIdentity":"rs-5296185","link":"https://doi.org/10.1038/s41586-025-09242-1","journal":{"identity":"nature","isVorOnly":false,"title":"Nature"},"publishedOn":"2025-07-09 04:00:00","publishedOnDateReadable":"July 9th, 2025"},"versionCreatedAt":"2025-04-17 11:51:05","video":"","vorDoi":"10.1038/s41586-025-09242-1","vorDoiUrl":"https://doi.org/10.1038/s41586-025-09242-1","workflowStages":[]},"version":"v1","identity":"rs-5296185","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5296185","identity":"rs-5296185","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}