1. Introduction
Deep convection that overshoots the tropopause (or some other atmospheric stable layer, e.g., the equilibrium level or level of neutral buoyancy) can inject cloud ice above said layer and the broader storm anvil, producing ice clouds called above-anvil cirrus plumes (AACPs; Setvák and Doswell 1991; Wang 2003; Levizzani and Setvák 1996; Setvák et al. 2010; Homeyer et al. 2017). AACPs are primarily identified in visible (VIS) satellite imagery as a region within a storm anvil that is relatively smoother in texture when compared to the broader anvil and casts a shadow on the underlying anvil, especially near sunset (Figs. 1a,b). Often, but not always, the VIS feature is collocated with warm brightness temperature anomalies in infrared (IR) imagery that are surrounded by colder brightness temperature anomalies characteristic of the broader storm anvil (Fig. 1c), a feature commonly referred to as the enhanced or cold U, V, or ring (Adler et al. 1983; McCann 1983; Brunner et al. 2007; Setvák et al. 2010; Púčik et al. 2013; Homeyer 2014). Most prior studies focus on such warm AACPs identified in IR imagery, occasionally supplemented with confirmation from VIS imagery. In these studies, warm AACPs have been demonstrated to represent stratospheric injection of cloud material (and following sublimation, water vapor), thereby increasing their significance as important contributors to climate (Fujita 1982; Wang 2003; Mullendore et al. 2005; Wang et al. 2016; Anderson et al. 2017; Seguchi et al. 2019; Smith 2021; Zou et al. 2021). AACP-producing storms have also been recognized as prolific severe weather producers, and their utility for severe weather identification and forecasting has been documented for decades (Fujita 1982; McCann 1983; Brunner et al. 2007; Setvák et al. 2010, 2013; Bedka et al. 2015; Homeyer et al. 2017; Bedka et al. 2018; Kunz et al. 2018; Mecikalski et al. 2021).
While much of the early work on AACPs documented their satellite characteristics, recent efforts have focused on developing an understanding of mechanisms responsible for their formation. Numerous high-resolution numerical model simulations, complemented by observational analyses, have revealed that frequent gravity wave breaking borne out of strong storm-relative wind in the lower stratosphere in and near the overshooting top is responsible for AACP formation (Wang 2003, 2007; Luderer et al. 2007; Wang et al. 2016; Homeyer et al. 2017; O’Neill et al. 2021). Earlier work from Wang (2007) identified two distinct AACP modes in numerical simulations that both formed as a result of gravity wave breaking. Recent very high-resolution simulations by O’Neill et al. (2021) emphasized that the continuous gravity wave breaking responsible for AACPs is accomplished through the establishment of a hydraulic jump downstream of the overshooting top. Although AACP formation through gravity wave breaking has been demonstrated repeatedly in existing work, the presence of two (or more) distinct AACP modes have not been routinely observed or documented in AACP studies.
Despite improved understanding of AACP formation and their significance to weather and climate, there are characteristics of AACPs that remain poorly understood. One such characteristic is that some AACPs exhibit typical features in VIS imagery while appearing cold (or colder) than the broader storm top (Figs. 1b,d). Heymsfield et al. (1983) attributed the warm region near the overshoot to subsidence from the descending portion of gravity wave breaking, which was also suggested by Fujita (1974, 1982), and might suggest that mechanisms responsible for the formation of warm and cold AACPs differ. Alternatively, other studies have suggested that these IR brightness temperature differences could instead be evidence of unique microphysics in AACPs.
Setvák and Doswell (1991), Levizzani and Setvák (1996), Rosenfeld et al. (2008), Setvák et al. (2013), and Shou et al. (2019) utilized polar-orbiting satellite observations to show that AACPs often have unique microphysical signatures evidenced by higher radiance in shortwave IR imagery. However, it is exceedingly rare that these platforms encounter AACPs, as they do not sample over land during peak convection and overshooting periods. For example, Setvák et al. (2013) utilized observations from NASA’s A-Train constellation to find global instances where AACPs were sampled and documented only five cases from 2006 to 2010. Rosenfeld et al. (2008) and Lindsey et al. (2006) leveraged geostationary satellite observations to assess cloud-top microphysical structures given their higher temporal sampling for a given spatial domain, but did not focus on AACPs. In situ trace gas observations of AACPs have only recently been documented in Smith et al. (2017), highlighting elevated water vapor concentrations in the stratosphere, but unfortunately no microphysics observations are available from these cases. As such, these different microphysical characteristics could have implications for AACP temperature (e.g., due to differing optical thickness of AACPs).
Recently, Bedka et al. (2018) acknowledged the greater complexity of AACP IR signatures, provided several examples, discussed potential explanations based on existing literature, and presented several hypotheses for variable AACP IR brightness temperature: 1) sedimentation of large ice crystals that reduces the cloud optical depth, allowing a colder tropospheric anvil beneath the AACP to dominate the radiative signal; 2) AACP injection into nearly isothermal UTLS environments or above-anvil layers that are cooling with height; 3) AACP subsidence into layers with colder temperature; and/or 4) cooling of the local UTLS temperature through AACP sublimation. These hypotheses from Bedka et al. (2018), as well as potential alternative explanations for variable IR signatures, have not been evaluated to date.
In this study, we identify 89 warm and 89 cold AACPs from 1-min Geostationary Operational Environmental Satellite 16 (GOES-16) satellite imagery, coupled with ground-based radar observations and environmental information from reanalysis to answer the following research questions: 1) Why do some AACPs exhibit a warm feature in IR imagery while others do not, and 2) what observable storm and environment differences exist between warm and cold AACPs? We have outlined three key hypotheses, which are illustrated in Fig. 2, and add contextual information to several of the points outlined in Bedka et al. (2018).
For hypothesis 1 (Fig. 2, top row), warm AACPs occur in environments with a single tropopause, reside in the lower stratosphere, and appear warm because the lower-stratosphere environment is warmer than the tropopause below (this is consistent with warm AACP analyses from past studies). Cold AACPs for this hypothesis, however, occur in environments with a double tropopause (i.e., where temperature decreases significantly above the primary tropopause until eventually increasing again above the secondary tropopause), and reside in the layer between tropopauses where the lower stratosphere is colder than the primary tropopause. Double tropopauses occur most often during spring and early summer in the midlatitudes when AACPs are also frequently observed (Randel et al. 2007; Añel et al. 2008; Manney et al. 2017; Xian and Homeyer 2019; Wilhelmsen et al. 2020).
Hypothesis 2 (Fig. 2, middle row) states that warm AACPs occur in UTLS environments with midlatitude characteristics, residing in the lower stratosphere, while cold AACPs occur in UTLS environments with tropical characteristics, residing in the upper troposphere where temperature is still decreasing with height. The wide variability in tropopause height over the North American midlatitudes, both zonally and meridionally (e.g., Hoinka 1998; Li et al. 2017), suggests that such a difference could be responsible for the variable IR signatures so long as the depths of warm and cold AACP-producing storms do not differ substantially.
Last, hypothesis 3 (Fig. 2, bottom row) indicates that warm and cold AACPs occur in similar UTLS environments, but differ microphysically. Warm AACPs in this case are optically thick, such that the satellite senses the temperature of the warm stratosphere environment (compared to the anvil and tropopause layer below). On the other hand, cold AACPs are optically thin in this hypothesis, such that the satellite is sensing temperatures consistent with the broader storm anvil below, rather than that characteristic of the AACP. Differences in apparent AACP translucence in VIS imagery suggests that variability of AACP optical thickness and thus AACP IR brightness temperature could be a likely control.
Hypotheses 1 and 2 are explicitly tested in this study given the availability and utility of environmental reanalyses and ground-based radar observations in diagnosing tropopause structures and storm-relative characteristics. Thorough evaluations of hypotheses 1 and 2, devoid of explicit hypothesis 3 assessment due to the lack of available microphysical data for these AACPs, could provide sufficient evidence in favor of one explanation; if such results indicate neither hypothesis 1 nor 2 are supported, then hypothesis 3 or one of the more nuanced processes outlined in Bedka et al. (2018) would prove the most likely alternative explanation.
2. Data and methods
a. Satellite observations
The latest generation of NOAA/NASA geostationary satellites, including GOES-16, features the Advanced Baseline Imager (ABI; Schmit et al. 2017, 2018). The ABI samples 16 spectral bands at 2 km horizontal resolution every 5 min for the CONUS domain. Additionally, the ABI can perform more frequent sampling over two fixed areas, each approximately 1000 km2, that can provide more detailed observations for areas of interest (e.g., severe storms, blizzards, or tropical cyclones). When the ABI is operating in this mesoscale mode, scans are completed every 60 s, such that if the two mesoscale domains overlap, the overlapping area would be scanned every 30 s. Given the manual and highly complex nature of AACPs and their identification (Bedka et al. 2018), we only evaluated mesoscale mode imagery between March and July from years 2017–20 in this study to increase confidence in AACP identification and focus on times when AACPs are common.
AACPs are manually identified using visible imagery (0.64 μm band) and classified as either warm or cold based on their IR imagery characteristics (10.3 μm band). We first searched for cold AACPs, given their relative infrequency, recording the latitude and longitude of the corresponding overshoot at hourly intervals. We required that all AACPs used in this study were sustained (continuously emitted) for at least one hour to compare only long-lived AACPs and to ensure that hourly model analyses were representative. Once 100 cold AACPs were identified, we repeated the process for warm AACPs, but only searching during the months and years that cold AACPs had been identified to mitigate biases associated with seasonality, thus initially resulting in 100 warm and 100 cold manually identified AACP storms for analysis (reduced slightly to 89 storms each for final analysis based on additional criteria outlined in section 2e). It is important to note that regardless of their IR signatures, AACPs in this study had no discernable differences in their visible signatures.
b. Environmental data
ERA5 is the fifth and most recent generation of the global European Centre for Medium-Range Weather Forecasts (ECMWF) reanalysis and is available from 1979 to present (Hersbach et al. 2020). ERA5 features 137 vertical hybrid sigma–pressure (model) levels from the surface up to 0.01 hPa, with approximately 400–500 m resolution in the UTLS. ERA5 has 0.28 125° native horizontal resolution, but here the 1° horizontal resolution data are utilized due to the lower relative importance of horizontal gradients in storm UTLS environments compared to vertical resolution. We use vertical profiles of temperature, pressure, geopotential height and winds to examine tropopause characteristics and environmental differences between those associated with warm or cold AACPs. When defining the tropopause in this study, we use the lapse rate tropopause definition (World Meteorological Organization 1957) given its capability for multiple tropopause identification in a profile and its well-demonstrated global reliability (Gettelman et al. 2011; Hoffmann and Spang 2022; Pan et al. 2018).
c. Radar observations
To evaluate warm and cold AACP-producing storm characteristics and quantify their degree of tropopause overshooting, we utilize ground-based radar observations. Radar volumes are available on a ∼2 km horizontal resolution and 0.5–1 km vertical resolution grid and at 5-min intervals, created using version 4.2 of the Gridded NEXRAD WSR-88D Radar (GridRad) data algorithm (Homeyer and Bowman 2022). GridRad uses space- and time-weighted binning to merge individual NEXRAD volume scans into the common GridRad volume. The resulting multiradar volumes include radar reflectivity at horizontal polarization to establish echo-top heights and to calculate the level of maximum detrainment (LMD; see section 2e). GridRad echo-top heights and intrastorm features typically have near-zero bias and an uncertainty of ±1 km (e.g., Homeyer and Bowman 2022, and references therein). To enable storm-relative analyses for the AACP-producing storms in this study, objective storm tracks are computed using an echo-top algorithm developed in Homeyer et al. (2017) and updated in Lagerquist et al. (2020). The radar tracks include 5-min latitude/longitude position (interpolated to 1-min frequency for matching with severe weather reports), LMD heights, echo-top heights for multiple radar reflectivity thresholds, and 30-min mean storm motion.
d. Severe weather reports
The National Centers for Environmental Information hosts the Storm Event Database (SED) which encompasses the time and/or duration, location, magnitude, and reporting source for all U.S. severe weather reports (NOAA/NCEI 2020). A severe weather event, following the NOAA Storm Prediction Center (SPC) definition, is defined as wind gusts ≥ 50 kt, hail ≥ 1 in. (∼2.5 cm) in diameter, or any tornado report. To reduce limitations from population-based reporting, we follow methods from Murillo and Homeyer (2019), such that only severe weather reports that occurred in a population dense region (>25 people per square mile) are retained for analysis, according to version 4 of the Gridded Population of the World dataset (CIESIN 2018).
SED reports were leveraged here, as opposed to the commonly used alternative quality-controlled version released by the SPC, due to the additional tornado report information available in the original SED reports that is excluded when incorporated in the SPC. The inherent filtering that results from objective storm-matching techniques used here (see section 2e) result in a similar quality dataset to that of the SPC.
e. Data synthesis and analysis techniques
The 100 warm and 100 cold AACPs identified using satellite imagery were manually matched with their corresponding radar storm track to enable comprehensive analysis. Warm and cold AACP storms that were not fully captured in radar observations due to data coverage limitations were removed from final analysis. After reevaluating storm populations so that equal seasonal distributions for both storm sets remained, 89 warm and 89 cold AACPs were retained. The full radar storm tracks are shown in Fig. 3, with storms exhibiting warm (cold) AACPs in red (blue). We used the closest ERA5 grid box to the storm location every hour for the environmental analysis. We tested the sensitivity of the ERA5 results to the precise model grid box chosen by performing similar analyses with a randomly selected nearby grid box and found little to no impact on the results summarized below. Storm-relative winds, defined as the difference between ERA5 winds and the radar-tracked 30-min storm motion, are computed for warm and cold AACPs to characterize the potential for gravity wave breaking and thus AACP formation in the UTLS (e.g., Homeyer et al. 2017; O’Neill et al. 2021, and references therein).
Severe weather reports were matched to the radar-based storm tracks if they occurred at a time the storm was tracked and within 30 km of storm center. To complement the report analyses, we evaluate one of the recently revised hail parameters from Murillo and Homeyer (2019) and Murillo et al. (2021), the linear discriminant analysis (LDA)-filtered 75th percentile of the maximum estimated size of hail [LDA-filtered MESH75; see Murillo et al. (2021) for more details]. We also calculate the LMD in each storm to compare transport characteristics, defined here as the altitude of the column-maximum anvil ice water content, utilizing methods similar to those of Mullendore et al. (2009), Carletta et al. (2016), and Starzec et al. (2020). Several steps were taken to minimize errors in LMD height due to limitations of the radar observations and downward settling of precipitation-sized hydrometeors in the anvil separate from the convective core (e.g., Homeyer 2014; Homeyer and Bowman 2022), as was done in previous studies using GridRad data (e.g., Starzec et al. 2020). Namely, we limited our search radius to 30 km from storm center to be consistent across the analysis and applied the Storm Labeling in 3 Dimensions (SL3D; Starzec et al. 2017) algorithm to identify anvil grid boxes adjacent to convective grid boxes for ice water content analysis. We only considered convective regions where reflectivity observations exist below 4 km. Additionally, we only considered anvil regions that featured ≥15 dBZ above 6 km, ≥5 grid boxes of observed reflectivity at individual heights, and ≥25 grid boxes of observed reflectivity within the full search volume. Carletta et al. (2016) demonstrate that an LMD based on anvil ice water content is generally biased ∼750 m low compared to that derived from 3D winds. The uncertainty of GridRad-retrieved LMDs is expected to be comparable to the vertical grid spacing of the data in anvil regions, ±1 km.
f. Statistical significance
The two-sample Kolmogorov–Smirnov (KS) test is used in this study to determine statistical significance with 99% confidence (α = 0.01) for one-dimensional radar and environmental metrics of warm and cold AACP storms. The null hypothesis of the two-sample KS test states that the two populations originate from the same distribution. Statistically significant differences between samples are found when the p value < α, and thus, the null hypothesis is rejected. Although warm and cold AACP vertical profile analyses are not explicitly evaluated for statistical significance, vertical layers where there is clear separation (a lack of overlap) between the full distribution of values for each AACP population likely indicate statistically significant differences. Moreover, the approximate coincidence in the locations of warm and cold AACP storm populations implies that radar data coverage (and therefore, quality) is similar such that statistically significant differences between the two populations are robust, especially given the relatively large uncertainty (±1 km) of the GridRad echo tops and LMD heights.
3. Results
Although warm and cold AACP environments in the lower and middle troposphere are quite similar, the largest differences between them are found in the UTLS region. When looking at vertical profiles of temperature, we find that warm AACPs are associated with lower-tropopause heights and warmer UTLS temperatures, as opposed to cold AACPs environments that feature higher-tropopause heights and colder UTLS temperatures (Fig. 4, left). Namely, primary tropopause altitudes in warm AACP environments are most frequently found near 13 km, consistently lower than cold AACP environments, which commonly occur near 15 km (Fig. 6, left). We also find that temperature starts to increase immediately above the tropopause in cold AACP environments, while an inversion/isothermal layer exists above the tropopause in warm AACP environments (Fig. 4, right). These characteristics indicate that warm AACPs are associated with midlatitude environments, given the low tropopause height and tropopause inversion layer; meanwhile, cold AACPs are associated with tropical environments, given the resemblance to the cold-point tropopause and transition layer that is characteristic of tropical environments.
We also independently evaluate warm (cold) AACPs associated with double-tropopause environments, which includes 38 (13) of the 89 storms. These environments exhibit similar primary and secondary tropopause heights, but feature unique lapse rate characteristics above the primary tropopause (Fig. 5, left). In cold AACP environments, temperature continues to decrease with height, however at a slower rate, and satisfies the WMO second tropopause condition several kilometers higher (Fig. 5, right). Above the primary tropopause in warm AACP environments, however, there are clear inversion and isothermal segments up to the secondary tropopause, which is found at even higher altitudes than that in cold AACP environments. Such separation between tropopause heights is commonly ∼4 km in warm AACP environments and ∼3 km in cold AACP environments, despite similar secondary tropopause altitudes (Fig. 6, right). Thus, these results demonstrate that in a select few cold AACP cases (∼15%), the UTLS contains a double tropopause and is characterized as more of a subtropical environment, with no inversion layer above the primary tropopause and relatively small separation between tropopauses. Warm AACP environments with a double tropopause are far more common (∼43%) and, similar to the analysis for all cases, exhibit a strong tropopause inversion layer above the primary tropopause.
Next, we evaluate radar-observed characteristics of each storm population in the context of associated environmental factors. When assessing storm-relative wind speed, we find comparable structure throughout the troposphere (Fig. 7, left), but key differences are revealed in tropopause-relative analyses (Fig. 7, right). Most notably, warm AACP storms have higher storm-relative wind speeds in the first 3 km above the tropopause, while storm-relative wind in cold AACP storms remains relatively weak (below 15 m s−1) up to ∼4 km above tropopause level (Fig. 7, right). Although there are somewhat larger overall differences between warm and cold AACP storm-relative wind profiles within double-tropopause environments, tropopause-relative analyses exhibit similar above-tropopause features to that for all environments (Fig. 8). Given that previous work indicates the necessity of storm-relative wind speeds ≥ 15 m s−1 to drive frequent gravity wave breaking and AACP development (e.g., Homeyer et al. 2017; O’Neill et al. 2021, and references therein), these results suggest that AACP development above the tropopause in cold AACP environments is unlikely.
To better assess the joint relationships between AACP storms and their environments, we examine the observed depth and detrainment levels of each storm here. Absolute LMD altitudes are essentially identical in warm and cold AACP storms (Figs. 9a,c), commonly spanning altitudes from 9 to 12 km. However, tropopause-relative analysis shows that LMD heights are significantly closer to the tropopause in warm AACP storms than cold AACP storms (Figs. 9b,d), as expected based on the environmental results discussed previously. Ten-dBZ echo-top altitudes in warm and cold AACP storms also overlap considerably (Figs. 10a,c). Consistent with LMD analyses, we find that tropopause-relative 10-dBZ echo-top altitudes within warm AACP storms are significantly higher than cold AACP storms, commonly reaching 4 km above the tropopause (Figs. 10b,d). Tropopause-relative 10-dBZ echo tops within double-tropopause environments follow similar patterns to the full distributions, with slightly less overlap (Fig. 11).
Last, we assess the severity of each AACP storm population using severe weather reports and MESH75. To adequately account for the different severe storm population sizes, storm lifetimes, and additional limitations in the report database, we compute the number of severe weather reports for every 5 min that storms were within a population dense area (Table 1). Warm AACP storms produced approximately twice as many severe wind and hail reports per 5 min than cold AACP storms and comparable tornadoes. However, warm and cold AACP storms exhibit similar lifetimes with severe MESH75 (Fig. 12). Given the extensively documented limitations of hail reports, radar products, such as MESH75, are often viewed as equally reliable as severe hail reports after appropriate methods have been implemented to the reports (e.g., Allen and Tippett 2015; Murillo and Homeyer 2019; Allen et al. 2020; Murillo et al. 2021). As such, these MESH75 results cast reasonable doubt on the reliability of the severe hail report results.
Severe weather report statistics for warm and cold AACP storms.
4. Discussion and conclusions
This study evaluated the UTLS environments and storm characteristics of 89 warm and 89 cold AACP storms to answer two main research questions: 1) Why do some AACPs exhibit a warm feature in IR imagery while others do not, and 2) what observable storm and environment differences exist between warm and cold AACPs? Our conclusions are as follows:
-
Warm AACPs frequently occur in low-tropopause, midlatitude environments, while cold AACPs frequently occur in high-tropopause, tropical environments, thus supporting hypothesis 2.
-
Stratospheric AACP formation necessitates consistently strong storm-relative winds within the lower stratosphere, which are observed in warm AACP environments but not in cold AACP environments.
-
Storm characteristics (LMD and echo-top heights) are largely consistent between warm and cold AACP storms.
Analysis of tropopause characteristics in warm and cold AACP storms reveal lower-tropopause heights, broad isothermal regions above, and more frequent double tropopauses for warm AACPs compared to higher-tropopause heights with inflection point characteristics and infrequent double tropopauses for cold AACPs (Fig. 13). Similar storm-relative wind profiles and 10-dBZ echo-top altitudes for warm and cold AACP storms, despite different tropopause-relative characteristics, suggest similar storm updrafts within different environments. We have high confidence that these warm AACPs occur in the lower stratosphere, evidenced by ∼95% of warm AACP 10-dBZ echo tops exceeding the tropopause, where sufficient dynamical support for AACP formation exists. Conversely, only ∼50% of cold AACP 10-dBZ echo tops exceed the tropopause, suggesting that sustained AACP production, which is needed for cases in this study, in the lower stratosphere is unlikely. Given that there is also a lack of dynamical support (strong storm-relative wind) for gravity wave breaking in the lower stratosphere of cold AACP environments, when tropopause overshooting did occur, it was likely intermittent and unlikely to produce the observed sustained AACPs in the lower stratosphere (Fig. 13). Thus, cold AACPs most likely reside in the upper troposphere (i.e., below the tropopause) where sufficient storm-relative winds are present.
The aforementioned result that warm AACPs reside in the stratosphere and cold AACPs reside in the troposphere has important implications for studies that aim to assess the stratospheric water vapor impact of AACP-producing storms. While it is still possible that hypothesis 3 is also valid in some (or many) instances, it is likely a minimal contribution to AACP IR characteristics in comparison to hypothesis 2 given the clear differences in the storm-relative environments in warm and cold AACP storms found here. Although AACP microphysics were not evaluated in this study, work is ongoing in collaboration with the authors to leverage numerous channels of the ABI aboard GOES-16/-17 to explicitly evaluate the unique microphysics of AACPs. In addition, the recently conducted NASA Earth Venture’s Suborbital field project, Dynamics and Chemistry of the Summer Stratosphere (DCOTSS), was specifically designed to collect in situ observations of AACPs in the stratosphere. DCOTSS data may help to improve microphysics understanding in AACPs.
Finally, while some apparent differences in storm severity (hail and wind) were found using reports for warm and cold AACP storm populations, with warm AACP storms being more prolific severe weather producers, independent evaluation of storm characteristics from radar suggest that such differences (at least for hail) are an artifact of the limited report data (cf. Table 1 and Fig. 12). Analysis of AACP-producing storms using a more comprehensive report database (e.g., as in Blair et al. 2017) would help to resolve this inconsistency.
Acknowledgments.
This work was supported by the National Aeronautics and Space Administration (NASA) under Award 80NSSC19K0347. We thank Kris Bedka at NASA Langley Research Center for helpful discussions and feedback on this work.
Data availability statement.
All data used in this study are publicly available. GOES and NEXRAD observations are hosted by NCEI and openly available at https://www.ncei.noaa.gov/access/cloud-access as part of the NOAA Big Data Program as cited in NOAA/NCEI (2022) and NOAA/NWS/Radar Operations Center (1991), ERA5 model-level output was obtained at https://apps.ecmwf.int/data-catalogues/era5/ (Hersbach et al. 2020), severe weather reports were obtained from NCEI at https://www.ncdc.noaa.gov/stormevents/ as cited in NOAA/NCEI (2020), and population density was obtained at https://doi.org/10.7927/H49C6VHW as cited in CIESIN (2018). The warm and cold AACP storm hourly locations produced for this study are hosted by Zenodo and openly available at https://doi.org/10.5281/zenodo.6977684, as cited in Murillo (2022).
REFERENCES
Adler, R. F., M. J. Markus, D. D. Fenn, G. Szejwach, and W. E. Shenk, 1983: Thunderstorm top structure observed by aircraft overflights with an infrared radiometer. J. Climate Appl. Meteor., 22, 579–593, https://doi.org/10.1175/1520-0450(1983)022<0579:TTSOBA>2.0.CO;2.
Allen, J. T., and M. Tippett, 2015: The characteristics of United States hail reports: 1955–2014. Electron. J. Severe Storms Meteor., 10 (3), https://doi.org/10.55599/ejssm.v10i3.60.
Allen, J. T., I. M. Giammanco, M. R. Kumjian, H. J. Punge, Q. Zhang, P. Groenemeijer, M. Kunz, and K. Ortega, 2020: Understanding hail in the Earth system. Rev. Geophys., 58, e2019RG000665, https://doi.org/10.1029/2019RG000665.
Anderson, J. G., and Coauthors, 2017: Stratospheric ozone over the United States in summer linked to observations of convection and temperature via chlorine and bromine catalysis. Proc. Natl. Acad. Sci. USA, 114, E4905–E4913, https://doi.org/10.1073/pnas.1619318114.
Añel, J. A., J. C. Antuña, L. de la Torre, J. M. Castanheira, and L. Gimeno, 2008: Climatological features of global multiple tropopause events. J. Geophys. Res., 113, D00B08, https://doi.org/10.1029/2007JD009697.
Bedka, K. M., C. Wang, R. Rogers, L. D. Carey, W. Feltz, and J. Kanak, 2015: Examining deep convective cloud evolution using total lightning, WSR-88D, and GOES-14 super rapid scan datasets. Wea. Forecasting, 30, 571–590, https://doi.org/10.1175/WAF-D-14-00062.1.
Bedka, K. M., E. M. Murillo, C. R. Homeyer, B. Scarino, and H. Mersiovsky, 2018: The above anvil cirrus plume: An important severe weather indicator in visible and infrared satellite imagery. Wea. Forecasting, 33, 1159–1181, https://doi.org/10.1175/WAF-D-18-0040.1.
Blair, S. F., and Coauthors, 2017: High-resolution hail observations: Implications for NWS warning operations. Wea. Forecasting, 32, 1101–1119, https://doi.org/10.1175/WAF-D-16-0203.1.
Brunner, J. C., S. A. Ackerman, A. S. Bachmeier, and R. M. Rabin, 2007: A quantitative analysis of the enhanced-V feature in relation to severe weather. Wea. Forecasting, 22, 853–872, https://doi.org/10.1175/WAF1022.1.
Carletta, N. D., G. L. Mullendore, M. Starzec, B. Xi, Z. Feng, and X. Dong, 2016: Determining the best method for estimating the observed level of maximum detrainment based on radar reflectivity. Mon. Wea. Rev., 144, 2915–2926, https://doi.org/10.1175/MWR-D-15-0427.1.
CIESIN, 2018: Gridded Population of the World, version 4 (GPWv4): Population density, revision 11. NASA SEDAC, accessed 8 September 2021, https://doi.org/10.7927/H49C6VHW.
Fujita, T. T., 1974: Overshooting thunderheads observed from ATS and Learjet. Satellite and Mesometeorology Research Project Rep. 117, 29 pp.
Fujita, T. T., 1982: Principle of stereographic height computations and their application to stratospheric cirrus over severe thunderstorms. J. Meteor. Soc. Japan, 60, 355–368, https://doi.org/10.2151/jmsj1965.60.1_355.
Gettelman, A., P. Hoor, L. L. Pan, W. J. Randel, M. I. Hegglin, and T. Birner, 2011: The extratropical upper troposphere and lower stratosphere. Rev. Geophys., 49, RG3003, https://doi.org/10.1029/2011RG000355.
Hersbach, H., and Coauthors, 2020: The ERA5 global reanalysis. Quart. J. Roy. Meteor. Soc., 146, 1999–2049, https://doi.org/10.1002/qj.3803.
Heymsfield, G. M., R. H. Blackmer, and S. Schotz, 1983: Upper-level structure of Oklahoma tornadic storms on 2 May 1979. I: Radar and satellite observations. J. Atmos. Sci., 40, 1740–1755, https://doi.org/10.1175/1520-0469(1983)040<1740:ULSOOT>2.0.CO;2.
Hoffmann, L., and R. Spang, 2022: An assessment of tropopause characteristics of the ERA5 and ERA-Interim meteorological reanalyses. Atmos. Chem. Phys., 22, 4019–4046, https://doi.org/10.5194/acp-22-4019-2022.
Hoinka, K. P., 1998: Statistics of the global tropopause pressure. Mon. Wea. Rev., 126, 3303–3325, https://doi.org/10.1175/1520-0493(1998)126<3303:SOTGTP>2.0.CO;2.
Homeyer, C. R., 2014: Formation of the enhanced-V infrared cloud-top feature from high-resolution three-dimensional radar observations. J. Atmos. Sci., 71, 332–348, https://doi.org/10.1175/JAS-D-13-079.1.
Homeyer, C. R., and K. P. Bowman, 2022: Algorithm description document for version 4.2 of the three-dimensional gridded NEXRAD WSR-88D Radar (GridRad) dataset. GridRad Tech. Rep., 30 pp.
Homeyer, C. R., J. D. McAuliffe, and K. M. Bedka, 2017: On the development of above-anvil cirrus plumes in extratropical convection. J. Atmos. Sci., 74, 1617–1633, https://doi.org/10.1175/JAS-D-16-0269.1.
Kunz, M., U. Blahak, J. Handwerker, M. Schmidberger, H. J. Punge, S. Mohr, E. Fluck, and K. M. Bedka, 2018: The severe hailstorm in southwest Germany on 28 July 2013: Characteristics, impacts and meteorological conditions. Quart. J. Roy. Meteor. Soc., 144, 231–250, https://doi.org/10.1002/qj.3197.
Lagerquist, R., A. McGovern, C. R. Homeyer, D. J. Gagne II, and T. Smith, 2020: Deep learning on three-dimensional multiscale data for next-hour tornado prediction. Mon. Wea. Rev., 148, 2837–2861, https://doi.org/10.1175/MWR-D-19-0372.1.
Levizzani, V., and M. Setvák, 1996: Multispectral, high-resolution satellite observations of plumes on top of convective storms. J. Atmos. Sci., 53, 361–369, https://doi.org/10.1175/1520-0469(1996)053<0361:MHRSOO>2.0.CO;2.
Li, W., Y. Yuan, Y. Chai, Y. Liou, J. Ou, and S. Zhong, 2017: Characteristics of the global thermal tropopause derived from multiple radio occultation measurements. Atmos. Res., 185, 142–157, https://doi.org/10.1016/j.atmosres.2016.09.013.
Lindsey, D. T., D. W. Hillger, L. Grasso, J. A. Knaff, and J. F. Dostalek, 2006: GOES climatology and analysis of thunderstorms with enhanced 3.9-μm reflectivity. Mon. Wea. Rev., 134, 2342–2353, https://doi.org/10.1175/MWR3211.1.
Luderer, G., J. Trentmann, K. Hungershöfer, M. Herzog, M. Fromm, and M. O. Andreae, 2007: Small-scale mixing processes enhancing troposphere-to-stratosphere transport by pyro-cumulonimbus storms. Atmos. Chem. Phys., 7, 5945–5957, https://doi.org/10.5194/acp-7-5945-2007.
Manney, G., and Coauthors, 2017: Reanalysis comparisons of upper tropospheric–lower stratospheric jets and multiple tropopauses. Atmos. Chem. Phys., 17, 11 541–11 566, https://doi.org/10.5194/acp-17-11541-2017.
McCann, D. W., 1983: The enhanced-V: A satellite observable severe storm signature. Mon. Wea. Rev., 111, 887–894, https://doi.org/10.1175/1520-0493(1983)111<0887:TEVASO>2.0.CO;2.
Mecikalski, J. R., T. N. Sandmæl, E. M. Murillo, C. R. Homeyer, K. M. Bedka, J. M. Apke, and C. P. Jewett, 2021: A random-forest model to assess predictor importance and nowcast severe storms using high-resolution radar–GOES satellite–lightning observations. Mon. Wea. Rev., 149, 1725–1746, https://doi.org/10.1175/MWR-D-19-0274.1.
Mullendore, G. L., D. R. Durran, and J. R. Holton, 2005: Cross-tropopause tracer transport in midlatitude convection. J. Geophys. Res., 110, D06113, https://doi.org/10.1029/2004JD005059.
Mullendore, G. L., A. J. Homann, K. Bevers, and C. Schumacher, 2009: Radar reflectivity as a proxy for convective mass transport. J. Geophys. Res., 114, D16103, https://doi.org/10.1029/2008JD011431.
Murillo, E. M., 2022: Above-anvil cirrus plume producing storm tracks. Zenodo, accessed 8 August 2022, https://doi.org/10.5281/zenodo.6977684
Murillo, E. M., and C. R. Homeyer, 2019: Severe hail fall and hailstorm detection using remote sensing observations. J. Appl. Meteor. Climatol., 58, 947–970, https://doi.org/10.1175/JAMC-D-18-0247.1.
Murillo, E. M., C. R. Homeyer, and J. T. Allen, 2021: A 23-year severe hail climatology using GridRad MESH observations. Mon. Wea. Rev., 149, 945–958, https://doi.org/10.1175/MWR-D-20-0178.1.
NOAA/NCEI, 2020: Storm events database, version 3.1. NOAA/NCEI, accessed January 2022, https://www.ncdc.noaa.gov/stormevents/.
NOAA/NCEI, 2022: NCEI data in the NOAA big data program. NOAA/NCEI, accessed January 2022, https://www.ncei.noaa.gov/access/cloud-access.
NOAA/NWS/Radar Operations Center, 1991: NOAA Next Generation Radar (NEXRAD) level 2 base data. NOAA/NCEI, accessed January 2022, https://doi.org/10.7289/V5W9574V.
O’Neill, M. E., L. Orf, G. M. Heymsfield, and K. Halbert, 2021: Hydraulic jump dynamics above supercell thunderstorms. Science, 373, 1248–1251, https://doi.org/10.1126/science.abh3857.
Pan, L. L., S. B. Honomichl, T. V. Bui, T. Thornberry, A. Rollins, E. Hintsa, and E. J. Jensen, 2018: Lapse rate or cold point: The tropical tropopause identified by in situ trace gas measurements. Geophys. Res. Lett., 45, 10 756–10 763, https://doi.org/10.1029/2018GL079573.
Púčik, T., M. Valachová, and P. Zacharov, 2013: Upper tropospheric conditions in relation to the cloud top features of 15 August 2010 convective storms. Atmos. Res., 123, 249–267, https://doi.org/10.1016/j.atmosres.2012.10.006.
Randel, W. J., D. J. Seidel, and L. L. Pan, 2007: Observational characteristics of double tropopauses. J. Geophys. Res., 112, D07309, https://doi.org/10.1029/2006JD007904.
Rosenfeld, D., W. L. Woodley, A. Lerner, G. Kelman, and D. T. Lindsey, 2008: Satellite detection of severe convective storms by their retrieved vertical profiles of cloud particle effective radius and thermodynamic phase. J. Geophys. Res., 113, D04208, https://doi.org/10.1029/2007JD008600.
Schmit, T. J., P. Griffith, M. M. Gunshor, J. M. Daniels, S. J. Goodman, and W. J. Lebair, 2017: A closer look at the ABI on the GOES-R series. Bull. Amer. Meteor. Soc., 98, 681–698, https://doi.org/10.1175/BAMS-D-15-00230.1.
Schmit, T. J., S. S. Lindstrom, J. J. Gerth, and M. M. Gunshor, 2018: Applications of the 16 spectral bands on the Advanced Baseline Imager (ABI). Electron. J. Oper. Meteor., 6, 33–46, https://doi.org/10.15191/nwajom.2018.0604.
Seguchi, T., S. Iwasaki, M. Kamogawa, T. Ushiyama, and H. Okamoto, 2019: Observation of jumping cirrus with ground-based cameras, radiosonde, and Himawari-8. J. Meteor. Soc. Japan, 97, 615–632, https://doi.org/10.2151/jmsj.2019-033.
Setvák, M., and C. A. Doswell III, 1991: The AVHRR channel 3 cloud top reflectivity of convective storms. Mon. Wea. Rev., 119, 841–847, https://doi.org/10.1175/1520-0493(1991)119<0841:TACCTR>2.0.CO;2.
Setvák, M., and Coauthors, 2010: Satellite-observed cold-ring-shaped features atop deep convective clouds. Atmos. Res., 97, 80–96, https://doi.org/10.1016/j.atmosres.2010.03.009.
Setvák, M., K. M. Bedka, D. T. Lindsey, A. Sokol, Z. Charvát, J. St´ástka, and P. K. Wang, 2013: A-Train observations of deep convective storm tops. Atmos. Res., 123, 229–248, https://doi.org/10.1016/j.atmosres.2012.06.020.
Shou, Y.-X., F. Lu, H. Liu, P. Cui, S. W. Shou, and J. Liu, 2019: Satellite-based observational study of the Tibetan Plateau vortex: Features of deep convective cloud tops. Adv. Atmos. Sci., 36, 189–205, https://doi.org/10.1007/s00376-018-8049-y.
Smith, J. B., 2021: Convective hydration of the stratosphere. Science, 373, 1194–1195, https://doi.org/10.1126/science.abl8740.
Smith, J. B., and Coauthors, 2017: A case study of convectively sourced water vapor observed in the overworld stratosphere over the United States. J. Geophys. Res. Atmos., 122, 9529–9554, https://doi.org/10.1002/2017JD026831.
Starzec, M., C. R. Homeyer, and G. L. Mullendore, 2017: STORM Labeling in Three Dimensions (SL3D): A volumetric radar echo and dual-polarization updraft classification algorithm. Mon. Wea. Rev., 145, 1127–1145, https://doi.org/10.1175/MWR-D-16-0089.1.
Starzec, M., G. L. Mullendore, and C. R. Homeyer, 2020: Retrievals of convective detrainment heights using ground-based radar observations. J. Geophys. Res. Atmos., 125, e2019JD031164, https://doi.org/10.1029/2019JD031164.
Wang, P. K., 2003: Moisture plumes above thunderstorm anvils and their contributions to cross-tropopause transport of water vapor in midlatitudes. J. Geophys. Res., 108, 4194, https://doi.org/10.1029/2002JD002581.
Wang, P. K., 2007: The thermodynamic structure atop a penetrating convective thunderstorm. Atmos. Res., 83, 254–262, https://doi.org/10.1016/j.atmosres.2005.08.010.
Wang, P. K., K.-Y. Cheng, M. Setvák, and C.-K. Wang, 2016: The origin of the gullwing-shaped cirrus above an Argentinian thunderstorm as seen in CALIPSO images. J. Geophys. Res. Atmos., 121, 3729–3738, https://doi.org/10.1002/2015JD024111.
Wilhelmsen, H., F. Ladstadter, T. Schmidt, and A. K. Steiner, 2020: Double tropopauses and the tropical belt connected to ENSO. Geophys. Res. Lett., 47, e2020GL089027, https://doi.org/10.1029/2020GL089027.
World Meteorological Organization, 1957: Meteorology—A three-dimensional science: Second session of the Commission for Aerology. WMO Bull., 4, 134–138.
Xian, T., and C. R. Homeyer, 2019: Global tropopause altitudes in radiosondes and reanalyses. Atmos. Chem. Phys., 19, 5661–5678, https://doi.org/10.5194/acp-19-5661-2019.
Zou, L., L. Hoffmann, S. Griessbach, R. Spang, and L. Wang, 2021: Empirical evidence for deep convection being a major source of stratospheric ice clouds over North America. Atmos. Chem. Phys., 21, 10 457–10 475, https://doi.org/10.5194/acp-21-10457-2021.