1. Introduction
Tropical cyclone (TC) intensity forecasts have improved very slowly over the past few decades, and average National Hurricane Center (NHC) 48-h intensity forecast errors remain in the range of 10–20 kt (1 kt = 0.5144 m s−1) (National Hurricane Center 2014). Intensity forecasts are particularly poor for episodes of rapid intensification (RI; Kaplan and DeMaria 2003); in which TCs strengthen by at least 15 m s−1 within 24 h. Because of the large forecast errors associated with RI and the need to more accurately forecast the strength of landfalling storms, improving forecasts of rapidly strengthening TCs is a top priority of the research and forecasting communities.
The TC intensity change is influenced by both large-scale environmental conditions, and by meso- to microscale processes within the vortex core (Marks and Shay 1998; Hendricks et al. 2010; Rogers et al. 2013b). Past research has successfully identified the environmental conditions that tend to favor TC intensification: high sea surface temperature (SST), low vertical wind shear, and high low- to midlevel relative humidity. Warm SSTs are necessary for the development of mature TCs, and SSTs determine the maximum intensity that can be reached given otherwise favorable atmospheric conditions (Merrill 1988; DeMaria and Kaplan 1994). The entrainment of dry air at low or midlevels can weaken TCs (Dunion and Velden 2004), so relatively humid environments are preferred for intensification. Vertical shear greater than ~10 m s−1 prevents intensification of most TCs. Hypothesized detrimental influences from shear include increased inner-core static stability due to tilting of the vortex (DeMaria 1996), ventilation of the warm core at upper levels (Frank and Ritchie 2001), and entrainment of dry air at mid- to lower levels (Cram et al. 2007; Riemer et al. 2010; Tang and Emanuel 2010). Some of the lack of intensity forecast improvement might stem from failing to account for the interactions between different environmental conditions. For example, the midlevel ventilation enabled by vertical wind shear may make TCs more sensitive to midlevel environmental humidity in the presence of higher shear (Emanuel et al. 2004).
In addition to an improved knowledge of environmental effects, improvement in TC intensity forecasting might also be achieved by developing an improved understanding of internal processes and attempting to determine which inner-core precipitation characteristics tend to lead to TC intensification. Several studies have identified intense convection as an important predictor, and specifically precursor, to TC intensification; including intense convective regions that persist for several hours (Steranka et al. 1986) and smaller cores on the scale of individual convective cells (Kelley et al. 2004; Jiang 2012). Intensification from intense inner-core convection has been attributed, thermodynamically, to subsidence warming forced by this strong convection (Guimond et al. 2010; Chen and Zhang 2013) and dynamically through aggregation of cyclonic vorticity generated in the updraft cores [e.g., vortical hot towers (VHTs); Van Sang et al. 2008; Houze et al. 2009; Sanger et al. 2014]. In contrast, some recent studies (Kieper and Jiang 2012; Jiang and Ramirez 2013; Zagrodnik and Jiang 2014) have shown that intense convection does not necessarily precede periods of rapid intensification. Instead, these authors hypothesize that relatively high inner-core azimuthal and areal precipitation coverage are better indicators of future RI than individual intense convective bursts.
Inner-core convection occurs in the presence of the powerful TC primary and secondary circulations, and the radial location of latent heating relative to that circulation might determine the degree to which the heating leads to TC intensification. In idealized axisymmetric model experiments, the wind speed increase due to a prescribed latent heating region is greater when the heat source is placed at smaller radii (Schubert and Hack 1982; Pendergrass and Willoughby 2009), with the highest heating efficiencies inside the radius of maximum wind (RMW; Pendergrass and Willoughby 2009; Vigh and Schubert 2009). Recently, Rogers et al. (2013a) showed that strong convective updrafts tend to be located within the low-level RMW in intensifying TCs and outside of the low-level RMW in steady-state TCs, indicating that convective contributions to TC intensification are enhanced within the RMWs of real storms. So while intense convection may not be necessary for TC intensification, when intense convection is present its radial distribution relative to the RMW can help to determine the likelihood of intensification.
The current study investigates the relationship between environmental conditions, precipitation properties, and intensity change in Hurricane Earl (2010). Earl developed from an African easterly wave that entered the Atlantic Ocean on 23 August 2010, reaching tropical storm strength by 2100 UTC 25 August (Cangiolosi 2011). From 26 to 28 August, Earl tracked westward across the tropical central Atlantic and slowly strengthened (Cangiolosi 2011). On 29–30 August, Earl rapidly intensified from a 55-kt tropical storm to a 115-kt hurricane (Fig. 1). In this study, RI is defined as the 36-h period, from 0600 UTC 29 August to 1800 UTC 30 August, during which the intensification rate equaled or exceeded 15 m s−1 (30 kt) (24-h day)−1 [consistent with the rate defined in Kaplan and DeMaria (2003)] according to the NHC best track analysis (Cangiolosi 2011, hereafter “best track”). Aircraft sampling before and during RI includes the National Oceanic and Atmospheric Administration (NOAA) WP-3D (P3) and Gulfstream IV (GIV) aircraft [part of the ongoing NOAA Intensity Forecasting Experiment (IFEX); Rogers et al. 2006, 2013b], the National Aeronautics and Space Administration (NASA) DC-8 aircraft [part of the Genesis and Rapid Intensification Processes campaign (GRIP); Braun et al. 2013], as well as the Air Force WC-130J (C130) aircraft. With the frequent observations provided by these aircraft, the datasets available during the RI of Earl are unprecedented.
NHC best track and TRMM 3B42. (a) accumulated rainfall (between 0000 UTC 27 Aug and 0000 UTC 1 Sep) and (b) maximum 10-m wind speed. The red line segment in (a) and the gray shading in (b) indicate the RI period. Blue bars indicate approximate sampling periods for NOAA P3 flights.
Citation: Monthly Weather Review 143, 6; 10.1175/MWR-D-14-00283.1
Recent observational studies on the RI of Earl (Rogers et al. 2015; Stevenson et al. 2014) have described the environmental wind shear evolution and the spatial and temporal evolution of intense convection, as well as vortex-scale kinematic structure, using aircraft radar reflectivity and three-dimensional Doppler wind analyses, dropsonde observations, aircraft in situ measurements, lightning data, and global model analyses. Some of the same lightning, airborne radar, and in situ datasets are used here. However, the present study expands upon previous studies by also utilizing passive microwave satellite data. The inclusion of passive microwave observations allows for a documentation of the coverage of both intense convection and at-least moderate rainfall before, during, and immediately after RI at a temporal resolution of ≤6 h. With this comprehensive collection of observations and objective analyses, the present study will further examine the relationship between Earl’s RI and environmental conditions, intense convection, kinematic structure, and precipitation coverage and symmetry.
The remaining sections are organized as follows: section 2 describes the collection of datasets utilized in this study; section 3 describes the evolution of the large-scale environmental conditions around Earl, the kinematic structure and precipitation characteristics within the vortex core, and the interactions between environmental conditions and the internal structures of the storm; and section 4 provides a summary of the results and some conclusions.
2. Data and methodology
a. Global atmospheric and SST analyses
The National Centers for Environmental Prediction (NCEP) Final (FNL) operational analysis wind and humidity fields are utilized for an assessment of the atmospheric environmental conditions. The FNL is available four times daily at the synoptic times (0000, 0600, 1200, and 1800 UTC) at a horizontal resolution of 1°. Environmental parameters are calculated from the NCEP FNL analyses as average values within an annulus of 1.8°–7.2° from the best track center. The shear magnitude and direction are computed from the vector difference between the average wind vector at the lower and higher pressure levels within this annulus. SST data are taken from the NCEP Real Time Global (RTG) SST analyses. NCEP produces these analyses once daily by interpolating 24 h of buoy, ship, and satellite SST data to a ¼°- resolution grid.
b. Satellite data
Passive microwave (PMW) data are utilized to examine the precipitation characteristics in Earl, especially at times when the NOAA P3 aircraft radar data are unavailable. Special Sensor Microwave Imager/Sounder (SSMIS) brightness temperatures are obtained from the Fundamental Climate Data Record (FCDR; Sapiano et al. 2013) of intercalibrated microwave imager brightness temperatures, Tropical Rainfall Measuring Mission (TRMM) Microwave Imager (TMI) level-1 calibrated data are from the University of Utah Precipitation Feature Database (Liu et al. 2008), and Advanced Microwave Scanning Radiometer Earth Observing System (AMSR-E) data are obtained from the National Snow and Ice Data Center (NSIDC) AMSR-E/Aqua Global Swath Brightness Temperatures dataset (Ashcroft and Wentz 2013). Microwave Humidity Sounder (MHS) and Advanced Microwave Sounder Unit-B (AMSU-B) brightness temperatures are obtained from the NASA Jet Propulsion Laboratory (JPL) Tropical Cyclone Information System (http://tropicalcyclone.jpl.nasa.gov). Each of the aforementioned PMW instruments measures brightness temperatures between 85 and 91 GHz. The spatial resolution in this channel is around 12.5 km for the SSMIS and approximately 6 km for the TMI and AMSR-E instruments. AMSU-B and MHS have a spatial resolution of 16 km at nadir, decreasing to near 36 km at the limbs of the swaths. An image of the TRMM Precipitation Radar (PR) reflectivity and TMI 37-GHz color composite is also shown for a single overpass on 28 August 2010. The PR has a swath width of approximately 250 km and horizontal resolution of 5 km, while the TMI 37-GHz data have a resolution of approximately 13 km.
AMSR-E, TMI, and SSMIS all scan conically and measure 85–91-GHz brightness temperatures in both vertical and horizontal polarizations, while 85–91-GHz data from the cross-track-scanning AMSU-B and MHS have a polarization that varies with the scan angle. The 85–91-GHz brightness temperatures are depressed over frozen and mixed-phase precipitating regions due to scattering from frozen precipitation, and in clear air over oceans and lakes because of low water surface emissivity. Land emissivity is high at 85–91 GHz, so brightness temperatures are elevated over landmasses. To remove the contrast between land and ocean surface emissivity, polarization corrected temperature (PCT; Spencer et al. 1989) is calculated from the AMSR-E, TMI, and SSMIS data.
c. NOAA P3 radar data
Figure 1b shows the timing of the NOAA P3 flights during the time period of interest. The P3 flew once before the onset of RI, three times during RI, and once a few hours after the end of RI. Flights are named according to the starting date and a letter, “I” or “H,” indicating the aircraft (N43F and N42F, respectively). The NOAA P3s are equipped with vertically scanning 3.22-cm wavelength Doppler radars on the tail end of each aircraft [tail Doppler radars (TDRs)], which employ the “fore-aft” pseudo dual-Doppler scanning strategy (FAST; Gamache et al. 1995) as the P3s fly successive radial passes through TCs. The NOAA Hurricane Research Division (HRD) generates separate wind retrievals and reflectivity analyses for each transect; using the method described in Reasor et al. (2009), with horizontal and vertical resolutions of 2 and 0.5 km, respectively. These TDR analyses are the same set of Doppler and reflectivity analyses presented in Rogers et al. (2015).
The individual TDR analyses have limited data coverage, so “merged” analyses are developed for each P3 flight by compositing all flight legs [as in Rogers et al. (2015)]. Although compositing results in a loss of effective resolution, the merged analyses can still illustrate slowly evolving aspects of Earl’s structure, such as the azimuthally averaged winds. The RMW is calculated using the azimuthally averaged tangential winds from these merged analyses. The RMWs at 2 km (“2-km RMW”) and 8 km (“8-km RMW”) altitude are then interpolated between P3 flights to create hourly RMW locations at those altitudes. Low data coverage prevents calculation of the 8-km RMW from the 828I flight, so the 8-km RMW is only interpolated between the 829H and 830I flights. These interpolated RMW locations are created for the purpose of exploring the radial distribution of convection relative to the RMW.
d. WWLLN
The World Wide Lightning Location Network (WWLLN; Rodger et al. 2006) is a network of ground observation stations that detect very low frequency (VLF) radiation from lightning strikes. The WWLLN is able to detect both cloud-to-ground (CG) and intracloud (IC) strikes, but detection efficiency (DE) is higher for stronger peak currents (Rodger et al. 2006; Jacobson et al. 2006), thus the DE is higher for CG lightning (Abarca et al. 2010). Various validation efforts cite CG detection efficiencies between 10% and 20% around the location of Earl in the western Atlantic (Abarca et al. 2010; Hutchins et al. 2012; Rudlosky and Shea 2013). The WWLLN reporting technique ensures that the location accuracy is within 10 km (R. Holzworth 2014, personal communication) for the vast majority of detected strikes, and the results from Hutchins et al. (2012) strongly suggest that the location accuracy was approximately 5 km for the northern Caribbean and nearby Atlantic Ocean regions during the lifetime of Hurricane Earl.
Electrification of convective cells requires much stronger updrafts than what is typically observed in tropical maritime convection (Zipser and Lutz 1994; Black and Hallett 1999; Reinhart et al. 2014), so lightning can serve as a proxy for intense convective activity in tropical cyclones. Despite the low DE, WWLLN data accurately capture the mesoscale spatial structure of lightning within TCs (Abarca et al. 2011), so WWLLN data can serve as a useful indication of the frequency and spatial distribution of intense convection in Earl.
e. Definition of a convective burst
There are many different definitions of “convective bursts” in TC literature (Steranka et al. 1986; Rogers et al. 2013a). Unfortunately, these different definitions can lead to confusion, since some definitions refer to periods of intense convection on the meso-β spatial scale, which have temporal scales of several hours or more (Steranka et al. 1986), while others describe individual short-lived vigorous convective-scale features (Guimond et al. 2010; Rogers et al. 2013a). Hereafter, we will refer to a convective burst as a period of intense (PCT ≤ 200 K, lightning frequency > 1 strike min−1) inner-core meso-β convection that persists on a time scale of several hours.
f. Uncertainty in time of RI onset
Surface wind observations from airborne Stepped Frequency Microwave Radiometers (SFMR) and dropsondes are available during the 828I (4–9 h before the best track RI onset) and 829H (3–8 h after the best track RI onset) P3 flights, as well as from an Air Force C130 flight that made its first pass through Earl near 1200 UTC 29 August (6 h after best track RI onset). These observations contribute to the best track wind speed estimates at 0000 and 1200 UTC 29 August, however, the estimated 55-kt maximum wind at 0600 UTC 29 August is much more uncertain, given the lack of aircraft data available around that time. Therefore, a precise timing of the onset of RI does not exist. As a consequence, processes occurring from approximately 1800 UTC 28 August through 1200 UTC 29 August will all be considered as potential contributors to the onset of RI.
3. Results
a. Evolution of environmental properties
The SSTs near Earl (Fig. 2a) are at magnitudes that favor rapid intensification of Atlantic TCs (Kaplan and DeMaria 2003) as early as 30 h before the onset of RI. Improving SST conditions may have played some role in the timing of the start of RI, as the SSTs increase by approximately 1°C between 12 h before and 18 h after the onset of RI. Low-level and midlevel relative humidity (RH; Fig. 2b) changes very little before, during, and after RI, with the exception of a slight dip in RH toward the latter part of the RI period. The magnitudes of the low-level RH before and during RI are at values that are below the mean for rapidly intensifying Atlantic storms, but still high enough to support RI in many cases (Kaplan and DeMaria 2003; Kaplan et al. 2010).
(a) NCEP RTG analyses mean SST within 3° from Earl and (b) NCEP FNL low-level and midlevel layer-average environmental relative humidity.
Citation: Monthly Weather Review 143, 6; 10.1175/MWR-D-14-00283.1
The 850–200- and 850–150-mb (1 mb = 1 hPa) deep-layer shear (Figs. 3a,b) are low 24–48 h prior to RI onset. Earl then encounters moderate shear (>6 m s−1) between 18 h before and 12 h after RI onset due to northeasterly outflow from Hurricane Danielle (Figs. 4a,b); this outflow appears to be maximized near 150 mb, hence, the separate time series shown in Fig. 3. As Earl moves west of this outflow channel (Figs. 4c,d), deep-layer shear decreases after 6 h into the RI period and falls below 6 m s−1 during the final 12 h of RI.
NCEP FNL deep-layer environmental shear (a) magnitude and (b) direction. The RI time period is shaded gray.
Citation: Monthly Weather Review 143, 6; 10.1175/MWR-D-14-00283.1
FNL 200-mb wind and 850-mb absolute vorticity (purple contours for vorticity >20−5 s−1, interval 20−5 s−1) at (a) 1200 UTC 28 Aug, (b) 0600 UTC 29 Aug, (c) 0000 UTC 30 Aug, and (d) 1800 UTC 30 Aug. The red “×” marks the best track TC fixes. Danielle is the elevated vorticity region north of Earl (near 29°N, 61°W at 1200 UTC 28 Aug).
Citation: Monthly Weather Review 143, 6; 10.1175/MWR-D-14-00283.1
b. Evolution of convective and kinematic properties
Figures 5a–n show PMW overpasses at several times before and during the RI, while the time series of inner-core fractional coverage of PCT ≤ 250 K and PCT ≤ 200 K are displayed in Fig. 6. PCTs ≤ 250 K correspond to regions with at least moderate rainfall rates (Spencer et al. 1989), so the 250-K coverage serves as an indication of the inner-core precipitation coverage. Considering that PCTs ≤ 200 K represent the lowest 4.4% of the distribution of PCTs observed in hurricane eyewalls (Cecil et al. 2002), and are also associated with an elevated probability of lightning (Liu et al. 2011), the fractional area of PCT ≤ 200 K serves as a proxy for the coverage of intense convection. Likewise, the time series of inner-core lightning frequency (Fig. 7a) serves as a proxy for the frequency of intense convection. For the purpose of examining the radial distribution of intense convection and moderate rainfall relative to the RMW, Hovmöller diagrams of lightning density (Fig. 7b), azimuthal coverage of PCT ≤ 250 K (Fig. 8a), and PCT ≤ 200 K (Fig. 8b) are also shown.
The 85–91-GHz images at (a) 0839 UTC 27 Aug, (b) 2233 UTC 27 Aug, (c) 0413 UTC 28 Aug, (d) 1639 UTC 28 Aug, (e) 2227 UTC 28 Aug, (f) 0506 UTC 29 Aug, (g) 0835 UTC 29 Aug, (h) 0955 UTC 29 Aug, (i) 1305 UTC 29 Aug, (j) 1723 UTC 29 Aug, (k) 2108 UTC 29 Aug, (l) 0357 UTC 30 Aug, (m) 1226 UTC 30 Aug, and (n) 1806 UTC 30 Aug. The black circle outlines the 100-km radius. Center fixes are interpolated from NHC advisory data, with additional subjective adjustments made based on flight-level data, TDR analyses, and some of the PMW images themselves. For the overpass at 0413 UTC 28 Aug in (c), the center location is also adjusted based on TRMM PR reflectivity.
Citation: Monthly Weather Review 143, 6; 10.1175/MWR-D-14-00283.1
Best track wind speed and fractional areas of (a) PCT ≤ 250 K and (b) PCT ≤ 200 K within 100 km from the TC center from AMSR-E (asterisk), TMI (diamond), SSMIS (square), and SSM/I (triangle).
Citation: Monthly Weather Review 143, 6; 10.1175/MWR-D-14-00283.1
(a) WWLLN hourly lightning frequency within 100 km from TC center and (b) a Hovmöller plot of hourly lightning density. Radial bins are 10 km wide. Overlaid solid and dashed lines indicate the 2- and 8-km merged analysis hourly interpolated RMWs, respectively.
Citation: Monthly Weather Review 143, 6; 10.1175/MWR-D-14-00283.1
Hovmöller plots of azimuthal coverage of 85–91-GHz (a) PCT ≤ 250 K and (b) PCT ≤ 200 K. AMSR-E and TMI data are sorted into 10-km-wide bins, SSMIS data are sorted to 20-km-wide radial bins. Interpolated RMWs are as in Fig. 7.
Citation: Monthly Weather Review 143, 6; 10.1175/MWR-D-14-00283.1
Between 30 and 48 h before the onset of RI, the 200- and 250-K coverages are relatively low, with some overpasses exhibiting a complete lack of intense inner-core convection (Figs. 5a,b). By 26 h prior to RI onset, the areal coverage of both at least moderate precipitation and intense convection increases somewhat, and TMI 85-GHz PCT and PR reflectivity plan views at this time depict a ring of convection partially encircling the TC center (Figs. 5c and 9a), suggesting that an eyewall is beginning to form. A “cyan ring” (Kieper and Jiang 2012) appears in 37-GHz color composite imagery from this same overpass (Fig. 9b), indicating an inner-core precipitation organization already favorable for intensification.
Plots from the TRMM overpass at 0413 UTC 28 Aug, showing (a) TRMM reflectivity at 2 km (inner swath) and TMI 85-GHz PCT (outside of inner swath) and (b) 37-GHz color composite.
Citation: Monthly Weather Review 143, 6; 10.1175/MWR-D-14-00283.1
Coverage of intense convection once again increases between 6 and 12 h prior to RI onset (Fig. 6b), and inner-core lightning frequency peaks soon after (between 2 and 6 h before RI onset; Fig. 7a). Taken together, these observations indicate that an inner-core “convective burst” takes place in Earl 2–12 h prior to the start of RI. Perhaps more importantly, a large portion of this pre-RI burst occurs within the interpolated 2-km RMW (Figs. 7b and 8b), as well as within the 45-km-wide low-level RMW present during the 828I flight (4–9 h before RI onset; Fig. 10a). However, as the vertical shear increases to >6 m s−1 within 6–18 h of the onset of RI, deep convection becomes confined to the east and southeast (left of shear to slightly upshear left) quadrants [Figs. 5d,e; Fig. 7 in Rogers et al. (2015); Fig. 6b of Stevenson et al. (2014)]. TDR analyses from the 828I P3 flight also show that at 7 h prior to RI onset the upper-level vortex is displaced approximately 50 km to the east-southeast of the low-level center [Fig. 7c of Rogers et al. (2015)]. This misalignment is almost certainly a consequence of the moderate shear that Earl encountered later on 28 August. The earlier appearance of a cyan ring and the subsequent vortex misalignment both suggest that RI, which might have otherwise begun on 28 August, was temporarily delayed by the increased shear.
Azimuthally averaged tangential winds from the (a) 828I, (b) 829H, (c) 829I, (d) 830H, and (e) 830I merged analyses.
Citation: Monthly Weather Review 143, 6; 10.1175/MWR-D-14-00283.1
Both lightning data (Stevenson et al. 2014) and TDR reflectivity analyses (Rogers et al. 2015) show that the intense convection migrates to the northeastern and northern quadrants 4–6 h prior to the start of RI, acquiring a more decisively upshear orientation compared to the earlier hours of the convective burst (Figs. 5d,e). The TDR wind analysis presented in Fig. 7e of Rogers et al. (2015) provides some evidence that the upper-level vortex center may have precessed along with the convective burst into these upshear quadrants as well. Following the initial convective migration, PMW and lightning data indicate that the strongest inner-core convection remains predominantly upshear throughout the remainder of the convective burst period [2–4 h before RI onset; Fig. 6b in Stevenson et al. (2014)], and for at least a couple of hours following the decline in lightning frequency (Fig. 5f). This persistent upshear orientation of the convective maximum during, and immediately after the burst, is unusual as inner-core convection typically organizes downshear with a maximum in convective intensity in the downshear-left quadrant (Corbosiero and Molinari 2002; Reasor et al. 2013; DeHart et al. 2014). Likewise, if the upper-level vortex did indeed rotate into upshear quadrants, this would represent an unusual tilt orientation since both observational and modeling studies indicate that TCs tend to tilt downshear or downshear-left (Braun et al. 2006; Reasor et al. 2013). The implications of this unusual structure will be discussed in the next section.
During the ~7 h after the onset of RI, both the Hovmöller in Fig. 8a and the PMW overpasses in Fig. 5g–i indicate that the azimuthal precipitation coverage increases near the RMW relative to the period prior to the onset of RI. In approximately the same time period in which the PMW data indicate increasing precipitation symmetry, the 2- and 8-km vortex centers are nearly vertically aligned (Rogers et al. 2015; Stevenson et al. 2014). Following the appearance of an aligned vortex, the AMSR-E pass at 1723 UTC 29 August indicates continued increases in precipitation symmetry 6–12 h after RI onset (Figs. 5j and 8a). The similar timing of the vortex alignment and increasing precipitation symmetry suggests that these changes are linked to each other and to the onset of RI.
The PMW and lightning time series plots in Figs. 6b and 7a indicate that a second episode of intense inner-core convection occurs between 6 and 12 h after the onset of RI. Inner-core lightning is less frequent in this second convective burst (Figs. 7a,b), suggesting that the coverage or strength of intense convection is somewhat less than during the pre-RI burst. Also, while the pre-RI convective burst is associated with a persistent, spatially coherent convective maximum, the episode 6–12 h after RI onset appears to consist of several shorter-duration, transient features (Figs. 5i,j and 7b). However, as will be discussed in section 4, the existence of an aligned upper-level vortex probably allowed this later convective burst, although apparently weaker and less persistent, to contribute more efficiently to intensification than the earlier burst.
Between 3 and 8 h after RI onset, Earl still has a relatively broad tangential wind maximum and large RMW (80–100 km, Fig. 10b); however, within 12 h later (15–19 h after RI onset), the RMW contracts to less than half of its original size (40 km), and the wind field becomes more noticeably peaked (Fig. 10c). Not surprisingly, PMW overpasses shortly after this period indicate the development of a distinct eyewall convective ring (Fig. 5l). The eyewall contracts throughout the rest of the RI period (Figs. 5l–n), along with continued contraction of the RMW (Figs. 10d,e). With the exception of a lull in lightning around 18 h into the RI, the PMW (Fig. 8b) and lightning data (Fig. 7b) indicate that intense inner-core convection continues to occur through the later stages of RI. The PMW and lightning Hovmӧller plots also indicate that a relatively high degree of azimuthal precipitation coverage is maintained near the RMW for the remaining 18 h of RI, and intense convective events appear to occur mainly within the 8-km RMW. Both of these favorable precipitation characteristics may have aided the continued intensification of Earl into a 115-kt major hurricane.
c. Radial distribution of convection during RI
Evaluating the radial distribution of intense convection from TDR analysis vertical velocity and reflectivity data, Rogers et al. (2015) found that the strongest convection tended to occur within the low-level (2 km) RMW around the onset of RI (828I and 829H flights), as well as during the later stages of RI (829I through 830I). This result is supported by WWLLN data for the 9 h leading up to RI in Stevenson et al. (2014), and in this study for the 36-h period during RI (Fig. 11a). In contrast, observations from PMW overpasses examined (some of which occur at times with no coincident TDR data) indicate that intense convection tends to only be located within the 8-km RMW during RI, not the 2-km RMW (Fig. 12; radial distribution of PCT ≤ 200 K). Considering the limitations due to the spatial resolution of PMW data, which may result in an underestimation of the areal coverage of relatively small, transient intense convection, and the infrequency of PMW overpasses, the WWLLN data and accompanying lightning statistics (Fig. 11) are more likely to accurately capture the overall radial distribution of intense convection during RI.
Average lightning density during RI in radial bins relative to the interpolated (a) 2-km and (b) 8-km RMW.
Citation: Monthly Weather Review 143, 6; 10.1175/MWR-D-14-00283.1
Fractional areal coverage of AMSR-E and TMI 85–91-GHz PCT < 200 K during RI, in radial bins relative to the interpolated (a) 2-km and (b) 8-km RMW.
Citation: Monthly Weather Review 143, 6; 10.1175/MWR-D-14-00283.1
Regardless of which dataset better captures the radial distribution of convection, there are some times during which both PMW and lightning data indicate that the greatest coverage and frequency of inner-core intense convection occurs outside the 2-km RMW. For example, during the times between 15 and 24 h after RI onset, in which a distinct eyewall convective ring first develops (Fig. 5l), the strongest inner-core convection tends to occur outside the 2-km RMW, but within or very near the 8-km RMW (Figs. 7b and 8b). At these times, only considering the low-level RMW can lead to the false conclusion that intense convection is predominantly outside of the RMW—a location now considered less conducive to intensification.
4. Summary and conclusions
This study has documented the environmental conditions, precipitation characteristics, and kinematic structure of Hurricane Earl (2010) before and during rapid intensification (RI) using a synthesis of global analyses, passive microwave satellite data, lightning data, and P3 TDR retrieved wind and reflectivity analyses. Specifically, this comprehensive dataset has facilitated a unique opportunity to examine the time evolution of the entire RI phase of a TC, including the conditions leading up to intensification.
An examination of satellite observations and global analysis fields reveals that the environmental conditions around Earl, and within the storm’s inner core, may have been favorable for intensification as early as ~24 h before the onset of RI. However, shear increased to >6 m s−1 beginning 18 h before RI onset, presumably causing a ~50-km misalignment between the lower- and upper-level vortex centers (Rogers et al. 2015; Stevenson et al. 2014). Despite this unfavorable condition, any delay in the onset of RI due to vortex misalignment was brief.
Within several hours of the onset of RI, a sequence of processes occurred that might have helped to allow Earl to begin rapid intensification despite the moderate shear present. PMW and lightning data, as well as the radar analyses presented in Rogers et al. (2015), all indicate that a prolonged, intense “convective burst” occurred within 12 h of RI. Initially oriented left of shear, this convective burst rotated into the upshear quadrants of the inner core (Rogers et al. 2015; Stevenson et al. 2014). Passive microwave data presented in this study indicate that azimuthal precipitation coverage near the RMW increased in the first 12 h of RI. Likewise, Doppler analyses (Rogers et al. 2015; Stevenson et al. 2014) show that following the upshear convective migration; apparently coincident with an increase in precipitation symmetry, the initially tilted vortex became largely aligned by approximately 6 h after RI onset.
The convective burst before RI may have contributed to intensification through the commonly proposed processes of latent heating, forced subsidence, and vorticity generation. And while misalignment of the vortex may have initially impeded intensification, especially if the vortex was relatively aligned before the increase in shear on 28 August, the left-of-shear tilt might have been ultimately self-limiting. Results from a modeling study by Chen and Gopalakrishnan (2015) suggest that both down- and upshear-left orientations of the upper-level vortex and convective burst result in a region of enhanced warming in the upshear portions of the inner core due to a superposition of shear-induced vortex-scale descent and convective-scale adiabatic subsidence. In the simulation presented by these authors, this region of enhanced upshear adiabatic warming is advected over the low-level center by the upper-level circulation, leading to intensification and promoting vortex alignment. The upshear rotation of deep convection (along with the possible coincident precession of the upper-level vortex) may have also remedied the vertical misalignment by allowing the large-scale environmental flow to advect the convective burst as well as the entire upper-level vortex toward the low-level center, further encouraging vortex alignment (Stevenson et al. 2014).
Observational evidence presented for Hurricane Earl in this study, as well as previous examinations in Rogers et al. (2015) and Stevenson et al. (2014), strongly suggests that vortex alignment was a key ingredient in initiating RI. Vortex alignment is linked to an increase in the depth of the vorticity anomaly [see Fig. 5b in Rogers et al. (2015)], and although the strongest, most persistent convective episode appears to have occurred just before vortex alignment, the increased mid- to upper-level inner-core inertial stability associated with the aligned vortex probably allowed subsequent inner-core convective heating to enhance the warm anomaly more efficiently than previous convection.
The increasing trend in azimuthal precipitation coverage, observed in the first 12 h after the start of RI, may have contributed to intensification as well. Multistorm studies indicate a similar tendency for both elevated azimuthal precipitation coverage near RI onset (relative to lower future-intensity-change rates) and further increases in azimuthal precipitation coverage during RI (Kieper and Jiang 2012; Zagrodnik and Jiang 2014). In favorable environmental conditions, this increasing trend in precipitation symmetry appears to be tied to the formation of an organized ring of convection associated with a developing eyewall (Kieper and Jiang 2012; Harnos and Nesbitt 2011). Increasing precipitation symmetry is not a necessary condition for RI, as Molinari and Vollaro (2010) and Nguyen and Molinari (2012) have shown that in higher shear (>10 m s−1) conditions, RI can occur with limited azimuthal precipitation coverage and a highly amplified wavenumber-1 inner-core convective asymmetry. The behavior of Earl illustrates a route through which RI is still associated with increasing precipitation symmetry despite moderate shear: in Earl, a distinct eyewall convective ring does not form until 18–24 h after the start of RI; instead, the increasing trend in precipitation symmetry near RI onset appears to be intimately linked with the alignment of the vortex. However, whether increasing precipitation symmetry is a key contributor to Earl’s RI or simply a response to alignment remains unclear.
This study used passive microwave and lightning data to analyze the location of intense convection with respect to the RMW (determined by Doppler wind analyses) during RI. In agreement with Rogers et al. (2015), results indicate that the preferred location of intense convection is within the RMW during the RI of Earl, lending further support to the notion that intense convection tends to be located within the RMW in strengthening TCs (Rogers et al. 2013a). However, the current study adds an important caveat: in this type of analysis, one must consider the RMW at multiple vertical levels. Between 15 and 24 h after the onset of RI, intense convection was preferentially located outside of the 2-km RMW but within or very near the 8-km RMW, and the intensification rate did not decrease during this period. Simulations of both Earl (Chen and Gopalakrishnan 2015) and Wilma (2005; Zhang and Chen 2012) suggest that the dominant thermodynamic contribution to intensification came from upper-level warming in both of these rapidly intensifying storms. If this dependence upon upper-level warming is common, then the location of intense convection relative to the upper-level RMW might serve as a more useful indicator of future TC intensification than the location of intense convection relative to the low-level RMW.
Acknowledgments
This research was supported by NASA Grants NNX09AC44G and NNX11AB59G under the leadership of Dr. Ramesh Kakar. Drs. Paul Reasor and John Gamache at the NOAA/Hurricane Research Division (HRD) provided P3 radar dual-Doppler and reflectivity analyses, and Dr. Reasor contributed important information and instructions for the use of these datasets. We also thank the World Wide Lightning Location Network (http://wwlln.net), a collaboration among over 50 universities and institutions, for providing the lightning location data used in this study. This paper was reviewed by Drs. Kristen Corbosiero and Haiyan Jiang, and their constructive comments and recommendations resulted in significant improvements from the original manuscript.
REFERENCES
Abarca, S. F., K. L. Corbosiero, and T. J. Galarneau Jr., 2010: An evaluation of the Worldwide Lightning Location Network (WWLLN) using the National Lightning Detection Network (NLDN) as ground truth. J. Geophys. Res., 115, D18206, doi:10.1029/2009JD013411.
Abarca, S. F., K. L. Corbosiero, and D. Vollaro, 2011: The World Wide Lightning Location Network and convective activity in tropical cyclones. Mon. Wea. Rev., 139, 175–191, doi:10.1175/2010MWR3383.1.
Ashcroft, P., and F. J. Wentz, 2013: AMSR-E/Aqua L2A global swath spatially-resampled brightness temperatures, version 3. NASA DAAC at the National Snow and Ice Data Center, accessed 2012, doi:10.5067/AMSR-E/AE_L2A.003.
Black, R. A., and J. Hallett, 1999: Electrification of the hurricane. J. Atmos. Sci., 56, 2004–2028, doi:10.1175/1520-0469(1999)056<2004:EOTH>2.0.CO;2.
Braun, S. A., M. T. Montgomery, and Z. Pu, 2006: High-resolution simulation of Hurricane Bonnie (1998). Part I: The organization of eyewall vertical motion. J. Atmos. Sci., 63, 19–42, doi:10.1175/JAS3598.1.
Braun, S. A., and Coauthors, 2013: NASA’s Genesis and Rapid Intensification Processes (GRIP) field experiment. Bull. Amer. Meteor. Soc., 94, 345–363, doi:10.1175/BAMS-D-11-00232.1.
Cangiolosi, J. P, 2011: Tropical cyclone report Hurricane Earl (AL072010). National Hurricane Center, 29 pp. [Available online at http://www.nhc.noaa.gov/pdf/TCR-AL072010_Earl.pdf.]
Cecil, D. J., E. J. Zipser, and S. W. Nesbitt, 2002: Reflectivity, ice scattering, and lightning characteristics of hurricane eyewalls and rainbands. Part I: Quantitative description. Mon. Wea. Rev., 130, 769–784, doi:10.1175/1520-0493(2002)130<0769:RISALC>2.0.CO;2.
Chen, H., and D.-L. Zhang, 2013: On the rapid intensification of Hurricane Wilma (2005). Part II: Convective bursts and the upper-level warm core. J. Atmos. Sci., 70, 146–162, doi:10.1175/JAS-D-12-062.1.
Chen, H., and S. Gopalakrishnan, 2015: A study on the asymmetric rapid intensification of Hurricane Earl (2010) using the HWRF system. J. Atmos. Sci., 72, 531–550, doi:10.1175/JAS-D-14-0097.1.
Corbosiero, K. L., and J. Molinari, 2002: The effects of vertical wind shear on the distribution of convection in tropical cyclones. Mon. Wea. Rev., 130, 2110–2123, doi:10.1175/1520-0493(2002)130<2110:TEOVWS>2.0.CO;2.
Cram, T., J. Persing, M. Montgomery, and S. Braun, 2007: A Lagrangian trajectory view on transport and mixing processes between the eye, eyewall, and environment using a high-resolution simulation of Hurricane Bonnie (1998). J. Atmos. Sci., 64, 1835–1856, doi:10.1175/JAS3921.1.
DeHart, J. C., R. A. Houze Jr., and R. F. Rogers, 2014: Quadrant distribution of tropical cyclone inner-core kinematics in relation to environmental shear. J. Atmos. Sci., 71, 2713–2732, doi:10.1175/JAS-D-13-0298.1.
DeMaria, M., 1996: The effect of vertical shear on tropical cyclone intensity change. J. Atmos. Sci., 53, 2076–2088, doi:10.1175/1520-0469(1996)053<2076:TEOVSO>2.0.CO;2.
DeMaria, M., and J. Kaplan, 1994: Sea surface temperature and the maximum intensity of Atlantic tropical cyclones. J. Climate, 7, 1324–1334, doi:10.1175/1520-0442(1994)007<1324:SSTATM>2.0.CO;2.
Dunion, J. P., and C. S. Velden, 2004: The impact of the Saharan air layer (SAL) on Atlantic tropical cyclone activity. Bull. Amer. Meteor. Soc., 85, 353–364, doi:10.1175/BAMS-85-3-353.
Emanuel, K., C. Desautels, C. Holloway, and R. Korty, 2004: Environmental control of tropical cyclone intensity. J. Atmos. Sci., 61, 843–858, doi:10.1175/1520-0469(2004)061<0843:ECOTCI>2.0.CO;2.
Frank, W. M., and E. A. Ritchie, 2001: Effects of vertical wind shear on the intensity and structure of numerically simulated hurricanes. Mon. Wea. Rev., 129, 2249–2269, doi:10.1175/1520-0493(2001)129<2249:EOVWSO>2.0.CO;2.
Gamache, J. F., F. D. Marks, and F. Roux, 1995: Comparison of three airborne Doppler sampling techniques with airborne in situ wind observations in Hurricane Gustav (1990). J. Atmos. Oceanic Technol., 12, 171–181, doi:10.1175/1520-0426(1995)012<0171:COTADS>2.0.CO;2.
Guimond, S. R., G. M. Heymsfield, and F. J. Turk, 2010: Multiscale observations of Hurricane Dennis (2005): The effects of hot towers on rapid intensification. J. Atmos. Sci., 67, 633–654, doi:10.1175/2009JAS3119.1.
Harnos, D. S., and S. W. Nesbitt, 2011: Convective structure in rapidly intensifying tropical cyclones as depicted by passive microwave measurements. Geophys. Res. Lett., 38, L07805, doi:10.1029/2011GL047010.
Hendricks, E. A., M. S. Peng, B. Fu, and T. Li, 2010: Quantifying environmental control on tropical cyclone intensity change. Mon. Wea. Rev., 138, 3243–3271, doi:10.1175/2010MWR3185.1.
Houze, R. A., W. C. Lee, and M. M. Bell, 2009: Convective contribution to the genesis of Hurricane Ophelia (2005). Mon. Wea. Rev., 137, 2778–2800, doi:10.1175/2009MWR2727.1.
Hutchins, M. L., R. H. Holzworth, C. J. Rodger, S. Heckman, and J. B. Brundell, 2012: WWLLN absolute detection efficiencies and the global lightning source function. Proc. EGU General Assembly 2012, Vienna, Austria, European Geosciences Union, Vol. 14, EGU2012-12917. [Available online at http://meetingorganizer.copernicus.org/EGU2012/EGU2012-12917.pdf.]
Jacobson, A. R., R. Holzworth, J. Harlin, R. Dowden, and E. Lay, 2006: Performance assessment of the World Wide Lightning Location Network (WWLLN) using the Los Alamos Sferic Array (LASA) array as ground truth. J. Atmos. Oceanic Technol., 23, 1082–1092, doi:10.1175/JTECH1902.1.
Jiang, H., 2012: The relationship between tropical cyclone intensity change and the strength of inner-core convection. Mon. Wea. Rev., 140, 1164–1176, doi:10.1175/MWR-D-11-00134.1.
Jiang, H., and E. M. Ramirez, 2013: Necessary conditions for tropical cyclone rapid intensification as derived from 11 years of TRMM data. J. Climate, 26, 6459–6470, doi:10.1175/JCLI-D-12-00432.1.
Kaplan, J., and M. DeMaria, 2003: Large-scale characteristics of rapidly intensifying tropical cyclones on the North Atlantic basin. Wea. Forecasting, 18, 1093–1108, doi:10.1175/1520-0434(2003)018<1093:LCORIT>2.0.CO;2.
Kaplan, J., M. DeMaria, and J. A. Knaff, 2010: A revised tropical cyclone rapid intensification index for the Atlantic and eastern North Pacific basins. Wea. Forecasting, 25, 220–241, doi:10.1175/2009WAF2222280.1.
Kelley, O. A., J. Stout, and J. B. Halverson, 2004: Tall precipitation cells in tropical cyclones eyewalls are associated with tropical cyclone intensification. Geophys. Res. Lett., 31, L24112, doi:10.1029/2004GL021616.
Kieper, M., and H. Jiang, 2012: Predicting tropical cyclone rapid intensification using the 37 GHz ring pattern identified from passive microwave measurements. Geophys. Res. Lett., 39, L13804, doi:10.1029/2012GL052115.
Liu, C., E. J. Zipser, D. J. Cecil, S. W. Nesbitt, and S. Sherwood, 2008: A cloud and precipitation feature database from nine years of TRMM observations. J. Appl. Meteor., 47, 2712–2728, doi:10.1175/2008JAMC1890.1.
Liu, C., D. Cecil, and E. J. Zipser, 2011: Relationships between lightning flash rates and passive microwave brightness temperatures at 85 and 37 GHz over the tropics and subtropics. J. Geophys. Res., 116, D23108, doi:10.1029/2011JD016463.
Marks, F. D., and L. K. Shay, 1998: Landfalling tropical cyclones: Forecast problems and associated research opportunities. Bull. Amer. Meteor. Soc., 79, 305–323, doi:10.1175/1520-0477(1998)079<0305:LTCFPA>2.0.CO;2.
Merrill, R. T., 1988: Environmental influences on hurricane intensification. J. Atmos. Sci., 45, 1678–1687, doi:10.1175/1520-0469(1988)045<1678:EIOHI>2.0.CO;2.
Molinari, J., and D. Vollaro, 2010: Rapid intensification of a sheared tropical storm. Mon. Wea. Rev., 138, 3869–3885, doi:10.1175/2010MWR3378.1.
National Hurricane Center, 2014: National Hurricane Center forecast verification. [Available online at http://www.nhc.noaa.gov/verification/verify5.shtml.]
Nguyen, L. T., and J. Molinari, 2012: Rapid intensification of a sheared, fast-moving hurricane over the Gulf Stream. Mon. Wea. Rev., 140, 3361–3378, doi:10.1175/MWR-D-11-00293.1.
Pendergrass, A. G., and H. E. Willoughby, 2009: Diabatically induced secondary flows in tropical cyclones. Part I: Quasi-steady forcing. Mon. Wea. Rev., 137, 805–821, doi:10.1175/2008MWR2657.1.
Reasor, P. D., M. D. Eastin, and J. F. Gamache, 2009: Rapidly intensifying Hurricane Guillermo (1997). Part I: Low-wavenumber structure and evolution. Mon. Wea. Rev., 137, 603–631, doi:10.1175/2008MWR2487.1.
Reasor, P. D., R. Rogers, and S. Lorsolo, 2013: Environmental flow impacts on tropical cyclone structure diagnosed from airborne Doppler radar composites. Mon. Wea. Rev., 141, 2949–2969, doi:10.1175/MWR-D-12-00334.1.
Reinhart, B., and Coauthors, 2014: Understanding the relationships between lightning, cloud microphysics, and airborne radar-derived storm structure during Hurricane Karl (2010). Mon. Wea. Rev., 142, 590–605, doi:10.1175/MWR-D-13-00008.1.
Riemer, M., M. T. Montgomery, and M. E. Nicholls, 2010: A new paradigm for intensity modification of tropical cyclones: Thermodynamic impact of vertical wind shear on the inflow layer. Atmos. Chem. Phys., 10, 3163–3188, doi:10.5194/acp-10-3163-2010.
Rodger, C. J., S. Werner, J. B. Brundell, E. H. Lay, N. R. Thomson, R. H. Holzworth, and R. L. Dowden, 2006: Detection efficiency of the VLF World-Wide Lightning Location Network (WWLLN): Initial case study. Ann. Geophys., 24, 3197–3214, doi:10.5194/angeo-24-3197-2006.
Rogers, R., and Coauthors, 2006: The Intensity Forecasting Experiment: A NOAA multiyear field program for improving tropical cyclone intensity forecasts. Bull. Amer. Meteor. Soc., 87, 1523–1537, doi:10.1175/BAMS-87-11-1523.
Rogers, R., P. Reasor, and S. Lorsolo, 2013a: Airborne Doppler observations of the inner-core structural differences between intensifying and steady-state tropical cyclones. Mon. Wea. Rev., 141, 2970–2991, doi:10.1175/MWR-D-12-00357.1.
Rogers, R., and Coauthors, 2013b: NOAA’s Hurricane Intensity Forecasting Experiment: A progress report. Bull. Amer. Meteor. Soc., 94, 859–882, doi:10.1175/BAMS-D-12-00089.1.
Rogers, R., P. Reasor, and J. Zhang, 2015: Multiscale structure and evolution of Hurricane Earl (2010) during rapid intensification. Mon. Wea. Rev., 143, 536–562, doi:10.1175/MWR-D-14-00175.1.
Rudlosky, S. D., and D. T. Shea, 2013: Evaluating WWLLN performance relative to TRMM/LIS. Geophys. Res. Lett., 40, 2344–2348, doi:10.1002/grl.50428.
Sanger, N. T., M. T. Montgomery, R. K. Smith, and M. M. Bell, 2014: An observational study of tropical cyclone spinup in Supertyphoon Jangmi (2008) from 24 to 27 September. Mon. Wea. Rev., 142, 3–28, doi:10.1175/MWR-D-12-00306.1.
Sapiano, M., W. Berg, D. McKague, and C. Kummerow, 2013: Towards an intercalibrated fundamental climate data record of the SSM/I Sensors. IEEE Trans. Geosci. Remote Sens., 51, 1492–1503, doi:10.1109/TGRS.2012.2206601.
Schubert, W. H., and J. J. Hack, 1982: Inertial stability and tropical cyclone development. J. Atmos. Sci., 39, 1687–1697, doi:10.1175/1520-0469(1982)039<1687:ISATCD>2.0.CO;2.
Spencer, R. W., H. M. Goodman, and R. E. Hood, 1989: Precipitation retrieval over land and ocean with the SSM/I: Identification and characteristics of the scattering signal. J. Atmos. Oceanic Technol., 6, 254–273, doi:10.1175/1520-0426(1989)006<0254:PROLAO>2.0.CO;2.
Steranka, J., E. B. Rodgers, and R. C. Gentry, 1986: The relationship between satellite measured convective bursts and tropical cyclone intensification. Mon. Wea. Rev., 114, 1539–1546, doi:10.1175/1520-0493(1986)114<1539:TRBSMC>2.0.CO;2.
Stevenson, S. N., K. L. Corbosiero, and J. Molinari, 2014: The convective evolution and rapid intensification of Hurricane Earl (2010). Mon. Wea. Rev., 142, 4364–4380, doi:10.1175/MWR-D-14-00078.1.
Tang, B., and K. Emanuel, 2010: Midlevel ventilation’s constraint on tropical cyclone intensity. J. Atmos. Sci., 67, 1817–1830, doi:10.1175/2010JAS3318.1.
Van Sang, N., R. K. Smith, and M. T. Montgomery, 2008: Tropical-cyclone intensification and predictability in three dimensions. Quart. J. Roy. Meteor. Soc., 134, 563–582, doi:10.1002/qj.235.
Vigh, J. L., and W. H. Schubert, 2009: Rapid development of the tropical cyclone warm core. J. Atmos. Sci., 66, 3335–3350, doi:10.1175/2009JAS3092.1.
Zagrodnik, J. P., and H. Jiang, 2014: Rainfall, convection, and latent heating distributions in rapidly intensifying tropical cyclones. J. Atmos. Sci., 71, 2789–2809, doi:10.1175/JAS-D-13-0314.1.
Zhang, D.-L., and H. Chen, 2012: Importance of the upper-level warm core in the rapid intensification of a tropical cyclone. Geophys. Res. Lett., 39, L02806, doi:10.1029/2011GL050578.
Zipser, E. J., and K. R. Lutz, 1994: The vertical profile of radar reflectivity of convective cells: A strong indicator of storm intensity and lightning probability? Mon. Wea. Rev., 122, 1751–1759, doi:10.1175/1520-0493(1994)122<1751:TVPORR>2.0.CO;2.