914 WEATHER AND FORECASTING VOLUME 30

Diagnosing the Conditional Probability of Tornado Damage Rating Using

Environmental and Radar Attributes

BRYAN T. SMITH, RICHARD L. THOMPSON, ANDREW R. DEAN, AND PATRICK T. MARSH

NOAA/NWS/NCEP/Storm Prediction Center, Norman, Oklahoma

(Manuscript received 2 October 2014, in final form 22 April 2015)

ABSTRACT

Radar-identified convective modes, peak low-level rotational velocities, and near-storm environmental data were
assigned to a sample of tornadoes reported in the contiguous United States during 2009–13. The tornado segment data
were filtered by the maximum enhanced Fujita (EF)-scale tornado event per hour using a 40-km horizontal grid.
Convective mode was assigned to each tornado event by examining full volumetric Weather Surveillance Radar-1988
Doppler data at the beginning time of each event, and 0.58 peak rotational velocity (Vrot) data were identified
manually during the life span of each tornado event. Environmental information accompanied each grid-hour event,
consisting primarily of supercell-related convective parameters from the hourly objective mesoscale analyses calculated
and archived at the Storm Prediction Center. Results from examining environmental and radar attributes, featuring the
significant tornado parameter (STP) and 0.58 peak Vrot data, suggest an increasing conditional probability for greater
EF-scale damage as both STP and 0.58 peak Vrot increase, especially with supercells. Possible applications of these
findings include using the conditional probability of tornado intensity as a real-time situational awareness tool.

1. Introduction Convective mode is an additional component widely

recognized as a contributor to the occurrence and
Considerable effort in recent decades has focused on near-
nonoccurrence of severe weather. Recent work by
storm environment interrogation via observed soundings
Smith et al. (2012, hereafter S12) demonstrated re-
(e.g., Rasmussen and Blanchard 1998; Rasmussen 2003;
lationships of convective mode and storm-scale rota-
Craven and Brooks 2004), model-based planar fields (e.g.,
tion (when applicable) to tornado damage intensity, and
Stensrud et al. 1997), and model-based proximity soundings
the second part of that study, by Thompson et al. (2012,
in order to discriminate between nontornadic and significant
hereafter T12), went a step further and investigated the re-
[rated as category 2 or greater on the Fujita scale ($F2)]
lationships between the near-storm environment and these
tornado environments for supercells (e.g., Thompson et al.
storm attributes.
2003, 2007; Davies 2004; Davies and Fischer 2009). These
The infusion of real-time diagnostic parameters, like
investigations provided empirical evidence supporting the
the National Oceanic and Atmospheric Administration/
importance of several measures of moisture, buoyancy, and
National Weather Service/Storm Prediction Center’s
vertical wind shear for producing significant tornadoes
(NOAA/NWS/SPC) hourly mesoanalysis (Bothwell
with supercells, as reflected in the development of super-
et al. 2002), can contribute to greater awareness of
cell ingredients–based composite parameters [e.g., signifi-
potential tornado risk in an operational forecast and
cant tornado parameter (STP1); Thompson et al. (2003)].
warning setting. Magsig (2008) discussed techniques
for diagnosing radar-based storm attributes and in-
tegrating environmental information into the warning
1
STP effective-layer calculation (T12) only discussed herein, for decision-making process. Recent work by Brotzge et al.
the nearest hour preceding each event. (2013) revealed NWS tornado warning performance, as
measured by probability of detection, was maximized
for the more intense tornado events [i.e., higher en-
Corresponding author address: Bryan T. Smith, NOAA/NWS/
NCEP/Storm Prediction Center, 120 David L. Boren Blvd., Ste. hanced Fujita (EF)-scale damage ratings] when the
2300, Norman, OK 73072. tornadoes were produced by discrete supercells with
E-mail: [email protected] strong mesocyclones, close to the radar site, and in

DOI: 10.1175/WAF-D-14-00122.1

Ó 2015 American Meteorological Society

AUGUST 2015 SMITH ET AL. 915

environments strongly supportive of tornadic super- Though forecasts of tornado intensity, on the spatio-
cells. Real-time utilization of the multicomponent temporal scales of tornado warnings, remain a daunting
datasets described in T12 and Brotzge et al. (2013) may task, this study strives to construct a tornado database
contribute to the improved situational awareness of that can provide diagnostic information on tornado in-
tornado potential. tensity (as inferred by EF-scale damage ratings), given
In light of recent tornado disasters (e.g., 27 April and that a tornado has developed (i.e., conditional proba-
22 May 2011, among others), contemporary efforts bility). This study builds upon previous work by S12 and
within the NWS, as part of the Weather Ready Nation T12 by further developing a multicomponent dataset, which
initiative, have sought to better assess tornado vulner- includes 0.58 peak rotational velocity (Vrot) information,
ability and better communicate impact-based hazard rather than mesocyclone nomograms (Andra 1997; Stumpf
information (i.e., tornado risk) to the public. One such et al. 1998), to assess storm-scale rotation strength. In addi-
example involves NWS Central Region local forecast tion, other classifiable circulations [e.g., mesovortex; Trapp
offices tasked with issuing experimental impact-based and Weisman (2003)] and their strengths were also exam-
warnings (IBWs; Wagenmaker et al. 2014) for severe ined. This study advocates combining near-storm environ-
thunderstorms and tornadoes. IBWs are intended to ment information and a relatively simple, real-time radar
convey the potential impact to life and property within diagnosis to better assess the maximum conditional tornado
the disseminated warning text based on the predicted intensity risk—a necessary step in improving both the con-
intensity of the severe hazard (e.g., tornado). Future sistency of tornado warnings and near-term forecasts of
warn-on-forecast (Stensrud et al. 2009) work will likely tornado intensity.
include explicit probabilistic information in addition
to the binary tornado warning decision point in
current use. 2. Data and methodology
Developmental work from recent studies (e.g.,
a. Data and event filtering
Kingfield et al. 2012; LaDue et al. 2012) examined the
relationship between storm-scale circulation algo- Radar-identified convective modes, peak low-level
rithms using Weather Surveillance Radar-1988 Dopp- rotational velocities, and near-storm environmental
ler (WSR-88D) and EF-scale damage ratings in a data were assigned to a sample of tornadoes reported in
diagnostic manner. A manual user-defined maximum the CONUS during the 2009–13 period, which corre-
low-level velocity difference (LLVD) was found by sponds with most of the WSR-88D superresolution
LaDue et al. (2012) to exhibit a stronger linear re- data era (Torres and Curtis 2007). The tornado seg-
lationship to EF-scale rating on a small number of high- ment data were filtered by the maximum EF-scale
resolution tornado damage surveys compared to an tornado damage rating per hour on a 40-km horizon-
automated LLVD approach using either the mesocy- tal grid, and after additional filtering described herein
clone detection algorithm (MDA; Stumpf et al. 1998) (section 2c), yielded a total of 4770 tornado grid-hour
or the tornado detection algorithm (TDA; Mitchell events (hereafter tornado events). Convective mode
et al. 1998). Newman et al. (2013) found utility in ap- was assigned to each tornado event via manual exam-
plying range correction to the local, linear least squares ination of full volumetric WSR-88D data (section 2b)
derivatives (LLSDs; Smith and Elmore 2004) azi- at the beginning time of each event, and 0.58 peak Vrot
muthal shear algorithm, and this procedure aided in was determined manually using superresolution radar
differentiating between nontornadic and tornadic ra- data during the life span of each tornado event (section
dar scans for a small number of events. Blair and 2c). Environmental information, consisting primarily
Leighton (2014) noted the need for robust, scientific of supercell-related convective parameters from the
guidance for real-time tornado intensity estimates in hourly SPC objective analyses, accompanied each
their assessment of event confirmation in NWS warn- tornado event.
ings and statements across the central continental Within the framework described above, the authors
United States (CONUS) from 2007 to 2011. The early made careful manual adjustments to a small portion
studies investigating the relationship between tornado (7.9%) of the database. Many of the suspected report
intensity and radar (e.g., Kingfield et al. 2012; LaDue errors involved incorrectly listed report times, as de-
et al. 2012; Toth et al. 2013) have shown some ability to termined by time matching the reports to radar data.
identify different levels of tornado intensity in a di- Examples of this suspected error type included reports
agnostic manner, thereby lending credence to the un- well removed from existing radar echoes and time dis-
derlying idea of tornado intensity identification, from placed on the order of tens of minutes to an hour or
which IBW is based. more. In situations where a suspected error could not be
916 WEATHER AND FORECASTING VOLUME 30

corrected easily, the NCDC publication Storm Data peak rotational velocity values for the 1-, 2-, and 3.5-nautical
was used to examine the description of the question- mile (n mi; 1 n mi 5 1.852 km) mesocyclone nomograms.
able reports in an effort to identify the storm re- Storms that exhibited persistent, weak azimuthal shear just
sponsible for the event. Despite alleviating most errors, below the nomogram’s minimal mesocyclone threshold
small time discrepancies on the order of one or two and transient supercell reflectivity structure, or identifiable
volume scans (51% of all events exhibiting error had rotation (regardless of magnitude) for no more than two
time displacement errors #10 min) were found in consecutive volume scans (i.e., ,10 min), were binned in
Storm Data between the beginning time of a tornado the marginal supercell (i.e., other) category.
event and pertinent WSR-88D velocity signatures, A QLCS is defined as consisting of contiguous
similar to a finding by French et al. (2013) using a reflectivity at or above the threshold of 35 dBZ for a
mobile radar. Unless a time or location change was horizontal distance of at least 100 km and a length-to-
necessary based on a well-resolved circulation, we de- width aspect ratio of at least 3:1 at the time of the event,
ferred to the NWS-documented begin time and loca- similar to Trapp et al. (2005). Other modes included
tion, in an attempt to account for uncertainty and disorganized cellular modes that did not include super-
variability in distance between the tornado location cell structures (e.g., single cell, multicell) and consisted
relative to the WSR-88D circulation location (e.g., mainly of conglomerates meeting the reflectivity
Speheger and Smith 2006). threshold but not satisfying either supercell or QLCS
criteria (e.g., short line segment). Additionally, storms
b. Radar-based storm mode classification criteria
exhibiting transient (i.e., one or two volume scans) ro-
The Gibson Ridge Level II Analyst radar-viewing tation below supercell rotation criteria were assigned to
software (http://www.grlevelx.com/) was used to analyze the other modes category. For a more thorough discus-
archived WSR-88D level-II single-site radar data (Crum sion pertaining to the complexity and challenges of
et al. 1993) from the National Climatic Data Center categorizing convective mode, please refer to S12.
(http://www.ncdc.noaa.gov/nexradinv/) using the closest
c. Low-level rotational velocities
radar (within 101 mi) to classify convective mode based
on S12. Convective mode was determined using full Peak inbound and outbound velocities were examined
volumetric radar data, especially when data through a for each volume scan from immediately prior to tornado
deep layer were needed to perform a more thorough formation through tornado dissipation. Only combina-
assessment of storm structure. Convective mode was tions of velocity maxima exhibiting cyclonic or anticy-
assigned based on the volume scan and lower-elevation clonic azimuthal shear within 5 n mi and #458 angle
tilts (e.g., 0.58) of base reflectivity immediately prior to from one another were considered, to avoid primarily
the time of the tornado event. Emphasis herein is placed convergent or divergent signatures. The maximum peak
on the three major convective mode classes of tornadic rotational velocity [Vrot 5 (jVinj 1 jVoutj)/2], from all
storms: supercells2 (3392 events), quasi-linear convec- volume scans was assigned to each tornadic event
tive systems (QLCSs; 894 events), and disorganized (Fig. 1), and only tornado events sampled at or below
cells/clusters and marginal supercells (484 events; 10 000 ft (i.e., #101-mi range) height above radar level
hereafter referred to as other modes). (ARL) were analyzed and included in this study (Fig. 2).
Discrete or embedded cells with focused areas of cy- Although the peak Vrot only uses a pair of data points
clonic (or anticyclonic) azimuthal shear were further that can be influenced by errors due to aliasing or ‘‘noisy
scrutinized as potential supercells, following the meso- data’’ (Wood and Brown 1997), this dataset considers
cyclone nomograms developed using 4-bit radar data multiple possible pairs of peak velocity data for indi-
[after Andra (1997) and Stumpf et al. (1998)]. Supercells vidual volume scans during the tornado’s lifetime. This
required a peak rotational velocity $10 m s21 (i.e., a approach can effectively reduce the influence of any
peak-to-peak azimuthal velocity difference of roughly volume scan(s) with potential data errors by defaulting
20 m s21 over a distance of less than ;7 km), rotation $ to other candidate volume scans. Concerns such as radar
1/ 4 the depth of the storm, and rotation duration of at beam placement relative to the tornado circulation were
least 10–15 min. Circulations were classified as weak partially mitigated by the large sample size of tornado
shear (nonsupercell), and weak, moderate, or strong events and by two or more volume scans per tornado
supercells, following the range-dependent horizontal event. Underestimates of 0.58 peak Vrot owing to beam
offset would likely be applied randomly throughout the
dataset. Ancillary data such as the time of the volume
2
Includes right-moving supercells (3384) and left-moving scan and a subjective binary assessment of a clear/tight
supercells (8). circulation were also found for the majority of tornado
AUGUST 2015 SMITH ET AL. 917

FIG. 1. (a) WSR-88D base reflectivity (dBZ; color scale on left) at 0.58 beam tilt from Jackson, MS (KDGX), at 0852 UTC 30 Nov 2010.
A cell-in-cluster supercell produced an EF2 tornado in Smith County, MS (start time 0844 UTC). North is up, county borders are black,
and distance scale is at lower right. (b) As in (a), but for storm-relative velocity (kt; color scale on left), 458 angle insert, and curved arrows
signifying rotation. Denoted inserts display maximum inbound storm-relative velocity (max Vin, 48.6 kt), maximum outbound storm-
relative velocity (max Vout, 30.1 kt), 0.58 peak Vrot (39.4 kt), and velocity sampled height ARL (1800 ft).

events (3690 of 4770). Finally, the sampling of circu- WSR-88D to resolve storm-scale circulations. Unlike
lations by ARL (nearest 100 ft) using the highest radar Toth et al. (2013) and LaDue et al. (2012), velocity data
bin between the two peak Vrot data points was docu- were not dealiased manually beyond the existing deal-
mented in order to account for the effects of radar iasing algorithm capability for several reasons: 1) our peak
beam widening with range that reduce the ability of the Vrot method is easily reproduced in real-time forecast and

FIG. 2. Spatial plot of tornado events sampled at #10 000 ft ARL.

918 WEATHER AND FORECASTING VOLUME 30

FIG. 3. Box-and-whiskers plot of 0.58 peak Vrot (kt) of EF0–EF5 tornado events (2009–13;
at #10 000 ft ARL, with 1–101-mi radius) grouped by supercell (Sup; dark gray), QLCS (light
gray; EF3 events not shown), and other modes (Other; white). The shaded boxes span the 25th–
75th percentiles, and the whiskers extend upward to the 90th and downward to the 10th per-
centiles. Median values are marked within the box, and sample sizes for each storm mode and
EF-scale category are shown in parentheses.

warning operations with short time constraints and Although tornado circulations appeared to be re-
2) the impact of not dealiasing a small fraction of tor- solved explicitly in a few cases with large tornadoes close
nado velocity signatures is likely minimized by the to the radar site, an overwhelming majority of WSR-
large size of this sample (4770 tornado events). The 88D velocity signatures were representative of the larger
velocity dealiasing algorithm technique used by Gibson tornadic vortex (Mitchell et al. 1998) or the low-level
Ridge Analyst software is similar to current and legacy mesocyclone (Stumpf et al. 1998). A relatively small
dealiasing techniques for WSR-88D data (e.g., Eilts percentage of available cases (11%) consisted of 0.58
and Smith 1990; Zittel and Jing 2012). peak Vrot diameters exceeding 3.5 mi, which is clearly

FIG. 4. NEXRAD coverage at or above 3000, 6000, and 10 000 ft or less AGL (ROC 2014).
The level refers to the center of the beam height (assuming standard atmospheric refraction).
Terrain blockage indicated where 50% or more of the beam is blocked.
AUGUST 2015 SMITH ET AL. 919

FIG. 5. As in Fig. 3, but for tornado events at 100–2900 ft ARL, with 1–42-mi radius.

larger than any documented tornado diameter. While scan was recorded. A 5–10-knot (kt; 1 kt 5 0.51 m s21)
many 0.58 peak Vrot cases were easily assessed, 0.58 peak difference in 0.58 peak Vrot was typical between
Vrot identification at times was a challenging task and seemingly erroneous Vrot and the Vrot recorded, and
involved considerable effort and uncertainty in assign- resulted in a reduced value than otherwise would have
ing the peak inbound and peak outbound values. A been assigned. If a tight circulation couplet (i.e., likely
neighborhood approach was effectively used on a small resolving the tornado vortex) was clearly separate from
subset of events (;3.6%) because of increased un- other nearby higher-velocity bins, the velocity data as-
certainty in assigning the 0.58 peak Vrot. Apparent bad sociated with the smaller-scale circulation were prefer-
radials in velocity data were not used and other nearby entially recorded; otherwise, preference was given to
velocity bins were used instead. Other questionable recording velocity information within the larger-scale
velocity signatures included noisy data, which clearly circulation if the outer circulation Vrot value was more
suffered from dealiasing problems. These difficult-to- than 5 kt greater than the candidate Vrot value of the
assign cases were in most situations assigned peak in- inner circulation.
bound and outbound velocity values nearby. Less often, The manual analysis of velocity data presented here
the rotational velocity from the next highest volume is similar to techniques used in real-time warning

FIG. 6. As in Fig. 3, but for tornado events at 3000–5900 ft ARL, with 42–70-mi radius.
920 WEATHER AND FORECASTING VOLUME 30

FIG. 7. As in Fig. 3, but for tornado events at 6000–10 000 ft ARL, with 70–101-mi radius.

decision-making. The subjective analysis used to diagnose occasion (especially above the ground) and large
circulation strength can be advantageous compared to an enough to be of concern regarding expected storm
automated objective approach, especially in cases when evolution. The RUC model was replaced by the Rapid
radar algorithms do not resolve some tornadic circula- Refresh (RAP) model in May 2012, though compari-
tions [e.g., landspouts; Brady and Szoke (1989)] be- sons of the RUC and RAP in severe storm environ-
cause of resolution limitations, or when circulations are ments are lacking in the formal literature. Laflin
misidentified along squall lines aligned along the (2013) examined vertical profiles of temperature and
radar beam. moisture for rawinsonde observations and RAP
While it was common for velocity signatures to vary model soundings and quantified differences in terms
during the life cycle of the tornado event, the tornado of buoyancy [e.g., lowest 100-hPa mean-layer con-
events in this sample rarely had one outlier volume vective available potential energy (MLCAPE)] with a
scan at 0.58 tilt with much stronger Vrot compared to convective-related focus on the preconvective boundary
the other sampled volume scans. Many of the higher- layer. Substantial errors were found in RAP 6- and 12-h
end tornado cases exhibited consistent velocity values forecasts of boundary layer moisture, which resulted
that were just below the peak Vrot value at least for in underestimates of buoyancy (e.g., surface-based
several volume scans, including a substantial part of CAPE errors approaching 1000 J kg21) in dry, well-
the tornado segment grid hour (i.e., tornado event). mixed environments. Yet Laflin (2013) covered only a
Although there was a strong correspondence between limited domain (six Great Plains rawinsonde sites)
the highest EF-scale rating and the maximum 0.58 during 7 weeks in the late spring 2012, and the findings
peak Vrot, the two did not necessarily match in time may not be representative of 0- and 1-h RAP sound-
and space. ings used in the SPC objective analyses, or of other
environments supportive of tornadoes (e.g., Thompson
d. Quality of SPC hourly mesoscale analyses
et al. 2013).
Rapid Update Cycle (RUC; Benjamin et al. 2004) Mesoscale observations, such as trends in WSR-88D
model 0- and 1-h forecasts provided the basis for the vertical wind profile data, may aid situational aware-
SPC hourly mesoscale analyses from January 2003 ness, especially for cases when background objective
through April 2012. Coniglio (2012) evaluated the hourly guidance differs substantially from the strength
SPC hourly objective analyses via VORTEX2 field of radar-derived wind fields (Guyer and Hart 2012).
project soundings from the springs of 2009 and 2010 Potvin et al. (2010) discussed proximity sounding
across the Great Plains and found that the SPC ana- sensitivity to spatiotemporal distance from an event.
lyses improved upon the background 1-h RUC model When examining 0–1-h proximity sounding data,
forecasts of surface temperature and dewpoint tem- Potvin et al. (2010) identified a zone within 40–80 km
perature, as well as many derived thermodynamic of the launch site (see their Fig. 5) that served to
variables. However, errors were still substantial on best characterize the near-storm environment (e.g.,
AUGUST 2015 SMITH ET AL. 921

TABLE 1. Mean (median) values of 0.58 peak Vrot and STP80km for supercell (Sup) and QLCS tornado events by EF-scale class.

Sup EF0 Sup EF1 Sup EF2 Sup EF3 Sup EF41 QLCS EF0 QLCS EF1 QLCS EF2
Vrot 35 (34) 41 (40) 51 (50) 65 (63) 76 (73) 30 (30) 35 (34) 39 (38)
STP80km 2.7 (1.9) 3.3 (2.6) 4.0 (3.4) 5.5 (4.6) 8.4 (8.4) 1.6 (1.0) 2.1 (1.7) 2.6 (2.2)

minimize convective feedback effects, maintain close 3. Results

distance). The findings of Potvin et al. (2010) were
a. 0.58 peak Vrot
reinforced by Parker (2014), who examined the spa-
tiotemporal variability of VORTEX2 field project A strong relationship exists between 0.58 peak Vrot
soundings relative to both tornadic and nontornadic and EF scale for all convective modes. For higher EF-
supercells. Parker (2014) noted that pronounced dif- scale ratings, an increase in the 0.58 peak Vrot distribu-
ferences in environmental characteristics extended tion occurred (Fig. 3). Prior studies (e.g., Wood and
beyond the storm-induced inflow region, with more Brown 1997; Newman et al. 2013) have documented the
favorable combinations of low-level moisture and dependence of decreased circulation resolution as a
vertical wind shear evident well away from a small function of increased radar range (and height ARL).
sample of tornadic supercells compared to non- Therefore, 0.58 peak Vrot data used to sample tornado
tornadic supercells. Still, variability in the near-storm events were rounded to the nearest 100 ft ARL and
environment was substantial, and a single proximity separated into three ARL (radar range) groups: 100–
sounding is not necessarily reflective of supercell tor- 2900, 3000–5900, and 6000–10 000 ft (Fig. 4). A largely
nado potential. monotonic increase for 0.58 peak Vrot is displayed for
This study utilized the maximum neighborhood tornado events as the supercell EF scale increases for
grid-hour value within 80 km of each tornado event events sampled below 6000 ft ARL (Figs. 5 and 6), or
for STP (hereafter STP80km; T12) to account for within 70 mi of the radar site.3 For events sampled at
proximity concerns and the spatial variability of en- 6000–10 000 ft ARL (Fig. 7), little difference in the dis-
vironmental parameters while providing a relatively tribution is evident among EF3 and EF4 supercells or
simple characterization of the tornado environments greater (hereafter EF41) events. Comparing the su-
that were dominated by supercells. The maximum percell EF41 events among the three radar ARL groups
neighborhood approach reflects the ability of the (Figs. 5–7) shows a decrease in 0.58 peak Vrot magnitudes
operational meteorologist to consider more than a at the 10th, 25th, 50th, 75th, and 90th percentiles as ARL
single gridpoint value, and to alleviate potential height (range) increases, particularly between the 100–
spatial errors in the model-based parameter fields. 2900 and 6000–10 000 ft ARL groups. This suggests that
An example of using the neighborhood grid-hour radar sampling of velocities is limited as horizontal dis-
value versus the grid-hour value is demonstrated by tance from the radar increases, through broader beam-
the Rozel, Kansas, EF4 tornado on 18 May 2013: width and corresponding lower horizontal resolution
STP80km reached 4.5 compared to the 40-km grid- (Wood and Brown 1997). In other words, the WSR-88D
hour value of 0.0 in a case with a tornadic storm in more clearly resolves the stronger and smaller-diameter
proximity to sharp gradients of low-level moisture circulations with EF41 tornado events closer to the
and buoyancy. radar, while radar sampling primarily reflects the me-
socyclone at greater distances (elevations). Some cau-
e. Conditional tornado probabilities tion is warranted in the interpretation of the EF41
tornado events in Fig. 7, given a sample size of only 12
Conditional (i.e., upon the occurrence of a tornado)
cases.
probabilities of tornado intensity, as measured by EF-
Comparisons of 0.58 peak Vrot and storm mode were
scale damage, are calculated using STP80km and 0.58
also completed. One possible contributor to the mis-
peak Vrot. Given the large range in documented STP80km
match between 0.58 peak Vrot and storm mode for sim-
(0–24), 0.58 peak Vrot (0–124 kt), and EF scale (0–5), the
ilar EF-scale damage rating (cf. Figs. 3 and 5–7) is the
sample sizes for paired values of STP80km to EF scale
inability of the WSR-88D to resolve the generally
and 0.58 peak Vrot to EF scale are severely limited in
most cases. Therefore, each STP80km value was placed
within a bin (e.g., 4.00–5.99), and each 0.58 peak Vrot
value below 100 kt was placed within a 10-kt bin (e.g., 3
See Fig. 4 for an approximation of the areal coverage of the
60.0–69.9 kt). WSR-88D radar below 3000 and 6000 ft ARL (ROC 2014).
922 WEATHER AND FORECASTING VOLUME 30

TABLE 2. Mean differences in 0.58 peak Vrot and STP80km for supercells (Sup). Parameter units are the same. Boldface differences are
statistically significant at a , 0.001, and boldface and italic differences are considered to be sufficiently large to be of operational sig-
nificance (i.e., Vrot . 20, STP . 4).

Sup EF41 2 Sup EF41 2 Sup EF41 2 Sup EF41 2 Sup EF3 2 Sup EF3 2 Sup EF3 2
Sup EF0 Sup EF1 Sup EF2 Sup EF3 Sup EF0 Sup EF1 Sup EF2
Vrot 41 35 26 12 30 23 14
STP80km 5.8 5.1 4.4 2.9 2.8 2.2 1.5

shallower vertical depths and smaller horizontal di- statistically significant at a , 0.001 for a two-tailed
mensions of the QLCS and other modes of tornadic Student’s t test (Table 3). While differences in mean 0.58
storm circulations compared to their larger supercell peak Vrot values comparing lower EF-scale-rated su-
counterparts. However, miniature supercells (e.g., percell and QLCS were deemed statistically significant,
Davies 1993) and their smaller-scale mesocyclone cir- these differences are of little practical significance in
culations on radar (e.g., Kennedy et al. 1993; Burgess operations (Tables 3 and 4). Tornado events from QLCS
et al. 1995; Grant and Prentice 1996)—often found in and other modes possessed substantially weaker 0.58
low-CAPE and high-shear environments (e.g., Davis peak Vrot than supercells (Figs. 3 and 5–7). Storms in the
and Parker 2014)—can also pose a challenge in identi- other modes category were disproportionately more
fying rotation due partially to shallower and smaller- difficult to assign 0.58 peak Vrot because of their weaker
diameter circulations. and more ambiguous rotational signatures.
Supercell tornado events were rated roughly one EF-
b. Convective mode and 0.58 peak Vrot
scale category less than QLCS tornado events with
S12 found that variations in tornado EF-scale damage similar 0.58 peak Vrot distributions (Fig. 3; cf. the QLCS
ratings were more closely related to mesocyclone EF1 distribution to supercell EF0). Hence, QLCS and
strength than the specific type of right-moving supercell other modes tend to be associated with weaker 0.58 peak
(discrete cell, cell in cluster, or cell in line). Based on the Vrot values than supercells that produce similar damage.
findings of S12, weak mesocyclones were most common In addition, the majority of QLCS tornadoes in our
with weak tornadoes (EF0 and EF1), whereas strong sample were reported near and east of the Mississippi
mesocyclones were almost exclusively associated with River (S12), where tornado damage paths may be re-
EF31 tornadoes when examining the volume scan prior vealed more consistently by greater population and
to the tornado start time. This study revealed a general vegetation densities compared to the Great Plains. This
increase in 0.58 peak Vrot as EF scale increased for all assertion is supported by mobile radar observations
three convective modes (i.e., supercell, QLCS, and other collected primarily in the Great Plains (Alexander and
modes; Figs. 3 and 5–7). Around one quartile difference Wurman 2008).
in 0.58 peak Vrot was found when compared to 61 su-
c. Near-storm environment
percell EF-scale rating class (Fig. 3). Mean 0.58 peak Vrot
values increased for each supercell EF-scale rating class Based on past studies (e.g., Thompson et al. 2003;
increase (Table 1). Differences in mean 0.58 peak Vrot T12), STP exhibited greater skill in discriminating be-
values for EF41 versus EF0, EF41 versus EF2, and EF2 tween nontornadic and significantly tornadic supercell
versus EF0 were 41, 26, and 15 kt, respectively (Tables 2 environments compared to any of its individual com-
and 3). All differences in EF-scale rating classes among ponents or any other parameters among the 38-variable
supercells (Tables 2 and 3) were statistically significant database at the SPC (see Table A1 in the appendix). The
at a , 0.001 for a two-tailed Student’s t test with unequal near-storm environment portion of this study focuses on
variances (Wilks 2006). A quartile difference $1 is ev- the STP and provides two examples of associating STP
ident between QLCS EF0 and EF2 events and was also with a tornado event. Using either the nearest STP

TABLE 3. As in Table 2, but for Sup 2 Sup, QLCS 2 QLCS, and Sup 2 QLCS differences.

Sup EF2 2 Sup EF2 2 Sup EF1 2 QLCS EF2 2 QLCS EF2 2 QLCS EF1 2 Sup EF2 2 Sup EF2 2
Sup EF0 Sup EF1 Sup EF0 QLCS EF1 QLCS EF0 QLCS EF0 QLCS EF2 QLCS EF1
Vrot 15 9 6 4 9 4 11 16
STP80km 1.3 0.7 0.7 0.5 1.0 0.5 1.5 2.0
AUGUST 2015 SMITH ET AL. 923

TABLE 4. As in Table 3, but for Sup 2 QLCS differences.

Sup EF2 2 Sup EF1 2 Sup EF1 2 Sup EF1 2 Sup EF0 2 Sup EF0 2 Sup EF0 2
QLCS EF0 QLCS EF2 QLCS EF1 QLCS EF0 QLCS EF2 QLCS EF1 QLCS EF0
Vrot 20 2 7 11 4 0 5
STP80km 2.4 0.8 1.3 1.8 0.1 0.6 1.1

gridpoint value or the neighborhood maximum value It must be stressed that composite parameters such as
(i.e., STP80km) for the preceding 40-km grid hour, STP the STP80km should not be examined alone, but rather in
increases as tornado damage classifications increase concert with the individual components in the STP that
(Fig. 8). Supercell events tended to exhibit higher STP identify important supercell tornado ingredients. De-
values than QLCS and other modes for the same EF- spite the promise of STP80km as a relatively simple en-
scale damage rating. The STP for supercell, QLCS, and vironmental diagnostic to assess the potential for
other modes tended to increase monotonically with in- tornadoes, there is no replacement for a thorough di-
creasing damage class ratings (aside from the 10th per- agnosis of the spatiotemporal distribution of buoyancy,
centile). Substantial overlap exists in the distributions shear, and moisture. Furthermore, anticipating changes
between adjacent EF-scale ratings, though the higher to the near-storm environment via airmass modification
values of STP80km (i.e., $6) are more common for a near boundaries, storm interactions, etc., provides an
greater proportion of supercell events at higher EF-scale observational foundation for the effective use of SPC
rating classes (i.e., EF31). Tornadic supercells by EF mesoanalysis data.
scale had higher median STP80km values than QLCS,
d. Relationship among environment, 0.58 peak Vrot ,
and tornadic QLCSs had higher STP80km values than
and EF-scale rating
other modes. Statistically significant differences in
STP80km values (Table 2) were found between EF41, As shown in T12, the differences in effective storm-
EF3, and EF2 supercell classes using a two-sample tailed relative helicity (ESRH; Thompson et al. 2007) be-
difference of means Student’s t test, which complements tween weak and strong tornado environments are
findings by Brotzge et al. (2013) from a similar, in- larger than the differences in MLCAPE. The mean
dependent dataset. values of ESRH increase more rapidly than the mean

FIG. 8. Box-and-whiskers plot of effective-layer STP (dimensionless) for all supercell, QLCS
EF0–EF2, and other modes EF0 and EF1 by EF-scale damage rating classes (40-km grid data
are shaded gray, labels on right). Black overlays (labels on left) denote STP80km values, at the
analysis time immediately preceding the event time. Other conventions are the same as in
Fig. 3.
924 WEATHER AND FORECASTING VOLUME 30

FIG. 9. Scatterplot of EF0–EF5 tornado events (2009–13; inverted triangle symbol) by EF-scale rating (legend at top right) of 100-hPa
mixed layer CAPE (J kg21; x coordinate) vs effective storm-relative helicity (m2 s22; y coordinate) and 0.58 peak Vrot proportionately sized
to velocity strength. The circles represent the mean values of ESRH and MLCAPE for each EF-scale rating.

values of MLCAPE at the lower EF-scale tornado Weakly damaging tornado events primarily occupy
damage ratings (i.e., from EF0 to EF2; Fig. 9). How- the distribution space featuring weaker velocities and
ever, MLCAPE exhibits an increasing influence on low STP80km values (Fig. 10). Strong tornadoes (i.e., EF2
mean EF-scale tornado damage ratings transitioning and EF3) tend to mostly occur with Vrot $ 40 kt but
from the middle to upper EF scale (i.e., from EF2 across much of the STP80km parameter space (e.g., 0.5–
to EF5). 15). Violent tornadoes (EF41) were found to generally

FIG. 10. As in Fig. 9, but for 0.58 peak Vrot (kt) vs STP80km (dimensionless).
AUGUST 2015 SMITH ET AL. 925

FIG. 11. Conditional probability of meeting or exceeding a given EF-scale rating (legend) for STP80km (dimensionless; x coordinate;
sample size) for all convective mode tornado events (2009–13; at #10 000 ft ARL, with 1–101-mi radius).

favor both extreme 0.58 peak Vrot ($90 kt) and near- resolved by the WSR-88D and are sampled by a greater
storm environments (STP80km $ 8). Mean EF-scale number of radar volume scans per tornado life cycle
tornado damage ratings increase owing to both compared to small, weak, and short-track (short lived)
strengthening 0.58 peak V rot and higher values of tornadoes.
STP 80km (Fig. 10).
e. Conditional probabilities of tornado intensity
The direct relationship between 0.58 peak Vrot and
tornado damage ratings (i.e., stronger Vrot associated Another potentially valuable way to extract mean-
with higher ratings) can be extended to other attributes ingful real-time information involves examining the
of tornadoes. For example, Brooks (2004) suggests that relationship among STP80km, 0.58 peak Vrot, and EF-
stronger tornadoes are associated with both 1) wider scale damage via conditional exceedance probabili-
damage paths and 2) longer pathlengths. Furthermore, ties of EF-scale damage ratings (i.e., given a tornado,
tornado intensity is a fundamental component of the what is the probability of damage of at least a certain
destruction potential index (DPI; Thompson and rating?). The relatively large sample size of tornado
Vescio 1998; Doswell et al. 2006), which is a parameter EF-scale ratings enables this study to examine dif-
that characterizes tornado impact. As a logical exten- ferences and similarities across EF-scale ratings. The
sion of these relationships, very strong 0.58 peak Vrot conditional probability of tornado damage intensity
signatures (conditional upon a tornado; .80 kt) imply exhibits a strong, yet seemingly robust and stable,
the possibility of more intense, wider, and longer-track signal of increasing probability for higher EF scale
tornadoes that can impact larger areas and potentially as both STP80km and 0.58 peak Vrot increase (Figs. 11
exert a greater societal impact in the form of damage and 12). An overwhelming majority (i.e., $90%) of
and fatalities. We find that these larger, more intense, tornado events within environments ,1 STP80km
and longer-lived tornadic circulations are better or ,30 kt for 0.58 peak Vrot are weak tornadoes (EF0
926 WEATHER AND FORECASTING VOLUME 30

FIG. 12. As in Fig. 11, but for 0.58 peak Vrot (kt; x coordinate).

and EF1). This is illustrated in Figs. 11 and 12, where the 50–59.9-and 90–99.9-kt ranges, respectively. Grouping
probability of EF21 events is less than or equal to 10% EF-scale tornado events into weak (EF0 and EF1),
at the aforementioned STP80km and 0.58 peak Vrot strong (EF2 and EF3), and violent (EF4 and EF5)
thresholds and below. As the environment becomes categories (Fig. 13) offers a simple way to amplify
more favorable for tornadic supercells and STP80km rises differences in the distribution of conditional proba-
from the lower single digits to 10 or higher, the condi- bilities. The conditional probability of a weak tor-
tional probabilities for an EF21 increase from 15%– nado is larger than the other EF-scale categories at
20% to 45%–50% (Fig. 11). A similar overall trend is weaker Vrot (i.e.,,60 kt), whereas strong and violent
displayed in Fig. 12 as 0.58 peak Vrot increases, but a tornado event categories have probability maxima
larger increase in conditional probabilities is evident displaced at stronger 0.58 peak V rot magnitudes
with higher EF scale compared to STP80km. Conditional (80.0–89.9 and 100.0–124.2 kt, respectively).
probabilities based solely on 0.58 peak Vrot for EF21 Both STP80km and 0.58 peak Vrot information were
events are 55%–60% when 0.58 peak Vrot ranges from binned and plotted together (Fig. 14) to provide the
60.0 to 69.9 kt. Even though the sample size of the tor- conditional probability of an EF21 tornado, similar
nado events with 0.58 peak Vrot of 80 kt or higher is small, to a forecaster having both datasets to consider in
the proportion of very damaging tornadoes increases real time. Conditional EF21 tornado probabilities
markedly (i.e., rapid increase in conditional probabili- increase by ;50% as 0.58 peak Vrot increases from 40
ties). For example, 0.58 peak Vrot ranging from 80.0 to to 80 kt for all STP80km values. A smaller increase
89.9 kt results in conditional probabilities of 65%–70% (;15%) in conditional EF21 tornado probabilities is
for EF31 events and 20% for EF41 events. To achieve evident as STP80km increases from 0 to 10 within a
high confidence (i.e., .75%) in damage ratings of range of 40–80-kt 0.58 peak Vrot. Combining both
EF11 or EF31, 0.58 peak Vrot must meet or exceed the pieces of information yields the largest increase in
AUGUST 2015 SMITH ET AL. 927

FIG. 13. Conditional probability of grouped EF-scale rating classes (legend on right; EF0 and EF1, EF2 and EF3, and EF4 and EF5) for
0.58 peak Vrot (kt; x coordinate; sample size) for all convective mode tornado events (2009–13; at #10 000 ft ARL, with 1–101-mi radius).

conditional EF21 probabilities. Specifically, an in- convective mode, and strong mesocyclones yielded the
crease from 40-kt 0.58 peak V rot with near-zero greatest risk for EF31 tornadoes. However, T12 found
STP80km to 80-kt 0.58 peak Vrot in an environment substantial overlap in STP distributions by EF-scale
with STP80km around 10 results in a more than 65% rating (T12, their Fig. 12) and emphasized the follow-
increase in conditional EF21 tornado probability, ef- ing statement: ‘‘confident delineation in damage cate-
fectively demonstrating the utility of combining both gories will prove difficult for individual storms during a
datasets to best discriminate between EF0 and EF1 particular hour based on storm mode and environment
and EF2–EF5 tornado events. alone.’’ Their assertion served as the primary motivation
to develop a dataset with greater precision of the parent
tornadic storm low-level circulation intensity than was
4. Summary and discussion
done previously in S12 (i.e., 1-kt Vrot increments vs three
Over 4700 tornado events during 2009–13 were ana- broad categories of mesocyclone strength). The addi-
lyzed from a spatially diverse sampling of tornadic storm tional work was completed by manually assigning 0.58
modes and environments within 101 mi of operational peak Vrot to tornado events, and this highlighted a dis-
WSR-88D radars. As part of a comprehensive convec- tinct relationship between tornado event characteristics
tive mode–environment investigation at the SPC, both (i.e., radar attributes) and tornado damage ratings.
previous foundational studies (i.e., S12 and T12) high- This study demonstrates the usefulness of a multiple-
lighted the relationship between convective mode, dataset approach to better assess the conditional prob-
mesocyclone strength, and tornado damage ratings. ability of maximum tornado EF scale by combining
Additionally, T12 combined near-storm environment information on the near-storm environment, convective
data (e.g., STP) with a large sample of tornado events mode, and 0.58 peak Vrot (Fig. 15). This approach can
and validated that high STP, right-moving supercell be applied operationally by considering the following:
928 WEATHER AND FORECASTING VOLUME 30

FIG. 14. Smoothed conditional probability of EF21 tornado rating (shaded) of 0.58 peak Vrot
(kt; x coordinate) vs STP80km (dimensionless; y coordinate). The conditional probability is only
calculated and shown for bins with at least one EF21 tornado.

1) real-time comparison of available observations with awareness approach. While the conditional probability
the model-based estimates of the storm environment, approach is not intended explicitly for tornado warnings
2) real-time monitoring of storm structure and rota- with lead time, the STP80km and 0.58 peak Vrot can aid in
tional characteristics via WSR-88D sampling, and anticipating decision thresholds as the warning decision-
3) supporting evidence of a tornado via the development making process evolves. Specifically, making use of 0.58
of a dual-polarization tornadic debris signature peak Vrot trends and corresponding changes in condi-
(DPTDS; e.g., Bodine et al. 2013; Schultz et al. 2012a,b; tional tornado strength probabilities during tornado
Bunkers and Baxter 2011) with vertical and temporal warnings can be used by the warning forecaster to both
continuity, since a majority of tornadoes are not re- assess changes in intensity and confidence to convey
ported to local National Weather Service offices in real tornado intensification in severe weather statements
time (Blair and Leighton 2014). When observations accompanying tornado warnings on an event-driven
corroborate the model-based estimates, and velocity basis.
signatures show spatial, temporal, and vertical continu- False alarms are probable with strict application of the
ity in the storm’s low levels, confidence can be higher in technique (e.g., Fig. 15) when there is no confirming
the application of the conditional probabilities derived evidence of a tornado. A case from the evening of
from 0.58 peak Vrot. Conversely, disagreement between 31 August 2014 illustrates the potential for false alarms.
observations and model-based estimates of the envi- A supercell in Fremont County, Iowa, exhibiting a 0.58
ronment, or sharp gradients among the meteorological peak Vrot of 63 kt at 0102 UTC and an estimated
variables with few corresponding observations, would STP80km value near 7 in the 0000 UTC SPC mesoanalysis
suggest lower confidence in an expected outcome. data (not shown), resulted in a ;53% conditional EF21
Undoubtedly, a continued critical evaluation by opera- probability. However, the strongest Vrot with the storm
tional forecasters (Guyer and Hart 2012) is needed to was confined to only a 5-min period (the 0059–0104 UTC
provide a case-by-case diagnosis and short-term pre- scans), and there was no corresponding DPTDS.
diction of the atmosphere while considering the inherent Additionally, though the SPC mesoanalysis data from
assumptions and limitations of this simplified situational 0100 UTC admittedly would not have been available
AUGUST 2015 SMITH ET AL. 929

FIG. 15. (a) Observed sounding from Lamont, OK, at 0000 UTC 15 Apr 2012, with STP 6.7. (b) Harper County,
KS, tornadic storm location (green circle), SPC mesoanalysis 40-km grid (black square), and 80-km radius (black
circle), with STP80km 12.0. (c) As in Fig. 1a, but for Vance Air Force Base, OK (KVNX), at 0142 UTC 15 Apr 2012.
A discrete-cell supercell produced an EF1 tornado event in Harper County, KS. (d) As in Fig. 1b, but for KVNX
for max Vin (87.4 kt), max Vout (96.2 kt), and 0.58 peak Vrot (91.8 kt) sampled at 1000 ft ARL. (e) KVNX WSR-88D
dual-pol cross-correlation coefficient (rhv) indicative of a tornadic debris signature. (f) As in (a), but for Norman,
OK, at 1700 UTC 20 May 2013, with STP 2.7. (g) As in (b), but for Cleveland County, OK. Note that the smoothed
planar STP value (5–6) is different than STP80km 7.2. (h) As in (c), but for Twin Lakes, OK (KTLX), at
2012 UTC 20 May 2013. A discrete-cell supercell produced an EF5 tornado event in Cleveland County, OK.
(i) As in (d), but for KTLX for max Vin (102.0 kt), max Vout (81.6 kt), and 0.58 peak Vrot (91.8 kt), sampled at 900 ft
ARL. (j) As in (e), but for KTLX. (k) As in (a), but for Omaha, NE, at 0000 UTC 1 Sep 2014, with STP 1.3. As in (b),
but for Fremont County, IA, with STP80km 7.2. (m) As in (c), but for Omaha/Valley, NE (KOAX), at 0102 UTC 1
Sep 2014. A cell-in-cluster supercell was nontornadic and produced a wind damage report in Fremont County, IA.
(n) As in (d), but for KOAX for max Vin (62.2 kt), max Vout (63.1 kt), and 0.58 peak Vrot (62.7 kt) sampled at 3500 ft
ARL. (o) As in (e), but for KOAX with no evidence of a tornadic debris signature.
930 WEATHER AND FORECASTING VOLUME 30

TABLE A1. Archived SPC mesoanalysis variables. Parameters are nondimensional where no units are listed, and abbreviations are as
follows: lowest 100-hPa mean layer (ML), surface based (SB), most unstable lowest 300 hPa (MU), and maximum 50-hPa mean
layer (best).

STP80km (fixed layer) STP (fixed layer)

STP80km (effective layer with CIN) STP (effective layer with CIN)
Effective inflow base STP (Thompson et al. 2003)
SCP80km (effective layer) SCP (effective layer)
ML, SB, MU, best CAPE (J kg21) 0–8, 0–6, 0–3, 0–1-km effective bulk wind difference (kt)
ML, SB, MU, best CIN (J kg21) 0–3, 0–1 effective storm-relative helicity (m2 s22)
ML, MU, LCL (m AGL) 0–3-km, 850–500-hPa, 700–500-hPa lapse rate (8C km21)
ML, MU, LFC (m AGL) Surface temp (8F)
Downdraft CAPE (J kg21) Surface dewpoint temp (8F)
0–3-km CAPE (Jkg21) Surface relative humidity (%)
Significant hail parameter Precipitable water (in.)

until ;0125 UTC, a substantial decrease in STP80km Bodine, D. J., M. R. Kumjian, R. D. Palmer, P. L. Heinselman, and
from ;7 to ;2 was noted from 0000 to 0100 UTC. The A. V. Ryzhkov, 2013: Tornado damage estimation using po-
larimetric radar. Wea. Forecasting, 28, 139–158, doi:10.1175/
lack of a real-time tornado report in this example,
WAF-D-11-00158.1.
combined with the lack of a DPTDS and a decrease in Bothwell, P. D., J. A. Hart, and R. L. Thompson, 2002: An in-
STP80km over time, suggested that an EF21 was un- tegrated three-dimensional objective analysis scheme in use at
likely. Work is ongoing to develop a large null sample to the Storm Prediction Center. Preprints, 21st Conf. on Severe
create unconditional tornado probabilities to aid in the Local Storms, San Antonio, TX, Amer. Meteor. Soc., J117–
J120. [Available online at https://ams.confex.com/ams/pdfpapers/
real-time diagnosis of tornado potential.
47482.pdf.]
Brady, R. H., and E. J. Szoke, 1989: A case study of nonmesocyclone
Acknowledgments. This study benefited from early tornado development in northeast Colorado: Similarities to
discussions with Dr. Israel Jirak and Steven Weiss waterspout formation. Mon. Wea. Rev., 117, 843–856,
(SPC), Dr. Harold Brooks (NSSL), and James LaDue doi:10.1175/1520-0493(1989)117,0843:ACSONT.2.0.CO;2.
(WDTB), as well as a manuscript review by Dr. Jirak. Brooks, H. E., 2004: On the relationship of tornado path length and
width to intensity. Wea. Forecasting, 19, 310–319, doi:10.1175/
We thank three anonymous reviewers for valuable 1520-0434(2004)019,0310:OTROTP.2.0.CO;2.
suggestions and constructive criticisms that prompted us Brotzge, J. A., S. E. Nelson, R. L. Thompson, and B. T. Smith, 2013:
to better clarify our thoughts in this manuscript. Tornado probability of detection and lead time as a function of
convective mode and environmental parameters. Wea. Fore-
casting, 28, 1261–1275, doi:10.1175/WAF-D-12-00119.1.
APPENDIX Bunkers, M. J., and M. A. Baxter, 2011: Radar tornadic debris
signatures on 27 April 2011. Electron. J. Oper. Meteor., 12 (7).
[Available online at http://www.nwas.org/ej/2011/2011.php.]
SPC Mesoanalysis Variables Burgess, D. W., R. L. Lee, S. S. Parker, S. J. Keighton, and D. L.
Floyd, 1995: A study of mini supercells observed by WSR-88D
Table A1 provides information on archived SPC
radars. Preprints, 27th Conf. on Radar Meteorology, Vail, CO,
mesoanalysis variables. Amer. Meteor. Soc., 4–6.
Coniglio, M. C., 2012: Verification of RUC 0–1-km forecasts and
REFERENCES SPC mesoscale analyses using VORTEX2 soundings. Wea.
Forecasting, 27, 667–683, doi:10.1175/WAF-D-11-00096.1.
Alexander, C. R., and J. M. Wurman, 2008: Updated mobile radar Craven, J. P., and H. E. Brooks, 2004: Baseline climatology of
climatology of supercell tornado structures and dynamics. sounding derived parameters associated with deep, moist
Preprints, 24th Conf. Severe Local Storms, Savannah, GA, convection. Natl. Wea. Dig., 28, 13–24.
Amer. Meteor. Soc., 19.4. [Available online at https://ams. Crum, T. D., R. L. Alberty, and D. W. Burgess, 1993: Recording, ar-
confex.com/ams/pdfpapers/141821.pdf.] chiving, and using WSR-88D data. Bull. Amer. Meteor. Soc., 74, 645–
Andra, D. L., Jr., 1997: The origin and evolution of the WSR-88D 653, doi:10.1175/1520-0477(1993)074,0645:RAAUWD.2.0.CO;2.
mesocyclone recognition nomogram. Preprints, 28th Conf. on Davies, J. M., 1993: Small tornadic supercells in the central plains.
Radar Meteorology, Austin, TX, Amer. Meteor. Soc., 364–365. Preprints, 17th Conf. Severe Local Storms, St. Louis, MO,
Benjamin, S. G., and Coauthors, 2004: An hourly assimilation– Amer. Meteor. Soc., 305–309.
forecast cycle: The RUC. Mon. Wea. Rev., 132, 495–518, ——, 2004: Estimations of CIN and LFC associated with tornadic
doi:10.1175/1520-0493(2004)132,0495:AHACTR.2.0.CO;2. and nontornadic supercells. Wea. Forecasting, 19, 714–726,
Blair, S. F., and J. W. Leighton, 2014: Assessing real-time tornado doi:10.1175/1520-0434(2004)019,0714:EOCALA.2.0.CO;2.
information disseminated through NWS products. Wea. ——, and A. Fischer, 2009: Environmental characteristics asso-
Forecasting, 29, 591–600, doi:10.1175/WAF-D-13-00126.1. ciated with nighttime tornadoes. Electron. J. Oper. Meteor.,
AUGUST 2015 SMITH ET AL. 931

10 (3). [Available online at http://www.nwas.org/ej/2009/2009. Parker, M. D., 2014: Composite VORTEX2 supercell environ-
php]. ments from near-storm soundings. Mon. Wea. Rev., 142, 508–
Davis, J. M., and M. D. Parker, 2014: Radar climatology of tornadic 529, doi:10.1175/MWR-D-13-00167.1.
and nontornadic vortices in high-shear, low-CAPE envi- Potvin, C. K., K. L. Elmore, and S. J. Weiss, 2010: Assessing the
ronments in the Mid-Atlantic and southeastern United impacts of proximity sounding criteria on the climatology of
States. Wea. Forecasting, 29, 828–853, doi:10.1175/ significant tornado environments. Wea. Forecasting, 25, 921–
WAF-D-13-00127.1. 930, doi:10.1175/2010WAF2222368.1.
Doswell, C. A., III, R. Edwards, R. L. Thompson, J. A. Hart, and Rasmussen, E. N., 2003: Refined supercell and tornado forecast
K. C. Crosbie, 2006: A simple and flexible method for ranking parameters. Wea. Forecasting, 18, 530–535, doi:10.1175/
severe weather events. Wea. Forecasting, 21, 939–951, 1520-0434(2003)18,530:RSATFP.2.0.CO;2.
doi:10.1175/WAF959.1. ——, and D. O. Blanchard, 1998: A baseline climatology
Eilts, M. D., and S. D. Smith, 1990: Efficient dealiasing of Doppler of sounding-derived supercell and tornado forecast para-
velocities using local environment constraints. J. Atmos. Oceanic meters. Wea. Forecasting, 13, 1148–1164, doi:10.1175/
Technol., 7, 118–128, doi:10.1175/1520-0426(1990)007,0118: 1520-0434(1998)013,1148:ABCOSD.2.0.CO;2.
EDODVU.2.0.CO;2. ROC, 2014: NEXRAD coverage below 10,000 feet AGL. NOAA/
French, M. M., H. B. Bluestein, I. PopStefanija, C. A. Baldi, and Radar Operations Center, accessed 1 September 2014.
R. T. Bluth, 2013: Reexamining the vertical development of [Available online at http://www.roc.noaa.gov/WSR88D/Maps.
tornadic vortex signatures in supercells. Mon. Wea. Rev., 141, aspx.]
4576–4601, doi:10.1175/MWR-D-12-00315.1. Schultz, C. J., and Coauthors, 2012a: Dual-polarization tornadic
Grant, B. N., and R. Prentice, 1996: Mesocyclone characteristics of debris signatures Part I: Examples and utility in an operational
mini supercell thunderstorms. Preprints, 15th Conf. on setting. Electron. J. Oper. Meteor., 13 (9), 120–137. [Available
Weather Analysis and Forecasting, Norfolk, VA, Amer. Me- online at http://www.nwas.org/ej/2012/2012.php.]
teor. Soc., 362–365. ——, and Coauthors, 2012b: Dual-polarization tornadic debris
Guyer, J. L., and J. A. Hart, 2012: Examination of WSR-88D VWP signatures Part II: Comparisons and caveats. Electron. J. Oper.
data in proximity to strong tornadoes. Preprints, 26th Conf. on Meteor., 13 (10), 138–150. [Available online at http://www.
Severe Local Storms, Nashville, TN, Amer. Meteor. Soc., 62. nwas.org/ej/2012/2012.php.]
[Available online at https://ams.confex.com/ams/26SLS/ Smith, B. T., R. L. Thompson, J. S. Grams, C. Broyles, and H. E.
webprogram/Paper211652.html.] Brooks, 2012: Convective modes for significant severe thun-
Kennedy, P. C., N. E. Westcott, and R. W. Scott, 1993: Single- derstorms in the contiguous United States. Part I: Storm
Doppler radar observations of a mini-supercell tornadic classification and climatology. Wea. Forecasting, 27, 1114–
thunderstorm. Mon. Wea. Rev., 121, 1860–1870, doi:10.1175/ 1135, doi:10.1175/WAF-D-11-00115.1.
1520-0493(1993)121,1860:SDROOA.2.0.CO;2. Smith, T. M., and K. L. Elmore, 2004: The use of radial velocity
Kingfield, D. M., J. G. LaDue, and K. L. Ortega, 2012: An evalu- derivatives to diagnose rotation and divergence. Preprints,
ation of tornado intensity using velocity and strength attri- 11th Conf. on Aviation, Range, and Aerospace, Hyannis, MA,
butes from the WSR-88D mesocyclone detection algorithm. Amer. Meteor. Soc., P5.6. [Available online at http://ams.
Preprints, 26th Conf. on Severe Local Storms, Nashville, TN, confex.com/ams/pdfpapers/81827.pdf.]
Amer. Meteor. Soc., 3.2. [Available online at https://ams. Speheger, D. A., and R. D. Smith, 2006: On the imprecision of
confex.com/ams/26SLS/webprogram/Paper211488.html.] radar signature locations and storm path forecasts. Natl. Wea.
LaDue, J. G., K. L. Ortega, B. R. Smith, G. J. Stumpf, and D. M. Dig., 30, 3–10.
Kingfield, 2012: A comparison of high resolution tornado Stensrud, D. J., J. V. Cortinas Jr., and H. E. Brooks, 1997: Dis-
surveys to Doppler radar observed mesocyclone parameters: criminating between tornadic and nontornadic thunderstorms
2011–2012 case studies. Preprints, 26th Conf. on Severe Local using mesoscale model output. Wea. Forecasting, 12, 613–632,
Storms, Nashville, TN, Amer. Meteor. Soc., 6.3. [Available doi:10.1175/1520-0434(1997)012,0613:DBTANT.2.0.CO;2.
online at https://ams.confex.com/ams/26SLS/webprogram/ ——, and Coauthors, 2009: Convective-scale warn-on-forecast
Paper212627.html.] system: A vision for 2020. Bull. Amer. Meteor. Soc., 90,
Laflin, J. M., 2013: Verification of RAP model soundings in pre- 1487–1499, doi:10.1175/2009BAMS2795.1.
convective environments. J. Oper. Meteor., 1 (6), 66–70, Stumpf, G. J., A. Witt, E. D. Mitchell, P. L. Spencer, J. T. Johnson,
doi:10.15191/nwajom.2013.0106. M. D. Eilts, K. W. Thomas, and D. W. Burgess, 1998: The
Magsig, M. A., 2008: New techniques for integrating environmental National Severe Storms Laboratory mesocyclone detection
information with radar base data analysis in National Weather algorithm for the WSR-88D. Wea. Forecasting, 13, 304–326,
Service warning decision making. Preprints, 24th Conf. on doi:10.1175/1520-0434(1998)013,0304:TNSSLM.2.0.CO;2.
Severe Local Storms, Savannah, GA, Amer. Meteor. Soc., Thompson, R. L., and M. D. Vescio, 1998: The destruction po-
P6.10. [Available online at https://ams.confex.com/ams/24SLS/ tential index: A method for comparing tornado days. Pre-
techprogram/paper_142264.htm.] prints, 19th Conf. Severe Local Storms, Minneapolis, MN,
Mitchell, E. D., S. V. Vasiloff, G. J. Stumpf, A. Witt, M. D. Eilts, Amer. Meteor. Soc., 280–282.
J. T. Johnson, and K. W. Thomas, 1998: The National Severe ——, R. Edwards, J. A. Hart, K. L. Elmore, and P. M. Markowski,
Storms Laboratory tornado detection algorithm. Wea. Fore- 2003: Close proximity soundings within supercell environ-
casting, 13, 352–366, doi:10.1175/1520-0434(1998)013,0352: ments obtained from the Rapid Update Cycle. Wea. Fore-
TNSSLT.2.0.CO;2. casting, 18, 1243–1261, doi:10.1175/1520-0434(2003)018,1243:
Newman, J. F., V. Lakshmanan, P. L. Heinselman, M. B. Richman, CPSWSE.2.0.CO;2.
and T. M. Smith, 2013: Range-correcting azimuthal shear in ——, C. M. Mead, and R. Edwards, 2007: Effective storm-relative
Doppler radar data. Wea. Forecasting, 28, 194–211, helicity and bulk shear in supercell thunderstorm environ-
doi:10.1175/WAF-D-11-00154.1. ments. Wea. Forecasting, 22, 102–115, doi:10.1175/WAF969.1.
932 WEATHER AND FORECASTING VOLUME 30

——, B. T. Smith, J. S. Grams, A. R. Dean, and C. Broyles, 2012: ——, S. A. Tessendorf, E. S. Godfrey, and H. E. Brooks, 2005:
Convective modes for significant severe thunderstorms in the Tornadoes from squall lines and bow echoes. Part I:
contiguous United States. Part II: Supercell and QLCS tornado Climatological distribution. Wea. Forecasting, 20, 23–34,
environments. Wea. Forecasting, 27, 1136–1154, doi:10.1175/ doi:10.1175/WAF-835.1.
WAF-D-11-00116.1. Wagenmaker, R., G. Mann, and M. Hudson, 2014: IBW project:
——, ——, A. R. Dean, and P. T. Marsh, 2013: Spatial distributions Introduction and overview. NOAA/NWS, accessed 15 July
of tornadic near-storm environments by convective mode. 2014. [Available online at http://www.crh.noaa.gov/images/dtx/
Electron. J. Severe Storms Meteor., 8 (5). [Available online at GLOMW/Presentations/2014_IBW_Project_Intro_Overview.pdf.]
http://www.ejssm.org/ojs/index.php/ejssm/issue/view/47.] Wilks, D. S., 2006: Statistical Methods in the Atmospheric Sciences.
Torres, S. M., and C. D. Curtis, 2007: Initial implementation of super- 2nd ed. International Geophysics Series, Vol. 59, Academic
resolution data on the NEXRAD network. Preprints, 23rd Int. Press, 627 pp.
Conf. on Interactive Information Processing Systems, San Anto- Wood, V. T., and R. A. Brown, 1997: Effects of radar sampling
nio, TX, Amer. Meteor. Soc., 5B.10. [Available online at http:// on single-Doppler velocity signatures of mesocyclones and
ams.confex.com/ams/pdfpapers/116240.pdf.] tornadoes. Wea. Forecasting, 12, 928–938, doi:10.1175/
Toth, M., R. J. Trapp, J. Wurman, and K. A. Kosiba, 2013: Com- 1520-0434(1997)012,0928:EORSOS.2.0.CO;2.
parison of mobile-radar measurements of tornado intensity Zittel, W. D., and Z. Jing, 2012: Comparison of a 2-D velocity
with corresponding WSR-88D measurements. Wea. Fore- dealiasing algorithm to the legacy WSR-88D velocity dealiasing
casting, 28, 418–426, doi:10.1175/WAF-D-12-00019.1. algorithm during Hurricane Irene. Preprints, 30th Conf. on
Trapp, R. J., and M. L. Weisman, 2003: Low-level mesovortices Hurricanes and Tropical Meteorology, Ponte Vedra Beach, FL,
within squall lines and bow echoes. Part II: Their genesis and Amer. Meteor. Soc., 7C.7. [Available online at https://
implications. Mon. Wea. Rev., 131, 2804–2823, doi:10.1175/ ams.confex.com/ams/30Hurricane/webprogram/Paper205076.
1520-0493(2003)131,2804:LMWSLA.2.0.CO;2. html.]