1. Introduction
On 8–9 February 2013, a nor’easter moved up the North Atlantic coast of the United States, resulting in copious snowfall and high winds over much of the region. This historic winter weather event exhibited several distinctive polarimetric radar signatures, including exceptionally high reflectivities, differential attenuation, downward excursions of the melting layer to the surface, depositional growth layers, depolarization streaks, nonuniform beamfilling, and a “snow flare” signature. These observed polarimetric features provide valuable insights into the system’s ongoing microphysical processes.
With the recent polarimetric upgrade to the U.S. Weather Surveillance Radar-1988 Doppler (WSR-88D) network, dual-polarization data are now available for locations throughout the country, thereby allowing for observations of the microphysical properties of precipitation in regions of the country never before sampled by polarimetric radar. Several fascinating polarimetric signatures in winter storms have recently been discovered. One of the more interesting signatures is an elevated region of enhanced specific differential phase KDP and differential reflectivity ZDR, and reduced copolar correlation coefficient ρhv. Kennedy and Rutledge (2011) reported on S-band measurements of an elevated layer of KDP in four Colorado winter storms that had local maxima of ~0.15°–0.4° km−1 near the −15°C isotherm. They further showed that the passage of these regions was associated with an increase in surface precipitation. Calculations using an electromagnetic scattering model indicated that highly oblate spheroidal particles with diameters between ~0.8 and 1.2 mm in range and moderate ice densities produced KDP values that were consistent with radar observations. Their calculations were unable to reproduce ZDR. They further concluded that the persistent collocation of this signature with the −15°C isotherm was an indication that rapidly growing dendrites likely played a significant role in producing the elevated KDP signature. In another study of winter storms, Andrić et al. (2013) reported on isolated pockets of enhanced KDP and ZDR that were collocated with reduced ρhv. These signatures were also found to be located at temperatures between −10° and −15°C, and were coincident with a zone of large radar reflectivity vertical gradient, with the reflectivity factor ZH increasing toward the ground. Using a simple kinematical, one-dimensional, two-moment bulk microphysical model that was coupled with an electromagnetic scattering model, Andrić et al. (2013) were able to approximately reproduce the correct profile shape and magnitude of ZH and ρhv and the correct shape (but not magnitude) of ZDR and KDP. They concluded that their inability to reproduce the correct profiles and magnitudes of all of the signatures indicated that microphysical processes not included in the model, such as secondary ice production, were likely important factors in producing the observed signature. Bechini et al. (2013) also documented enhanced ZDR and KDP values near the model-indicated −15°C isotherm in the ice region of precipitating clouds, using C- and X-band radars in northwestern Italy. They found that these regions of enhancement (KDP values peaked around 2.0° km−1 at C band) were likely associated with dendritic growth and were correlated with the ZH below.
Ryzhkov et al. (2011) and Kumjian et al. (2011) presented preliminary examinations of distinct and recurring polarimetric signatures in winter storms, including the aforementioned elevated features, providing useful groundwork for future winter weather polarimetric studies. As recognized by Hogan et al. (2002) and Field et al. (2004), plumes of enhanced ZDR are associated with updrafts and the generation of supercooled liquid water in winter convective storms. According to Ryzhkov et al. (2011) and Kumjian et al. (2011), it is likely that some of these updrafts can produce enough graupel for charge separation sufficient to generate electric fields. These electrostatic fields can change the orientation of ice crystals atop these updrafts, which can cause the transmitted radar signal to become depolarized (Ryzhkov and Zrnić 2007; Hubbert et al. 2010a,b; Kumjian et al. 2011; Ryzhkov et al. 2011). The resulting polarimetric signatures reveal the depolarization through radial streaks of enhanced positive and/or negative ZDR.
Other distinct polarimetric signatures presented by Ryzhkov et al. (2011) include downward excursions of the melting layer, characterized by reduced ρhv and locally maximized ZH and ZDR extending from the melting layer down to the surface. These excursions are usually associated with melting snowflakes extracting heat from the local environment, resulting in the cooling of ambient air temperatures (Kain et al. 2000). A gradual downward propagation of melting-induced cooling can produce a 0°C isothermal layer, allowing precipitation to fall farther toward the surface before melting entirely; if the melting layer is near the ground, frozen precipitation is more likely to reach the surface (e.g., Findeisen 1940; Kain et al. 2000). Furthermore, this melting effect is strongest in locations of high precipitation rates (Kain et al. 2000). Downward excursions of the melting layer can also be associated with weak embedded convection, which facilitates the riming of snow above the freezing level (Ryzhkov et al. 2011). Rimed snowflakes have larger terminal velocities and their complete melting within the melting layer occurs at lower heights compared to unrimed snow. Polarimetric signatures of these melting-layer downward excursions to the surface have been associated with abrupt changes in precipitation type reaching the ground, which present a challenge to forecasters. Therefore, polarimetric radar observations of this phenomenon could prove useful in winter weather nowcasting.
The most recently documented winter polarimetric feature is the refreezing signature, which forms in the lower levels beneath the melting layer and is characterized by enhanced ZDR and KDP, and reduced ZH and ρhv. Kumjian et al. (2013) documented this signature, and associated environmental conditions, for four winter storms over central Oklahoma, using the S-band polarimetric WSR-88D in Norman, Oklahoma (KOUN), and the University of Oklahoma’s C-band Polarimetric Radar for Innovations in Meteorology and Engineering (OU-PRIME), which is also located in Norman. During the events, the refreezing signature occurred when ice pellets were reaching the surface (Kumjian et al. 2013). Possible microphysical mechanisms of the low-level ZDR enhancement include preferential freezing of the smaller drops (so that the raindrop size spectrum becomes more skewed toward larger drops that may freeze later) and the possible local generation of anisotropic ice crystals in the subfreezing air (Kumjian et al. 2013). Further analysis is needed to thoroughly understand the origin of the ice crystals, as well as to explain the microphysical processes responsible for the occurrence of this unique polarimetric feature. Nevertheless, the refreezing signature will be valuable in detecting transitions of precipitation from freezing rain to ice pellets (Ryzhkov et al. 2011; Kumjian et al. 2013).
This study investigates the evolution and nature of intriguing and previously undocumented polarimetric signatures observed during the 8–9 February blizzard, and examines them in light of the thermodynamic environment within which they developed and the apparent microphysical processes that were active when they appeared. We examine data from the polarimetric WSR-88D S-band radar in Upton, New York (Long Island; KOKX) and environmental thermodynamic analyses from the operational 13-km Rapid Refresh (RAP) model. The model output is used to interpret the polarimetric signatures of different types of ice crystal habits (e.g., needles, plates, stellars, and dendrites) that form in different temperature regimes and regions of ice supersaturation. Polarimetric signatures are also analyzed alongside surface precipitation type observations from Meteorological Phenomena Identification Near the Ground (mPING); the National Weather Service (NWS) Forecast Office in Upton, New York; and Stony Brook University in Stony Brook, New York [observations from Ganetis et al. (2013); approximately 284° azimuth and 23-km range from KOKX]. These data aid in the interpretation of the radar observations.
2. Storm overview
On 8 February 2013, two low pressure systems merged over the northeastern United States, producing a historic winter weather event that ranked among the top five worst blizzards in the region (NART 2014). Figure 1 illustrates the evolution of synoptic-scale weather patterns during the event using the 13-km RAP (Cifelli et al. 2005) model analyses. At 1200 UTC, a low was located just off the North Carolina coast, while a second low was located over the Midwest and the 0°C isotherm at 2 m above ground level was located just south of Long Island, New York (Fig. 1a). At this time, precipitation was primarily over the mid-Atlantic states. Over the next 12 h, the two systems merged off of the East Coast and the resultant low deepened and progressed northeastward, while the precipitation shield moved over the northeastern United States and New England (Figs. 1b,c). By 0000 UTC (9 February 2013), reflectivity values over Long Island were in excess of 50 dBZ. Between 0600 and 1200 UTC (9 February 2013), the surface low moved slowly eastward and precipitation continued to fall over New England (Figs. 1d,e). By 1800 UTC, precipitation was no longer falling over Long Island and Connecticut (Fig. 1f).
The observed mosaic of composite reflectivity (shaded) and RAP model analyses of sea level pressure (hPa; solid) and 2-m temperature (°C; dashed). The 273-K (0°C) isotherm is set in boldface.
Citation: Weather and Forecasting 29, 6; 10.1175/WAF-D-14-00056.1
The 24-h accumulated liquid-equivalent precipitation, beginning at 1200 UTC, was in excess of 38 mm over south-central Connecticut and most of Long Island (Fig. 2a). Accumulations exceeded 50 mm over central and eastern Long Island and southeastern New Jersey. Precipitation type observations collected by the National Oceanic and Atmospheric Administration/National Severe Storms Laboratory’s (NOAA/NSSL) mPING project (Elmore et al. 2014) show a transition of surface precipitation type between 1200 and 1800 UTC (Fig. 2b). Rain was falling over southern New Jersey, southeastern Pennsylvania, and along the southern coast of Long Island, while a narrow zone of freezing rain and ice pellets was positioned over central New Jersey and along northern Long Island. Snow was the predominant precipitation type over Connecticut, Rhode Island, Massachusetts, and most of New York State. As time progressed, the region of freezing rain and ice pellets shifted slightly southward (Figs. 2c,d) with most locations reporting snow by 1200 UTC 9 February (Fig. 2e).
(a) Stage IV analyzed liquid-equivalent 24-h accumulated precipitation beginning at 1200 UTC 8 Feb 2013, (b)–(e) observed precipitation type from mPING, and (f) snow depth from NWS observers and the Community Collaborative Rain, Hail & Snow Network (CoCoRaHS; Cifelli et al. 2005).
Citation: Weather and Forecasting 29, 6; 10.1175/WAF-D-14-00056.1
By 1200 UTC (9 February 2013), total snow accumulations exceeded 50 cm over central Long Island, Connecticut, and Massachusetts (Fig. 2f). Heavy snowfall and blizzard conditions occurred from northern New Jersey, inland to New York, and northward through Maine (Fig. 3). Storm-total snow accumulations of 30–61 cm were common, with amounts surpassing 61 cm over a SW–NE-oriented band from Long Island to southern Maine. Maximum accumulations up to 102 cm, as well as snowfall rates exceeding 15 cm h−1, were reported in parts of Connecticut. In addition to record-setting snow accumulations, significant coastal flooding and hurricane-force wind gusts were recorded along the coast (Picca et al. 2014). Impacts of the event included at least 18 fatalities, thousands of flight cancellations at major airports, travel bans, and hundreds of thousands of customers throughout the region left without electricity for several days (NWA 2014; Associated Press 2014).
Storm-total snow accumulation from the 8–9 Feb 2013 blizzard. (Courtesy B. Vincent, of the NWS Forecast Office in Raleigh, North Carolina.)
Citation: Weather and Forecasting 29, 6; 10.1175/WAF-D-14-00056.1
3. Observational analysis
Here, we present polarimetric radar observations of some of the intriguing features of the Northeast blizzard during the observational period from 1000 UTC 8 February through 0400 UTC 9 February 2013. This time period includes the onset and intensification phases of the system and the subsequent period during which the Midwest and mid-Atlantic surface low pressure centers merged, resulting in a strengthening of the surface low, rapid cooling of the thermal profile, lower-density snow, and reduced ZH across the region.
a. High ZH near the surface
One of the more remarkable features of this storm was the extremely high ZH observed near the surface (especially after 2100 UTC), which reached magnitudes greater than 50 dBZ in regions of wet snow. This is considerably higher than the ~40-dBZ maximum typically reported in previous studies (e.g., Ryzhkov and Zrnić 1998). Table 1 provides a summary of typical polarimetric values of cold-season precipitation types. Note that in the case of an anomalous event, it is possible to observe values outside of the ranges provided.
Typical ranges of values of some polarimetric variables of cold-season precipitation types, at S band. [Adapted from Ryzhkov and Zrnić (1998), Straka et al. (2000), Park et al. (2009), Andrić et al. (2013), Kumjian (2013), Kumjian et al. (2013), 2014, and Picca et al. (2014).]
During the time period shown in Figs. 4a and 4d, the maximum ZH near the surface (at 0.5° elevation, and below 1.5 km) remained >50 dBZ and the precipitation band appeared to pivot to a more north–south orientation as the low pressure center traversed northeastward. The type of precipitation at 2216 UTC, as inferred from the polarimetric observations, is indicated in Figs. 4a–c. Pure, dry snow exists north of the model-indicated 0°C wet-bulb temperature TW isotherm (ZH < 35 dBZ, ZDR of 0–0.5 dB, and ρhv near 1). Wet snow, snow, and ice pellets are indicated south of the 0°C isotherm and in regions of ZH ranging from 40 to 60 dBZ, ZDR from 1.2 to 3.5 dB, and ρhv as low as 0.75. These precipitation types are in agreement with mPING observations and those taken by meteorologists and cooperative observers for the NWS Forecast Office in Upton, which indicate snow and a mixture of ice pellets, snow, and various snow mixtures to the north and south of the 0°C isotherm, respectively.
PPI displays of the polarimetric variables at (a)–(c) 2216 UTC 8 Feb and (d)–(f) 0236 UTC 9 Feb 2013 at 0.5° elevation. The 0°C RAP model TW at the surface is overlaid (boldface, dashed). At 2216 UTC, pure dry snow was located within colder temperatures north of the 0°C isotherm, while wet snow and mixed-phase hydrometeors occurred within warmer temperatures south of the 0°C isotherm in (a)–(c). The solid black line indicates the location of the 144° azimuth RHI in Fig. 5. At 0236 UTC, dry snow was predominant, while wet snow and ice pellets were also observed within the max ZH region, within negative surface temperatures, north of the 0°C isotherm in (d)–(f).
Citation: Weather and Forecasting 29, 6; 10.1175/WAF-D-14-00056.1
Figure 5 depicts a reconstructed RHI through a high-reflectivity band at 2216 UTC. Regions of exceptionally large ZH (>50 dBZ) are confined to a shallow layer below 1.5 km throughout the entire observational period, while ZH greater than 24 dBZ remains below approximately 6 km. According to Ganetis et al. (2013), these extremely high ZH values near the surface may be partly attributed to frontogenesis. They found that, as the system intensified, the environment became more unstable and convective as temperatures exceeded 0°C at approximately 850–800 hPa in the warm layer within the band, which was associated with the trough of warm air aloft as well as subsidence warming within the band circulation; these processes resulted in the production of sleetlike hydrometeors and the ZH values >50 dBZ. After frontogenesis and subsidence warming decreased in magnitude, cold-air advection became dominant and led to a rapid transition (i.e., within an hour) to lower ZH values (Ganetis et al. 2013). The corresponding ZDR is high and ρhv is low in the shallow layer of extremely high ZH, which testifies that large wet snowflakes or even melting ice hydrometeors (e.g., graupel and/or hail) are dominant scatterers there, despite the model profile indicating that no melting would be expected. This illustrates an apparent inconsistency between the type of precipitation identified by the radar and the wet-bulb temperatures retrieved from the RAP model at distances closer than 20 km from the radar, where TW < 0°C in the whole depth of the atmosphere (where melting is not anticipated). The radar reflectivity is highest just below the “nose” of the melting layer, where the transition from dry hydrometeors with low ZDR and high ρhv to melting precipitation with high ZDR and low ρhv occurs at lower levels than in surrounding areas. This is an indication of larger-sized rimed snow and/or hail, which is melting at lower heights due to its higher terminal velocity. It is possible that ice crystals generated aloft descend into a layer of localized convection and abundant moisture along the front and start growing via riming, which results in rapid enhancement of ZH in a relatively shallow surface layer. The pocket of low ZDR above the high-ZH region indicates the localized region where riming is occurring, leading to larger fall speeds of the hydrometeors and a lowering of the melting level. Another possible mechanism for the depressed melting level is localized cooling due to the melting of heavy precipitation and large aggregates (e.g., Stewart 1984; Stewart et al. 1984; Oraltay and Hallett 2005). This type of cooling can produce local near-0°C isothermal layers, allowing partially melted hydrometeors to reach the surface (e.g., Findeisen 1940; Wexler et al. 1954; Szeto et al. 1988; Ryzhkov et al. 2011). Figure 6 provides a conceptual model illustrating these two possible mechanisms.
RHIs at 2216 UTC 8 Feb 2013 of (a) ZH, (b) ZDR, (c) ρhv, and (d) KDP at 144° azimuth. In each RHI, ZH is contoured at 20, 30, 40, and 50 dBZ. Contours of RAP model TW are also overlaid (°C). The model-indicated 0°C isotherm is identified by the white contour.
Citation: Weather and Forecasting 29, 6; 10.1175/WAF-D-14-00056.1
Conceptual model showing two possible mechanisms for producing a downward excursion in the radar bright band: (a) a localized region of enhanced cooling due to evaporation and melting (indicated by the yellow-shaded oval) associated with heavier precipitation and (b) a localized region of enhanced updraft (indicated by the yellow-shaded arrow) that produces rimed particles that have a larger fall speed than the surrounding hydrometeors. Solid arrows offer a comparison of the relative fall speeds between lightly aggregated stellars/dendrites and heavily rimed stellars/dendrites. When compared to (a), rimed particles in (b) would likely produce slightly enhanced ZH and reduced ZDR above the bright band and slightly enhanced ZH and ZDR below the bright band. Each process might be responsible for possible changes in surface precipitation type when occurring at sufficiently low altitudes.
Citation: Weather and Forecasting 29, 6; 10.1175/WAF-D-14-00056.1
Surface reports of a mixture of ice pellets (mPING) and “large sleet” resembling pea-sized hail (R. Hanrahan 2013, personal communication; Picca et al. 2014) confirmed that, in addition to contributions from melting snow in this region of maximum ZH, there may have also been contributions from wet growth and melting of refrozen hydrometeors. The presence of wet snow and melting graupel/hail beneath the melting layer, such as exists in this event, can serve to alert forecasters to the potential for localized convection and the associated riming of ice crystals aloft to locally alter the precipitation type at the surface. It is interesting to compare the location of the RAP 0°C isotherm with the apparent 0°C isotherm in the radar observations (given by the strong north-to-south gradients in ZDR and ρhv along the northern shore of Long Island; Figs. 4b,c). The RAP 0°C isotherm is approximately 40 km north of this line. Without 2-m temperature observations over Long Island Sound, one cannot definitely state the RAP model is errant. However, the comparison does highlight the advantage of using dual-polarization data over model data to distinguish where transitions in precipitation type exist.
By 0236 UTC (Figs. 4d–f), the region of maximum ZH > 50 dBZ shifted north of KOKX (over Connecticut), and north of the model-indicated 0°C isotherm (e.g., Fig. 4d; 0236 UTC 9 February 2013), into colder temperatures. After the main snowband traversed north of the 0°C isotherm, polarimetric observations show that dry snow became the predominant precipitation type, while ice pellets were also observed at the surface (Figs. 4d–f; mPING). The polarimetric variables in Figs. 4e and 4f also depict a north–south transition line extending from central Long Island to the southern shore of Connecticut that is not reflected in the RAP-analyzed 2-m temperature. Indeed, mPING observations reveal that the precipitation type to the west of this line is snow and to the east is ice pellets. Given that the height of the elevation angle is between 0.5 and 1.5 km above ground level in this region, the disagreement between the RAP analysis [and the RAP temperatures interpolated to the radar surface (not shown)] and the radar observations may indicate that the RAP model failed to capture the temperature field accurately at this time, again underscoring the value in using dual-polarization observations for inferring precipitation type transition zones. It is also important to note that the RAP data aloft at this time (not shown) exhibited no evidence of an elevated warm layer in this region, precluding the possibility of a warm layer causing the occurrence of the mixed-phase precipitation in this localized region.
b. Differential attenuation
Another prominent feature observed throughout the event was the remarkable differential attenuation manifested by large negative values of ZDR down to −2 dB along the radar beams propagating through regions of heavy, wet snow (e.g., Figs. 5 and 7). To the authors’ knowledge, the differential attenuation values presented in this manuscript are larger than previously reported for other winter storms. After a radar beam propagates through heavy wet snow and mixed-phase hydrometeors, the incident electromagnetic energy is absorbed and scattered, resulting in loss of power and reduction in echoes farther down the radial (Doviak and Zrnić 1993, p. 38). Given the occurrence of intense (>50 dBZ) heavy wet snow at the surface, it is not surprising that attenuation and differential attenuation observations in this system reached magnitudes that appear to exceed those previously documented for S-band radar observations.
RHIs at 2233 UTC 8 Feb 2013 of (a) ZH, (b) ZDR, (c) ρhv, and (d) ΦDP at 216° azimuth. In each RHI, ZH is contoured at 20, 30, 40, and 50 dBZ. Contours of the RAP model TW are also overlaid (°C). The model-indicated 0°C isotherm is identified by the white contour. The max TW within the model-indicated warm layer was approximately 4°C.
Citation: Weather and Forecasting 29, 6; 10.1175/WAF-D-14-00056.1
Figure 8 displays line plots of ΦDP, ZDR, and ρhv as functions of range, at the 2.4°-elevation scan along the 2233 UTC 216° azimuth. It should be noted that ΦDP has been smoothed, according to the preprocessing procedures detailed by Ryzhkov et al. (2005). Beginning at a range of ~10 km, there was a rapid increase in ΦDP, with values reaching 125° at 100 km (Fig. 8a). In wet snow, ZDR peaked at ~2.7 dB and then decreased and remained mostly negative (due to substantial differential attenuation) over a large range beyond approximately 30 km (Fig. 8b). At ranges between 10 and 35 km, ρhv was reduced when the beam propagated through the wet snow and mixed-phase precipitation at the lower heights near the surface (Fig. 8c). Note that KDP values in wet snow were approximately 1° km−1 (Fig. 5d), which is consistent with high KDP values of ~1°–1.3° km−1 in melting snow documented by Ryzhkov and Zrnić (1998).
Line plots at 2233 UTC 8 February 2013 of (a) ΦDP, (b) ZDR, and (c) ρhv as functions of range from KOKX, at 2.4° elevation, along the 216° azimuth. At this time, the 2.4°-elevation scan was sampling the heavy, wet snowband with ZH up to 60 dBZ, located between ranges of approximately 10 and 55 km.
Citation: Weather and Forecasting 29, 6; 10.1175/WAF-D-14-00056.1
To better quantify the factor 
Differential attenuation occurred throughout the entire observational period, becoming more distinct as the event progressed. The 1.45° plan position indicator (PPI) ZDR imagery at 0001 UTC 9 February 2013 (Fig. 9) depicts differential attenuation, indicated by ZDR values from −2 to 0 dB to both the southwest and northeast of KOKX. These negative ZDR values occur down radial of the region of ZH > 55 dBZ, ZDR > 1 dB, and ρhv < 0.97 (Fig. 9) within a 50-km range of the radar. An analysis of the polarimetric observations at this time suggests the precipitation type in the region of greatest attenuation is wet snow. Special observations taken by observers at Stony Brook University with an ice microscope [presented in Ganetis et al. (2013) and Picca et al. (2014)] also indicate there was a mixture of different frozen habits, including a new variety of hydrometeor coined “asteroid ice.” This type of hydrometeor had the appearance of small, irregular hail and exhibited a heavily rimed exterior with a diameter of approximately 1.5–2.5 mm, indicating a formation mechanism similar to that of hail: it is expected that ice crystals were generated aloft due to depositional growth, descended into the updraft, and then became heavily rimed due to abundant supercooled water within a warm, moist, and shallow region above the surface.
PPIs at 0001 UTC 9 Feb 2013 of (a) ZH, (b) ZDR, (c) KDP, and (d) ρhv at 1.45° elevation. Distances are relative to KOKX.
Citation: Weather and Forecasting 29, 6; 10.1175/WAF-D-14-00056.1
c. Downward excursion of the melting layer
A downward excursion of the melting-layer bright band (MLBB) became evident by 1300 UTC 8 February (e.g., 1354 UTC 8 February; Fig. 10). This feature is indicated by the north end of the semicircular region of low ρhv, moderate-to-high ZDR, and low-to-moderate ZH located along the southern shore of Long Island. During this time, mPING observations show a corresponding transition from rain and ice pellets in the south to snow in the north (Figs. 2b and 10d). Figure 11 provides an RHI perspective of the downward excursion of the melting layer toward the surface, supplemented by a range-versus-height schematic to aid in the interpretation of this phenomenon. Areas where the 0.5° radar beam (with a 1° beamwidth) is at least partially contaminated by the MLBB (high ZDR and low ρhv) are indicated by the yellow shading, and areas where the beam passes through pure rain (low ZDR and high ρhv; center of the semicircle) are indicated by the green shading.
PPIs at 1354 UTC 8 Feb 2013 of (a) ZH, (b) ZDR, and (c) ρhv at 0.5° elevation. (d) The mPING surface precipitation type reports for the 30 min centered at 1354 UTC (i.e., 1339–1409 UTC). The downward excursion of the MLBB to the surface (distinct in ρhv and ZDR) was associated with an observed transition line of precipitation types at the surface.
Citation: Weather and Forecasting 29, 6; 10.1175/WAF-D-14-00056.1
RHIs at 2350 UTC 8 Feb 2013 of (a) ZH, (b) ZDR, and (c) ρhv at 143° azimuth, illustrating a downward excursion of the MLBB. (d) Range-vs-height schematic of the MLBB sloping downward toward the surface, with a hypothetical 0.5°-elevation scan (with a 1° beamwidth) overlaid. The radar beam increases in range and height from KOKX. Areas where the 1°-wide radar beam is contaminated by the bright band (high ZDR and low ρhv) are indicated by the yellow shading and areas where the beam passes through pure rain (low ZDR and high ρhv) are indicated by the green shading.
Citation: Weather and Forecasting 29, 6; 10.1175/WAF-D-14-00056.1
As time progressed, a bubble-like feature became evident in the ρhv and ZDR imagery, located slightly northeast of KOKX (i.e., centered at approximately x = 10 km, y = 20 km in Figs. 12b,c and x = 8 km, y = 35 km in Figs. 13b,c); it was initially attached to the north end of the MLBB semicircle and then curiously fluctuated in size and distance from it during approximately 3 h. The bubble was characterized by enhanced ZDR (1–2.5 dB) and reduced ρhv (as low as 0.9), and was first observed just after 1400 UTC. The semicircular MLBB-to-ground signature became progressively more distinct with increasing elevation and time. By 1459 UTC, the bubble had become larger and more distinct (Fig. 12). Eventually, it began to detach from the semicircle, becoming nearly completely detached by 1557 UTC (Fig. 13). By 1659 UTC, it had nearly completely dissipated, leaving the semicircle with a more distinct horizontal line (i.e., transition line) of reduced ρhv and enhanced ZDR (Figs. 14b,c). According to the polarimetric imagery, the bubble was located north of the transition line, within the region of snow indicated by ρhv > 0.996, ZDR < 0.75, and ZH < 30 dBZ (e.g., Figs. 12a–c). Therefore, the bubble was potentially a pocket of wet snow associated with local convection and riming aloft within the associated updraft. Before dissipating, it propagated into the colder air temperatures, predominantly carried by the prevailing winds from the south. By 1900 UTC, the downward excursion of the MLBB had become less distinct, as heavy precipitation bands moved northward and meshed with the semicircle, appearing as a widespread region of precipitation with a horizontal leading transition line (not shown).
As in Fig. 10, but for 1459 UTC 8 Feb 2013. The mPING observations are shown for the 30 min centered at 1459 UTC (i.e., 1444–1514 UTC). The bubble-like feature is centered at approximately x = 10 km, y = 20 km in (b) and (c).
Citation: Weather and Forecasting 29, 6; 10.1175/WAF-D-14-00056.1
As in Fig. 10, but for 1557 UTC 8 Feb 2013. The mPING observations are shown for the 30 min centered at 1557 UTC (i.e., 1542–1612 UTC). The bubble-like feature is centered at approximately x = 8 km, y = 35 km in (b) and (c).
Citation: Weather and Forecasting 29, 6; 10.1175/WAF-D-14-00056.1
As in Fig. 10, but for 1659 UTC 8 Feb 2013. The mPING observations are shown for the 30 min centered at 1659 UTC (i.e., 1644–1714 UTC).
Citation: Weather and Forecasting 29, 6; 10.1175/WAF-D-14-00056.1
During the evolution period, the polarimetric detection of the MLBB extension to the surface was verified by an observed abrupt transition of mPING surface precipitation types. Rain was predominant to the south of the transition line, over the southern coast of Long Island, while a narrow zone of freezing rain and ice pellets was positioned along the line, over the central and northern regions of Long Island, with snow predominant to the north (Figs. 2b, 12d, 13d, and 14d). After the dissipation of the polarimetric signature, the region of freezing rain and ice pellets shifted slightly southward (Figs. 2c,d). Meteorologists and cooperative observers from the NWS Forecast Office in Upton reported a finer temporal resolution of a transition of precipitation type to sleet along the transition line. Therefore, considering the transition line corresponded well with these ice pellet reports, there potentially may have been some contribution of wet growth and melting of refrozen hydrometeors in addition to the melting effects of the downward extension of the MLBB.
d. Depositional growth layer
An elevated horizontal layer of enhanced ZDR, KDP, and reduced ρhv was observed above the melting layer, within the comma-head region of the cyclone. The enhanced ZDR likely signified rapid depositional growth of large, horizontally oriented ice crystals in the subfreezing temperatures aloft. Further aggregation of large ice crystals causes a decrease of ZDR below. Figure 15 illustrates this intriguing feature along the 0° azimuth at 0305 UTC 9 February 2013. Enhanced ZDR (1–3 dB) and KDP (up to 2° km−1) and reduced ρhv (as low as 0.9) were observed in a layer above ~3 km, within the model-indicated −12° and −20°C TW isotherms (Fig. 15). The largest KDP values were preferentially located between the −14° and −18°C isotherms and there was a pronounced vertical ZH gradient directly below the layer, consistent with observations documented by Kennedy and Rutledge (2011), Andrić et al. (2013), and Bechini et al. (2013). Also at this time, ZH near the surface exceeded 50 dBZ, while exceptionally heavy snow (fine, low-density snow with little rime) was found at the surface by observers at the NWS Forecast Office in Upton.
As in Fig. 5, but for 0305 UTC 9 Feb 2013 at 0° azimuth. In each RHI, ZH is contoured at 10, 20, 30, and 40 dBZ. Contours of RAP model TW are also overlaid (°C), with the −12° and −20°C contours in white.
Citation: Weather and Forecasting 29, 6; 10.1175/WAF-D-14-00056.1
The depositional growth layer was first observed after 1200 UTC, as much as 10 h prior to the greatest ZH values near the surface and when large ZH values were most widespread. As time progressed, and as ZH exceeded 50 dBZ near the surface, the layer became more evident and occurred above the regions of large ZH. By 2100 UTC, the layer had become yet more evident (particularly in terms of KDP) as ZH further increased in the heavy snowband and as colder and drier air was introduced near the surface. This layer was particularly apparent in the PPI plots above the 6°-elevation scans. Figure 16 (2106 UTC) displays a striking example of double rings of enhanced ZDR, reduced ρhv, and slightly enhanced KDP encircling KOKX; the outer ring indicates the depositional growth layer (between approximately −12° and −20°C), while the inner ring indicates the MLBB. The depositional growth layer occurred at greater heights southeast of KOKX, where temperatures were warmer at the surface, while the layer occurred at lower heights northwest of KOKX, above colder surface temperatures. At this time, mPING surface precipitation type reports indicate snow, wet snow, and ice pellets. Also at this time, and during the following hour, cooperative observers from the NWS office in Upton reported large dendrites and aggregates within heavy snow along the northern shore. Lightly rimed dendrites were falling at Stony Brook University, as observed by Ganetis et al. (2013) at 0409 UTC. Note that this signature (Fig. 16) preceded the rapid increase in ZH near the surface (>50 dBZ) by less than 1 h. It was less visible between 2300 and 0100 UTC 9 February 2013, but reappeared and became most distinct from 0100 to 0400 UTC, particularly north of KOKX, above the greatest ZH values over Connecticut (not shown). The physical explanation for the waning of the layer is not yet clear, but it may have been attributed to the instability being relieved for a short time, before being reinvigorated thereafter.
PPIs at 2106 UTC 8 Feb 2013 of (a) ZH, (b) ZDR, (c) ρhv, and (d) KDP at 19.5° elevation. Contours of RAP model TW on the conical surface are overlaid. The boldface contours represent the 0°C wet-bulb isotherms, while the white contours represent the −12° and −20°C isotherms. Double rings of enhanced ZDR, reduced ρhv, and slightly enhanced KDP encircle KOKX; the outer ring indicates the depositional growth layer, while the inner ring indicates the MLBB. This signature preceded the rapid increase in ZH near the surface by less than 1 h.
Citation: Weather and Forecasting 29, 6; 10.1175/WAF-D-14-00056.1
e. Informative polarimetric artifacts
Several types of artifacts were observed in the polarimetric data. These features should not be overlooked, as they can also provide valuable information about storm microphysical processes (Kumjian et al. 2013).
1) Depolarization streaks
Depolarization streaks were observed during the time of heaviest precipitation, when the low-level ZH was >50 dBZ. According to Ryzhkov et al. (2011), weak convective updrafts in winter storms can produce a tangible amount of graupel and charge separation sufficient to generate electric fields. Furthermore, strong electrostatic fields can change the orientation of ice crystals atop these updrafts, causing the transmitted radar signal to become depolarized (Ryzhkov and Zrnić 2007; Ryzhkov et al. 2011). The resulting polarimetric signatures reveal the depolarization through radial streaks of enhanced positive and/or negative ZDR. Therefore, the numerous depolarization streaks observed during the Northeast blizzard event provided evidence of widespread electrification in the cloud.
Depolarization streaks occurred in the southern comma-head region of the synoptic-scale low pressure system from 2100 through 0300 UTC 9 February 2013, and were most frequent and pronounced during the 2300 UTC hour, when ZH exceeded 55 dBZ near the surface. The location of these streaks within the system supports the work by Rauber et al. (2014), who documented a climatology of electrification in the comma heads of 16 continental winter cyclones. They found that lightning originated in elevated convective cells within the southern region of the comma head, due to advancement of upper-tropospheric dry air within the dry slot over the low-level moist air. Although beyond the scope of this work, it is possible that a similar mechanism may have been responsible for triggering cloud electrification in the February 2013 Northeast blizzard.
The depolarization streaks were observed at the 0.5°–9.89°-elevation angles, especially at 1.45°–4.3°, and originated at uncharacteristically low heights (compared to streaks typically observed in warm-season convection) atop convective updrafts in regions of heavy, wet snow. Origination heights (i.e., the heights of the tips of the streaks) ranged from 1 to 5 km but were primarily at approximately 3 km, where model-indicated TW values were from −4° to −6°C. Average origination heights were slightly lower after 0000 UTC, when compared to those prior to 0000 UTC. The PPI image at 2314 UTC (Fig. 17) provides an example of the streaks observed during the event. The ZDR imagery displays radial streaks of enhanced positive and negative ZDR, with magnitudes of 5 and 2 dB, respectively. The streaks originated atop convective updrafts in regions of heavy wet snow (ZH > 50 dBZ, ρhv as low as 0.9) and in the vicinity of a cluster of total lightning flashes detected by the Earth Networks Total Lightning Network (Fig. 17). These observations bolster the results of Rauber et al. (2014), who documented that lightning flashes typically occur on the southern side of the comma head, where dry air aloft may overrun low-level moist air and trigger convective instability.
PPIs at 2314 UTC 8 Feb 2013 of (a) ZH and (b) ZDR at 3.34° elevation. Depolarization streaks are indicated by the radial streaks of positive and negative ZDR. Lightning flash locations at 2316 UTC, from the Earth Networks Total Lightning Network, are indicated by the white circles.
Citation: Weather and Forecasting 29, 6; 10.1175/WAF-D-14-00056.1
2) Nonuniform beamfilling
Another polarimetric artifact observed was nonuniform beamfilling (NBF; Ryzhkov 2007), revealed by a wedge of radial streaks of reduced ρhv (e.g., Fig. 18d). The RHI plots in Fig. 5 illustrate the NBF, with ρhv as low as 0.75 (Fig. 5c) and ΦDP as large as 95° (not shown). The wedge of low ρhv indicated large vertical gradients of ΦDP, due to a nonuniform mixture of precipitation types and sizes within the radar beam cross sections. Here, the diverse precipitation types were likely heavy wet snow and dry snow, heavy sleet, and wet-growth ice hydrometeors (Ganetis et al. 2013; Picca et al. 2014). Furthermore, progressive beam broadening enhanced the impact of NBF with increasing range from the radar. Significant accumulation of ΦDP down the radial provided further evidence of the melting of large aggregates and other larger ice hydrometeors (i.e., graupel, large hail-like ice pellets) near the surface.
As in Fig. 9, but for 0259 UTC 9 Feb 2013. NBF was indicated by the wedge of radial streaks of reduced ρhv.
Citation: Weather and Forecasting 29, 6; 10.1175/WAF-D-14-00056.1
3) A “SNOW FLARE” signature
An interesting “snow flare” signature, reminiscent of a hail three-body scatter signature, was observed northeast of KOKX from 0000 to 0200 UTC 9 February at the 6°–19.5°-elevation scans, revealed by enhanced values of ZDR and reduced ρhv, collocated with low ZH and KDP (e.g., Fig. 19). The signature flared outward from the radar and appeared similar to a three-body scattering signature (or hail flare) that is frequently associated with the presence of hailstones (e.g., Zrnić 1987; Hubbert and Bringi 2000; Kumjian et al. 2010; Zrnić et al. 2010; Picca and Ryzhkov 2012). However, in this case, the signature appears to have been associated with very large snowflakes and wet-growth ice hydrometeors. The flare occurred coincident with surface observations of snow and ice pellets in the mPING data. However, it also occurred at the same time that Ganetis et al. (2013) reported the presence of hail-like large ice pellets (or asteroid ice). The coincidental occurrence of asteroid ice with the formation of the snow flare supports that its growth process is analogous to that of hail. However, detailed analysis of RHIs in different azimuthal directions shows that strong sidelobe contamination should not be excluded as a possible source of the snow flare. As the flare began to appear less distinct, wet-growth ice hydrometeors were on a downward trend, with a decreasing density of rimed snow from the NWS Forecast Office in Upton toward the northern shore. During the hour following the disappearance of this signature, heavy, much less dense snow (with some aggregates) was observed at the Forecast Office in Upton and surrounding areas, while there was a rapid decrease in sleet, graupel, and heavily rimed snow.
PPIs at 0108 UTC 9 Feb 2013 of (a) ZH, (b) ZDR, (c) KDP, and (d) ρhv at 19.5° elevation. The snow flare was revealed by enhanced values of ZDR and reduced ρhv, collocated with low ZH and KDP, flaring outward from KOKX. This artificial signature was associated with very large snowflakes, ice pellets, and anomalous hail-like ice hydrometeors (or asteroid ice) at the surface.
Citation: Weather and Forecasting 29, 6; 10.1175/WAF-D-14-00056.1
4. Summary and conclusions
The 8–9 February 2013 Northeast blizzard was a unique event, exhibiting several intriguing dual-polarization radar signatures. This study investigates the evolution and nature of these signatures, and the thermodynamic conditions within which they developed, to obtain a better understanding of the fundamental microphysical processes within this system. Polarimetric data (from the S-band KOKX radar) were analyzed alongside RAP model wet-bulb temperature analyses, as well as surface precipitation type observations from both mPING and the NWS Forecast Office in Upton, New York, for interpretation of polarimetric signatures.
Values of ZH during this event were extraordinary for a winter storm, exceeding 50 dBZ and reaching as high as 60 dBZ within a shallow layer just above the surface. Also, as the incoming snowbands proceeded northward, the polarimetric data exhibited an exceptionally distinct transition from frozen to unfrozen precipitation, providing detail that was often unmatched by the numerical model output. During this event, the polarimetric observations were critical for accurately assigning the transition from liquid to frozen precipitation, illustrating how dual-polarization radar data could be a potentially valuable tool for forecasters when nowcasting transitional winter precipitation. Another prominent feature of the event was the remarkable differential attenuation, resulting from the radar beam propagating through regions of heavy wet snow and mixed-phase precipitation. These differential attenuation observations reached magnitudes that exceed anything previously documented for S-band radar observations in snow. This study also documents a downward excursion of the MLBB to the surface, characterized by reduced ρhv and locally maximized ZH and ZDR; this feature was correlated with an abrupt transition line of precipitation types at the surface.
Some of the most distinctive signatures observed during the event were elevated horizontal layers of enhanced ZDR and KDP, and reduced ρhv, located above the environmental freezing layer and within the comma-head region of the cyclone. The enhanced ZDR values likely signified the presence of large, horizontally oriented ice crystals at the subfreezing temperatures aloft, near the model-predicted −15°C TW isotherm, where the conditions for rapid depositional growth are most favorable. These depositional growth layers appeared to be correlated with the increase in heavy snowfall; ice crystals were generated aloft, aggregated, descended, and then contributed to the large ZH values near the surface. The layers appeared increasingly more evident as the period of greatest ZH values neared, with the initial layer observations preceding the greatest surface ZH by several hours, demonstrating the potential utility of this signature for nowcasting increases in precipitation at the surface.
Several polarimetric artifacts were also observed and provided valuable information about the system’s microphysical processes. Distinct depolarization streaks occurred with frequency during the 2300 UTC hour, when ZH exceeded 55 dBZ near the surface. These radial streaks of positive and negative ZDR indicated regions of atmospheric electrification (and possible regions of supercooled water), and they originated at uncharacteristically low heights, atop weak convective updrafts in regions of heavy wet snow. The effects of nonuniform beamfilling were also observed during the event, indicating large gradients of ΦDP within the radar resolution volume, due to a nonuniform mixture of precipitation types and sizes within the radar beam cross sections. Finally, a “snow flare” of reduced ρhv, enhanced ZDR, moderate KDP, and low ZH flared outward from the radar and appeared similar to a three-body scattering signature commonly reported in hailstorms; this signature could also be due to sidelobe contamination. This feature was associated with very large snowflakes and ice hydrometeors at the surface, including anomalous ice hydrometeors (Ganetis et al. 2013), which had the appearance of small, irregular hailstones.
This study provides a next step toward understanding the fundamental microphysical processes within winter precipitation and how polarimetric signatures relate to larger-scale storm structure and evolution. The radar signatures investigated herein convey the value of polarimetry in identifying features undetectable in conventional radar data. These signatures are associated with hazardous winter weather conditions that cause havoc on the public and transportation sectors, both at the surface and in the air. Therefore, polarimetry provides a valuable tool for short-term detection and prediction of winter weather precipitation types, especially transitional events.
Acknowledgments
The authors thank the NOAA/National Weather Service employees who maintain and operate the KOUN polarimetric radar for research-grade operations. Funding was provided by the NOAA/Office of Oceanic and Atmospheric Research under NOAA–University of Oklahoma Cooperative Agreement NA11OAR4320072, the U.S. Department of Commerce, and by the U.S. National Weather Service, Federal Aviation Administration, and Department of Defense program for modernization of NEXRAD. The authors also acknowledge the NSF’s support under Grant 1193948. We also thank Earth Networks for providing the total lightning data. Additionally, thank you to Dr. Stephen Cocks for providing a review of this manuscript, and three anonymous reviewers for useful comments that helped improve the manuscript.
REFERENCES
Andrić, J., Kumjian M. R. , Zrnić D. S. , Straka J. M. , and Melnikov V. M. , 2013: Polarimetric signatures above the melting layer in winter storms: An observational and modeling study. J. Appl. Meteor. Climatol., 52, 682–700, doi:10.1175/JAMC-D-12-028.1.
Associated Press, cited 2014: Northeast slowly recovering from blizzard. [Available online at http://bigstory.ap.org/article/northeast-trying-get-back-track-after-storm.]
Bechini, R., Baldini L. , and Chandrasekar V. , 2013: Polarimetric radar observations in the ice region of precipitating clouds at C-band and X-band radar frequencies. J. Appl. Meteor. Climatol., 52, 1147–1169, doi:10.1175/JAMC-D-12-055.1.
Cifelli, R., Doesken N. , Kennedy P. , Carey L. D. , Rutledge S. A. , Gimmestad C. , and Depue T. , 2005: The Community Collaborative Rain, Hail, and Snow Network: Informal education for scientists and citizens. Bull. Amer. Meteor. Soc., 86, 1069–1077, doi:10.1175/BAMS-86-8-1069.
Doviak, R. J., and Zrnić D. S. , 1993: Doppler Radar and Weather Observations. 2nd ed. Academic Press, 562 pp.
Elmore, K. L., Flamig Z. L. , Lakshmanan V. , Kaney B. T. , Farmer V. , and Rothfusz L. P. , 2014: mPING: Crowd-sourcing weather reports for research. Bull. Amer. Meteor. Soc., doi:10.1175/BAMS-D-13-00014.1, in press.
Field, P., Hogan R. , Brown P. , Illingworth A. , Choularton T. , Kaye P. , Hirst E. , and Greenway R. , 2004: Simultaneous radar and aircraft observations of mixed-phase cloud at the 100 m scale. Quart. J. Roy. Meteor. Soc., 130A, 1877–1904, doi:10.1256/qj.03.102.
Findeisen, W., 1940: The formation of the 0°C isothermal layer and fractocumulus under nimbostratus. Meteor. Z., 6, 882–888.
Ganetis, S. A., Colle B. A. , Sienkiewicz M. J. , Schultz D. M. , Heinselman P. L. , and Novak D. R. , 2013: Evolution of an intense mesoscale snowband during the 8–9 February 2013 Northeast U.S. blizzard. 15th Conf. on Mesoscale Processes, Portland, OR, Amer. Meteor. Soc., P16.5. [Available online at https://ams.confex.com/ams/15MESO/webprogram/Paper228054.html.]
Hogan, R., Field P. , Illingworth A. , Cotton R. , and Choularton T. , 2002: Properties of embedded convection in warm-frontal mixed-phase cloud from aircraft and polarimetric radar. Quart. J. Roy. Meteor. Soc., 128, 451–476, doi:10.1256/003590002321042054.
Hubbert, J. C., and Bringi V. N. , 2000: The effects of three-body scattering on differential reflectivity signatures. J. Atmos. Oceanic Technol., 17, 51–61, doi:10.1175/1520-0426(2000)017<0051:TEOTBS>2.0.CO;2.
Hubbert, J. C., Ellis S. M. , Dixon M. , and Meymaris G. , 2010a: Modeling, error analysis, and evaluation of dual-polarization variables obtained from simultaneous horizontal and vertical polarization transmit radar. Part I: Modeling and antenna errors. J. Atmos. Oceanic Technol., 27, 1583–1598, doi:10.1175/2010JTECHA1336.1.
Hubbert, J. C., Ellis S. M. , Dixon M. , and Meymaris G. , 2010b: Modeling, error analysis, and evaluation of dual-polarization variables obtained from simultaneous horizontal and vertical polarization transmit radar. Part II: Experimental data. J. Atmos. Oceanic Technol., 27, 1599–1607, doi:10.1175/2010JTECHA1337.1.
Kain, J. S., Goss S. M. , and Baldwin M. E. , 2000: The melting effect as a factor in precipitation-type forecasting. Wea. Forecasting, 15, 700–714, doi:10.1175/1520-0434(2000)015<0700:TMEAAF>2.0.CO;2.
Kennedy, P. C., and Rutledge S. A. , 2011: S-band dual-polarization radar observations of winter storms. J. Appl. Meteor. Climatol., 50, 844–858, doi:10.1175/2010JAMC2558.1.
Kumjian, M. R., 2013: Principles and applications of dual-polarization weather radar. Part III: Artifacts J. Oper. Meteor.,1, 265–274, doi:10.15191/nwajom.2013.0121.
Kumjian, M. R., Picca J. C. , Ganson S. M. , Ryzhkov A. V. , and Zrnić D. S. , 2010: Three-body scattering signatures in polarimetric radar data. NOAA/NSSL Rep., 12 pp. [Available online at https://www.nssl.noaa.gov/publications/wsr88d_reports/FINAL_TBSS.doc.]
Kumjian, M. R., Reeves H. D. , and Ryzhkov A. V. , 2011: Polarimetric signatures in winter storms. NOAA/NSSL Rep., 27 pp. [Available online at http://www.nssl.noaa.gov/publications/wsr88d_reports/FINAL2011-Tsk2b-PolSigWinter.pdf.]
Kumjian, M. R., Ryzhkov A. V. , Reeves H. D. , and Schuur T. J. , 2013: A dual-polarization radar signature of hydrometeor refreezing in winter storms. J. Appl. Meteor. Climatol., 52, 2549–2566, doi:10.1175/JAMC-D-12-0311.1.
Kumjian, M. R., Rutledge S. A. , Rasmussen R. M. , Kennedy P. C. , and Dixon M. , 2014: High-resolution polarimetric radar observations of snow-generating cells. J. Appl. Meteor. Climatol., 53, 1636–1658, doi:10.1175/JAMC-D-13-0312.1.
NART, cited 2014: News from NOAA in the North Atlantic. North Atlantic Regional Team Newsletter. [Available online at http://www.regions.noaa.gov/north-atlantic/wp-content/uploads/2013/03/NOAA_NART_Spring_2013_newsletter_FINAL.pdf.]
NWA, cited 2014: Update on the February 8–9, 2013 nor’easter/southern New England blizzard. National Weather Association. [Available online at http://www.nwas.org/archive/2013/WinterStorm2013.]
Oraltay, R. G., and Hallett J. , 2005: The melting layer: A laboratory investigation of ice particle melt and evaporation near 0°C. J. Appl. Meteor., 44, 206–220, doi:10.1175/JAM2194.1.
Park, H. S., Ryzhkov A. V. , Zrnić D. S. , and Kim K.-E. , 2009: The hydrometeor classification algorithm for the polarimetric WSR-88D: Description and application to an MCS. Wea. Forecasting, 24, 730–748, doi:10.1175/2008WAF2222205.1.
Picca, J., and Ryzhkov A. , 2012: A dual-wavelength polarimetric analysis of the 16 May 2010 Oklahoma City extreme hailstorm. Mon. Wea. Rev., 140, 1385–1403, doi:10.1175/MWR-D-11-00112.1.
Picca, J., Schultz D. M. , Colle B. A. , Ganetis S. , Novak D. R. , and Sienkiewicz M. , 2014: The value of dual-polarization radar in diagnosing the complex microphysical evolution of an intense snowband. Bull. Amer. Meteor. Soc., doi:10.1175/BAMS-D-13-00258.1, in press.
Rauber, R. M., and Coauthors, 2014: Stability and charging characteristics of the comma head region of continental winter cyclones. J. Atmos. Sci., 71, 1559–1582, doi:10.1175/JAS-D-13-0253.1.
Ryzhkov, A. V., 2007: The impact of beam broadening on the quality of radar polarimetric data. J. Atmos. Oceanic Technol., 24, 729–744, doi:10.1175/JTECH2003.1.
Ryzhkov, A. V., and Zrnić D. S. , 1995: Precipitation and attenuation measurements at a 10-cm wavelength. J. Appl. Meteor., 34, 2121–2134, doi:10.1175/1520-0450(1995)034<2120:PAAMAA>2.0.CO;2.
Ryzhkov, A. V., and Zrnić D. S. , 1998: Discrimination between rain and snow with a polarimetric radar. J. Appl. Meteor., 37, 1228–1240, doi:10.1175/1520-0450(1998)037<1228:DBRASW>2.0.CO;2.
Ryzhkov, A. V., and Zrnić D. S. , 2007: Depolarization in ice crystals and its effect on radar polarimetric measurements. J. Atmos. Oceanic Technol., 24, 1256–1267, doi:10.1175/JTECH2034.1.
Ryzhkov, A. V., Zrnić D. S. , and Gordon B. A. , 1998: Polarimetric method for ice water content determination. J. Appl. Meteor., 37, 125–134, doi:10.1175/1520-0450(1998)037<0125:PMFIWC>2.0.CO;2.
Ryzhkov, A. V., Giangrande S. E. , and Schuur T. J. , 2005: Rainfall estimation with a polarimetric prototype of WSR-88D. J. Appl. Meteor., 44, 502–515, doi:10.1175/JAM2213.1.
Ryzhkov, A. V., Reeves H. , Schuur T. , Kumjian M. , and Zrnić D. , 2011: Investigations of polarimetric radar signatures in winter storms and their relation to aircraft icing and freezing rain. 35th Conf. on Radar Meteorology, Pittsburgh, PA, Amer. Meteor. Soc., P13.197. [Available online at https://ams.confex.com/ams/35Radar/webprogram/Paper191245.html.]
Ryzhkov, A. V., Kumjian M. , Ganson S. , and Zhang P. , 2013: Polarimetric radar characteristics of melting hail. Part II: Practical implications. J. Appl. Meteor. Climatol., 52, 2871–2886, doi:10.1175/JAMC-D-13-074.1.
Stewart, R. E., 1984: Deep isothermal layers within precipitation bands over southern Ontario. J. Geophys. Res., 89, 2567–2572, doi:10.1029/JD089iD02p02567.
Stewart, R. E., Marwitz J. D. , Pace J. C. , and Carbone R. E. , 1984: Characteristics through the melting layer of stratiform clouds. J. Atmos. Sci., 41, 3227–3237, doi:10.1175/1520-0469(1984)041<3227:CTTMLO>2.0.CO;2.
Straka, J. M., Zrnić D. S. , and Ryzhkov A. V. , 2000: Bulk hydrometeor classification and quantification using polarimetric radar data: Synthesis of relations. J. Appl. Meteor., 39, 1341–1372, doi:10.1175/1520-0450(2000)039<1341:BHCAQU>2.0.CO;2.
Szeto, K. K., Lin C. A. , and Stewart R. E. , 1988: Mesoscale circulations forced by melting snow. Part I: Basic simulations and dynamics. J. Atmos. Sci., 45, 1629–1641, doi:10.1175/1520-0469(1988)045<1629:MCFBMS>2.0.CO;2.
Wexler, R. R., Reed J. R. , and Honig J. , 1954: Atmospheric cooling by melting snow. Bull. Amer. Meteor. Soc., 35, 48–51.
Zrnić, D. S., 1987: Three-body scattering produces precipitation signature of special diagnostic value. Radio Sci., 22, 76–86, doi:10.1029/RS022i001p00076.
Zrnić, D. S., Zhang G. , Melnikov V. , and Andric J. , 2010: Three-body scattering and hail size. J. Appl. Meteor. Climatol., 49, 687–700, doi:10.1175/2009JAMC2300.1.
