1. Introduction
a. Motivation
An extratropical cyclone that affected the eastern United States on 24–25 January 2000 produced heavy snow from the Carolinas northward into the mid-Atlantic region and set the storm-total snowfall record at Raleigh-Durham, North Carolina (RDU), of 51.5 cm (20.3 in.; NCDC 2000a). An objective analysis1 of liquid-equivalent precipitation totals from the National Weather Service (NWS) cooperative observer network shows that 30–70+ mm of liquid-equivalent precipitation fell across the central and eastern Carolinas and southeast Virginia during the 2-day period from 24 to 26 January 2000 (Fig. 1a). Maximum liquid-equivalent precipitation amounts exceeded 75 mm (3 in.) across eastern South Carolina and southeastern North Carolina.
In contrast to observations, the 48-h precipitation forecast ending at 0000 UTC 26 January from the 0000 UTC 24 January run of the National Centers for Environmental Prediction (NCEP) Eta Model was significantly lighter for interior sections of the Carolinas and the mid-Atlantic region, with liquid-equivalent precipitation amounts greater than 25 mm confined to the immediate coastal area of North Carolina and offshore (Fig. 1b). For example, the Eta Model forecasted less than 5 mm of liquid-equivalent precipitation at RDU, where over 60 mm was observed. Previous studies of this case have attributed the poor operational model forecasts to high sensitivity to initial conditions (Zhang et al. 2002; Langland et al. 2002; Jang et al. 2003) and the limits of mesoscale predictability in the presence of moist processes (Zhang et al. 2003). Zupanski et al. (2002) showed strong sensitivity of Eta Model forecasts of this event to the data assimilation cycle between 0000 and 1200 UTC on 24 January as forecasted precipitation totals over the Carolinas and Virginia improved when the data assimilation scheme better depicted surface convergence and moisture over Georgia at 1200 UTC 24 January.
The goal of this research is to identify the specific physical processes and associated synoptic or mesoscale features that determined the westward extent of precipitation over the Carolinas and Virginia in this event. Specifically we will quantify the impact of an evolving diabatically generated potential vorticity (PV) maximum that was initially generated in an area of precipitation that formed between 0600 and 1200 UTC 24 January over the lower Mississippi River valley and southeast United States. We have termed this precipitation feature “incipient precipitation” (IP), since it developed prior to the rapid deepening of the surface cyclone offshore of the Carolinas.
The IP was poorly predicted in short-term forecasts from the 0000 UTC 24 January 2000 run of the Eta Model. (From this point forward, date and time will be referenced as DD/HH, e.g., 24/00 denotes 0000 UTC 24 January). The IP can be seen in radar imagery at 24/09 as a band of moderate to heavy precipitation over southeastern Alabama and western Georgia (Fig. 2). Observed precipitation totals for the 6-h period ending at 24/12 were between 10 and 33 mm over this region while the Eta Model forecasted the heavy precipitation to remain farther south, generating less than 5 mm over the region where the heaviest precipitation was observed.
A sequence of Rapid Update Cycle (RUC) model analyses of 900–700-hPa PV and radar imagery from 24 January shows the development and intensification of a PV maximum that moves eastward with the IP from 24/06 through 24/18 (Fig. 3). The PV maximum reaches a magnitude of more than 1.25 potential vorticity units (PVU, where 1 PVU = 1.0 × 10−6 m2 s−1 K kg−1) by 24/18 (Fig. 3d). The inability of the Eta Model to properly forecast the IP would likely preclude accurate prediction of any diabatically generated PV maximum produced by the IP. This is confirmed by comparing the Eta Model 24-h forecast of 900–700-hPa PV, 700-hPa winds, and 700-hPa horizontal moisture flux valid at 25/00 to the RUC model analysis at this time. The Eta Model forecasts a 1-PVU maximum well offshore of the southeast U.S. coast but weak 700-hPa winds over the Carolinas and moisture flux values less than 10 g kg−1 m s−1 (Fig. 4a). In contrast, the RUC analysis, which reflects the impact of the IP, shows the same offshore PV maximum plus a second, larger 1.5-PVU maximum along the Georgia coast (Fig. 4b). As would be expected poleward of a large cyclonic PV feature in this location, onshore (easterly) 700-hPa winds exceed 35 kt over the eastern Carolinas, producing moisture fluxes greater than 50 g kg−1 m s−1 in this region. This comparison indicates that the PV maximum along the Georgia coast was a factor in the moisture transport into the Carolinas and Virginia in this event (see section 1b).
The above evidence leads us to hypothesize that the lower-tropospheric PV maximum along the Georgia coast at 25/00 was initially generated by the IP and contributed strongly to the inland moisture transport in the Carolinas and Virginia. We will present a PV budget, perform piecewise PV inversion, and analyze output from mesoscale model experiments to test the above hypothesis.
b. Latent heating and extratropical cyclones
In an adiabatic atmosphere, cyclogenesis can be conceptualized through the PV framework as the mutual interaction of finite-amplitude disturbances at the tropopause and the surface (Hoskins et al. 1985). This view is consistent with the Eady model2 in an adiabatic atmosphere (Morgan and Nielsen-Gammon 1998) and with the Sutcliffe–Pettersen self-development concept (Sutcliffe and Forsdyke 1950; Pettersen 1956) where low-level thermal advection contributes to cyclogenesis by inducing sea level pressure falls and amplifying the upper-level wave and creating a positive feedback as the surface cyclone strengthens. The PV counterpart to this view is succinctly summarized by Hoskins (1990, section 5.4).
The role of diabatic processes, such as latent heat release (LHR) in extratropical cyclogenesis has been well documented (e.g., Tracton 1973; Bosart 1981; Gyakum 1983a, b; Bosart and Lin 1984; Uccellini et al. 1987; Reed et al. 1988; Reed and Kuo 1988; Manobianco 1989; Davis and Emanuel 1991; Davis 1992; Davis et al. 1993; Stoelinga 1996). LHR can further enhance the cyclogenesis process in two ways that are readily evident in the PV framework. First, LHR reduces the effective static stability (Durran and Klemp 1982), which enhances the vertical penetration of the circulation associated with the upper- and lower-boundary PV anomalies, increasing the mutual amplification of these waves (Hoskins et al. 1985, section 6e). Second, a maximum of diabatic heating produces (destroys) PV below (above) the level of maximum heating along the absolute vorticity vector. This generally leads to the generation (destruction) of PV in the lower (upper) troposphere if the heating maximum is located in the midtroposphere (Raymond 1992). Owing to the fact that PV maxima are associated with negative geopotential height perturbations, diabatically generated PV maxima in the lower troposphere are thus directly linked to the location and intensity of surface cyclones (e.g., Davis 1992; Stoelinga 1996). This association between surface cyclones and lower-tropospheric PV maxima evident within the PV framework is consistent with traditional quasigeostrophic diagnosis via the height-tendency equation, and can physically be interpreted as hydrostatic pressure reduction below a local maximum in heating. The vertical gradient of the heating and the magnitude of the absolute vorticity determine the strength of the resultant diabatic PV anomaly (Stoelinga 1996). Reed et al. (1992, their Fig. 16) present a conceptual model of PV features in extratropical cyclones, including this lower-tropospheric diabatic PV maximum.
The principle of PV invertibility discussed by Hoskins et al. (1985) allows one to recover balanced atmospheric fields by inverting a portion of the PV field, while the conservation principle allows the impact of nonconservative processes, such as diabatic heating, to be quantified through the computation of nonadvective PV tendencies. Several studies have investigated the impact of diabatically produced positive PV anomalies in the lower troposphere on cyclones, with the findings showing large case-to-case variability. Davis and Emanuel (1991) found that the development of a positive diabatically generated PV anomaly in the lower troposphere added directly to the circulation of the surface cyclone and helped amplify the upper-level wave in a system that developed over eastern North America. They also showed that the diabatically produced PV anomaly eventually was responsible for 40% of the cyclone’s circulation. Reed et al. (1992) found that an anomalously large positive lower-tropospheric PV anomaly developed in a western Atlantic cyclone as boundary layer air ascended the warm frontal surface and underwent differential condensational heating. A model experiment in which latent heat was turned off showed that the surface cyclone was 19 hPa weaker after 24 h relative to a full-physics model simulation. Examining a cyclone over the central United States, Davis (1992) showed reduced importance of the lower-tropospheric diabatically produced PV maximum to the overall cyclone development; this feature contributed only 20% of the vorticity at the 850-hPa level in that system. In the case studied by Davis (1992), the low-level diabatic PV anomaly actually prevented phase locking of the upper and lower waves by contributing to the rapid eastward progression of the cyclonic surface potential temperature (θ) anomaly. Davis et al. (1993) studied two cases of marine cyclogenesis and one continental cyclone and found that latent heating enhanced each cyclone’s intensity by adding a positive PV anomaly onto the structure of the surface wave, a process that Davis and Emanuel (1991) describe as “superposition.” Stoelinga (1996) found that a low-level diabatically produced PV maximum contributed 70% of the mature cyclone’s intensity, with a much weaker cyclone developing in a model simulation withholding the effects of latent heat. The mutual amplification that occurred in that case involved the upper-level PV anomaly and the diabatically produced low-level PV anomaly, as the surface θ anomaly had little impact on the cyclone intensity.
Diabatically produced cyclonic PV anomalies in the lower troposphere have also been shown to contribute to moisture transport in cyclones and along low-level fronts. Whitaker et al. (1988), in a simulation of the 1979 Presidents’ Day storm, showed that a low-level diabatically produced PV anomaly contributed substantially to the moisture transport in that event. Lackmann and Gyakum (1999) found that a PV anomaly along a cold front was coupled with a low-level jet (LLJ) in a flooding event in the Pacific Northwest. Lackmann (2002) showed that a diabatically produced PV anomaly along a cold front contributed 15%–40% of the strength of an LLJ that developed along the edge of the warm sector of a cyclone in the central United States.
It is evident from previous research that diabatically generated low-level cyclonic PV maxima can be important in not only the development of extratropical cyclones, but also in the transport of moisture into cyclones, since these PV maxima are usually located in a region of the atmosphere with high moisture content. In the remainder of this paper, we use PV diagnostics to test the hypothesis that the PV anomaly generated by the IP was important to the moisture transport into the Carolinas and Virginia during the 24–25 January 2000 cyclone event. The methodology used in this study is outlined in section 2. A synoptic overview of the cyclone is presented in section 3. Section 4 shows the results of a mesoscale model simulation and PV diagnostics. Section 5 contains conclusions and suggestions for future work.
2. Methodology
a. Precipitation analysis
Some previous studies of this case (e.g., Zhang et al. 2002; Zupanski et al. 2002) appear to have used only Automated Surface Observation System (ASOS) gauge liquid precipitation observations as the verifying analysis for their model simulations. These analyses (Fig. 1 in Zhang et al. 2002; Fig. 3 in Zupanski et al. 2002) show only 10–20 mm of precipitation over central North Carolina. However, in instances of freezing and/or frozen precipitation, the ASOS tipping-bucket gauge will greatly underestimate the amount of liquid-equivalent precipitation, perhaps by as much as 40%–50% (NWS 2005). In this event, the ASOS at RDU reported 20.1 mm of liquid-equivalent precipitation for this event, while a more accurate estimate of the actual liquid-equivalent total was 60.2 mm (NCDC 2000b), indicating an ASOS underestimation of nearly two-thirds. As a more accurate representation of the true liquid-equivalent precipitation totals from this event, we used liquid-equivalent precipitation totals from 24–26 January 2000 reported by NWS cooperative observers, against which we will evaluate the performance of a mesoscale model simulation described below. Although this dataset is not ideal because the observations are taken at various times and are available only once daily, precluding the comparison of 6- or 12-h precipitation totals with model output, they nevertheless represent a major improvement over the use of ASOS tipping-bucket data in the regions that received heavy snowfall. All observations from Virginia, North Carolina, South Carolina, and Georgia were obtained and interpolated to a 77 × 53 grid with 0.21° latitude × 0.21° longitude grid spacing using a Barnes objective analysis (Koch et al. 1983). Any clearly erroneous observations were removed from the dataset (e.g., an observation of zero precipitation in an area that received substantial snowfall) before the objective analysis was performed; this analysis is shown in Fig. 1a.
b. Mesoscale model
To provide a dynamically consistent dataset for case analysis and to facilitate computations for PV diagnostics, this event was simulated using version 3.5 of the fifth-generation Pennsylvania State University–National Center for Atmospheric Research (NCAR) Mesoscale Model (MM5). MM5 is a limited-area, nonhydrostatic, sigma-coordinate model. An overview of the model and the prognostic equations can be found in Dudhia (1993) and Grell et al. (1994).
A satisfactory simulation of the event was achieved by initializing MM5 at 24/09 with initial conditions provided by the National Oceanic and Atmospheric Administration (NOAA) Forecast Systems Laboratory RUC model analysis (Benjamin et al. 1998). The model was run for 39 h through 26/00, and 3-hourly RUC analyses were used throughout the simulation to provide lateral boundary conditions. The model was run with 37 vertical sigma levels, and the horizontal domain had a grid spacing of 36 km, covering much of the eastern United States and adjacent coastal waters (Fig. 5). This initial time was chosen because by 24/09 the IP had already formed over Alabama and Georgia, allowing for the initial conditions to better capture the impact of the IP beyond this time and provide a reasonable simulation dataset for case analysis.
Physical parameterizations chosen for this simulation were the Reisner mixed-phase explicit moisture scheme (Reisner et al. 1998), the Grell cumulus parameterization (CP) scheme (Grell et al. 1994), the Blackadar high-resolution planetary boundary layer scheme (Zhang and Anthes 1982), and the Dudhia cloud radiation scheme (Dudhia 1989). Climatological snow-cover data were used. The lower boundary condition over water grid points in the simulation was provided by a sea surface temperature (SST) analysis with 14-km grid spacing generated every 48 h over North America and adjacent waters. In this analysis the SST is defined as the sea surface temperature tuned to in situ data at 1-m depth. (More information on this dataset is available online at http://www.saa.noaa.gov/nsaa/products/welcome.)
c. PV budget
d. Piecewise PV inversion
To quantify the impact of the diabatic PV maximum produced by the IP on the moisture transport and the cyclone evolution, nonlinear piecewise PV inversion was performed based on the methodology outlined in Davis and Emanuel (1991). Given an independent balance relationship between the wind and temperature fields, the balanced flow and geopotential height field can be recovered for a given distribution of PV as a boundary value problem where PV is defined [after Rossby (1940) and Ertel (1942)] as in (2). Here the Charney (1955) balance condition, which is similar to gradient wind balance and very accurate in high-curvature flows, is used following Davis and Emanuel (1991).
We solve for balanced fields of geopotential (Φ) and streamfunction (ψ) using perturbation values of PV on the model interior as well as upper- and lower-boundary θ. To compute mean-state values of PV, θ, Φ, and ψ for the inversion, pressure-level output was averaged daily from a 14-day MM5 simulation run from 17 to 31 January 2000. Initial and boundary conditions for this background simulation were supplied from the 2.5° NCEP–National Center for Atmospheric research (NCAR) reanalysis (Kalnay et al. 1996). Other parameters for the simulation were identical to those outlined in section 2b. Perturbation values of PV, θ, Φ, and ψ were obtained by subtracting the 14-day mean state from the instantaneous values from the original MM5 simulation at the time of inversion. The average 875–825-hPa PV field from the 14-day simulation shows a broad PV maximum of 0.6–0.8 PVU in the Ohio Valley, with smaller areas of 0.6–0.8 PVU along the Atlantic Coast (Fig. 6).
The remaining perturbation PV and boundary θ were also partitioned and inverted. The sum of the resultant height field from the four perturbation pieces equaled that found when all perturbation PV and boundary θ was inverted simultaneously (not shown). Once the balanced piecewise velocity field has been obtained, the piecewise horizontal moisture flux is computed by multiplying this vector field by the water vapor mixing ratio, similar to the method employed by Lackmann et al. (1998).
3. Synoptic overview
Geopotential heights, isotachs, winds, and divergence at the 250-hPa level from the RUC model analysis are presented in Fig. 7. At the time of this case, the RUC model had a horizontal grid spacing of 40 km and 40 vertical levels (Benjamin et al. 1998). The RUC data presented here were interpolated onto the “211” grid with an approximate grid spacing of 80 km and vertical levels every 50 hPa. At 24/00 a broad trough covers the central United States, with a 70 m s−1 jet at the 250-hPa level over New Mexico and Texas on the west side of the upper trough, and a lifting 75 m s−1 jet over the mid-Atlantic and New England (Fig. 7a). A broad divergence maximum is evident over eastern Texas and Louisiana with a stronger maximum over Georgia. By 24/12 the upper trough has deepened and moved into the Tennessee Valley as the western jet digs into the Gulf of Mexico, and divergence has increased over northern Florida and southern Georgia (Fig. 7b). Downstream ridging appears over North Carolina and Virginia by 25/00 (Fig. 7c) and spreads into New England by 25/12 (Fig. 7d) as the trough takes on a negative meridional tilt and the divergence increases downstream of the trough axis.
At the 500-hPa level a broad, positively tilted trough extends from the western Great Lakes to the southern plains at 24/00 (Fig. 8a). By 24/12, the vorticity within the southern portion of the trough has amplified considerably and moved eastward to the Mississippi–Alabama border (Fig. 8b). Downstream ridging has amplified in eastern Virginia by 25/00, as the trough has taken on a negative meridional tilt and closed off over eastern Georgia and the absolute vorticity continues to increase in the trough base (Fig. 8c). By 25/12, the 500-hPa cyclone has continued to deepen as it moves into eastern North Carolina (Fig. 8d).
At the surface, manual analysis at 24/00 shows a front stretching east–west along the Gulf Coast, connecting to a cold front trailing a departing cyclone off the mid-Atlantic coast (Fig. 9a). High pressure is centered over Kansas and is ridging eastward into the Mississippi Valley. By 24/12, a 1005-hPa surface low has developed along the Georgia–Florida border (Fig. 9b). A cold front trails the cyclone through the eastern Gulf of Mexico, and coastal frontogenesis has developed northeast of the low center along the coast from Georgia to the Carolinas, with a temperature difference approaching 15°C across the front. A second baroclinic zone is seen well offshore, with southerly winds east and south of the boundary, shifting to west or northwest across the front. By 25/00, the cyclone has deepened to 991 hPa (a central pressure decrease of 14 hPa in 12 h) and moved northeast to a position east-northeast of Charleston, South Carolina (Fig. 9c). Two cold fronts are seen trailing the cyclone, one farther east that marks a wind shift from southerly to westerly, and a second cold front in the offshore waters from the Carolinas to southern Florida. This second front marks the leading edge of cold, continental air surging southward behind the cyclone as the wind shifts from westerly to northwesterly across the front. A well-defined warm/coastal front still extends north-northeast of the low center parallel to the North Carolina coast and extending eastward offshore. By 25/12, the cyclone has moved to a position east of Norfolk, Virginia, as it continues to deepen to 983 hPa (Fig. 9d). The warm/coastal front remains prominent northeast of the low center, with a sharp wind shift from southeasterly to easterly across the front. Cold air continues to move offshore behind the trailing cold front as it pushes well east of the coastline.
From a satellite perspective, this cyclogenesis event appears to be of the instant occlusion type as described by Evans et al. (1994). In this conceptual model, cyclogenesis occurs in association with a distinct cold-air cloud cluster downstream of an upper-trough axis, within a jet entrance region on the cold side of a deep-tropospheric baroclinic zone. In the present case, the cloud mass associated with the IP represents the cold-air cloud cluster that merges with a baroclinic zone offshore of the East Coast. The initial stage in the instant occlusion process is seen in the infrared satellite imagery overlaid with 300-hPa RUC-analyzed isotachs at 24/12 (Fig. 10a). The coldest cloud tops in the cold-air cluster are located in the entrance region of the upper jet, and the surface low is located on the eastern edge of the cloud shield along of the Florida–Georgia border. The deep-tropospheric baroclinic zone is clearly separated from the cold-air cluster at this time and stretches along the East Coast of the United States from Florida to Long Island, New York. The southern portion of this baroclinic zone is evident in the surface data shown in Fig. 9b. By 25/00, the cloud mass associated with the cold-air cluster over North Carolina has linked with the northern portion of the offshore baroclinic zone and the surface low is found near the dry slot offshore of the Carolinas (Fig. 10b).
The evolution of the precipitation shield is shown in Fig. 11. At 24/00, radar imagery indicates a large area of precipitation from northern Florida extending northeastward along the southeast U.S. coast (Fig. 11a). The development of the IP is seen by 24/12, with a large area of moderate to heavy precipitation over much of Georgia and extending into western South Carolina (Fig. 11b). The evolving precipitation shield moves northeast and covers much of North and South Carolina by 25/00 in the comma head region of the developing cyclone (Fig. 11c). By 25/12, the precipitation has expanded northward into southern New England, while still continuing as far south as the eastern Carolinas (Fig. 11d).
The initial development of the IP occurs around 24/06 along a zone of frontogenesis at the 900-hPa level, as depicted in the RUC analysis (Fig. 12a). A cross section from northwest to southeast across the front shows upglide occurring in an area with mixing ratios greater than 8 g kg−1 below 850 hPa (Fig. 12b). This region is also experiencing synoptic-scale ascent ahead of the upper trough, evident in Q-vector convergence in the 850–500-hPa layer over Mississippi and Alabama (Fig. 12c). The vertical profile of equivalent potential temperature (θe) in this region shows a large area of weak potential instability in the midtroposphere, where θe is constant or decreasing with height, and values of moist geostrophic PV (MPVg = gηg · ∇θe) are negative (Fig. 12d). Negative values of MPVg indicate the presence of elevated potential instability, which could be released if sufficient lift occurs to saturate the layer (Schultz and Schumacher 1999). This raises the possibility that elevated upright and/or slantwise convection could be occurring in this region, as ample moisture and forcing for ascent are also present.
4. Results
a. MM5 synoptic evaluation
To use the MM5 simulation as a surrogate dataset for PV diagnostics, it is important to show that MM5 produced a reasonably accurate simulation of the cyclone event. The evolution of the MM5 cyclone and simulated radar reflectivity from the grid-scale precipitation (Fig. 13) are quite similar to that seen in the surface analysis (Fig. 9) and radar imagery (Fig. 11). At 24/12, the surface low is located over northern Florida and a large precipitation shield covers much of Georgia (Fig. 13a). By 25/00, the surface low is intensifying along the strong baroclinic zone offshore of the Carolinas as model grid-scale precipitation has spread into the Carolinas and southeastern Virginia (Fig. 13b). The surface low has moved northeast of Cape Hatteras, North Carolina, at 25/12 with a central pressure of 984 hPa as precipitation has spread into the mid-Atlantic and New England (Fig. 13c). The total precipitation produced by the simulation shows a broad maximum of 25–50 mm from eastern South Carolina to southeastern Virginia, with a sharp gradient to the west (Fig. 13d). The pattern and maximum amounts are similar to those seen in observations (Fig. 1a), although heavy precipitation was not produced far enough west, as amounts tapered off rapidly from central Virginia to western South Carolina in the simulation (Fig. 13d). These results, which show increased precipitation over the Carolinas and Virginia as well as a deeper surface cyclone tracking closer to the coast when compared to the operational Eta Model, are consistent with the hypothesis that the PV maximum generated by the IP was important to this event. Results from additional simulations with 12- and 4-km grid spacing show little difference in the overall evolution of the IP formation or total precipitation over the Carolinas and Virginia (not shown). This indicates that model grid spacing was not a significant factor in influencing the precipitation forecast if the model was initialized after the formation of the IP.
b. PV budget
The evolution of the latent heating and lower-tropospheric PV maximum is shown in the results of a PV budget. At 24/12, latent heating at 700 hPa from the grid scale and convective precipitation in the MM5 simulation is maximized over central and southwest Georgia, with values exceeding 180 × 10−5 K s−1 associated with the IP (Fig. 14a). A PV maximum in the 900–700-hPa layer of 1.25 PVU is collocated with the heating maximum over southwestern Georgia. At 24/15, the PV maximum has strengthened to 1.5 PVU and moved eastward with the latent heating in a band from eastern Georgia to northern Florida (Fig. 14b). By 24/18, the 700-hPa LHR maximum has moved north into eastern South Carolina, with values of 180 × 10−5 K s−1 as the PV maximum has become more consolidated in eastern Georgia (Fig. 14c). At 25/00 the strongest latent heating has shifted northeastward into eastern North Carolina with the coastal cyclone’s developing precipitation shield. The PV maximum seen earlier has moved to the Georgia coast while additional maxima have formed offshore of the Carolinas (Fig. 14d).
A cross section at 24/15 from 32.5°N, 86°W to 30.5°N, 79°W shows a PV maximum centered near 925 hPa with a magnitude greater than 2 PVU (Fig. 15). The PV maximum is located upstream from the level of maximum heating along the absolute vorticity vector (Fig. 15a). Nonadvective PV tendencies are strongly positive near the PV maximum as indicated by the convergence of nonadvective PV flux vectors in that area (Fig. 15b).
Over the period from 24/12 to 25/00, a lower-tropospheric PV maximum grew in size and magnitude over the southeastern United States, coincident with a 700-hPa LHR maximum that initially formed in the MM5 simulation in the region where the IP was seen in observations. Latent heating was maximized in the middle and lower troposphere with a significant increase in lower-tropospheric PV occurring upstream along the absolute vorticity vector from the heating maximum, consistent with the findings of Raymond (1992).
c. Piecewise PV inversion
To quantify the impact of the lower-tropospheric cyclonic PV anomaly produced by the IP, all cyclonic perturbation PV in the layer extending from 950 to 600 hPa was inverted. At 24/15, as the cyclone is beginning to organize over northern Florida, an 875–825-hPa cyclonic PV anomaly is located over southeastern Georgia with a nondimensionalized magnitude of 0.3–0.4 (Fig. 16a). The balanced flow at 850 hPa from the piecewise inversion is contributing 20–25 kt of southeasterly flow over the Carolinas at this time. The resultant 850-hPa height field from the inversion shows a –100 m minimum over eastern Georgia over the PV anomaly. By 24/18, the PV anomaly has moved northeastward to eastern Georgia and southern South Carolina, with the associated height minimum exceeding –120 m (Fig. 16b). The southeasterly balanced flow has increased to 25–30 kt at 850 hPa over the eastern Carolinas. At 24/21 the elongated PV anomaly over land has moved little, while an offshore anomaly of 0.7–0.8 associated with the model surface cyclone has strengthened and is associated with a –160 m 850-hPa height minimum (Fig. 16c). The southeasterly 850-hPa balanced flow continues to increase over North Carolina and Virginia, reaching a magnitude of 30 kt near the coast and 15–20 kt well inland. The 850-hPa height minimum and PV anomaly move slightly north by 25/00, as the balanced flow at the 850-hPa level has become more easterly over North Carolina, while remaining southeasterly over Virginia (Fig. 16d). It is clear that a long duration of onshore southeasterly flow in the lower troposphere was associated with the lower-tropospheric cyclonic PV anomaly through the period from 24/15 to 25/00 as the cyclone moved northeastward offshore and intensified. The negative height perturbation and onshore component of the balanced flow associated with the diabatic PV maximum are also consistent with the surface cyclone tracking close to the coast in this simulation.
The impact of the onshore flow associated with the lower-tropospheric PV anomaly on the moisture flux into the Carolinas and Virginia will be examined next. As in Lackmann and Gyakum (1999), the total (piecewise) horizontal moisture flux was computed by multiplying the mixing ratio by the total wind (balanced flow from PV inversion) at a given pressure level. The 750-hPa piecewise moisture flux at 24/21 shows moisture flux vectors directed onshore from the southeast in the cyclonic flow north of the PV anomaly (Fig. 17a). There is strong convergence of the moisture flux vectors over central North Carolina at this time, consistent with heavy precipitation in this region. A cross section from 37°N, 81.5°W to 32°N, 76°W shows the piecewise moisture flux extending well inland through the 950–600-hPa layer (Fig. 17b). The 750-hPa moisture flux by the total wind at 24/21 shows an onshore component to the flux vectors along the coast of the Carolinas with a more northeastward component to the flux over inland areas (Fig. 18). In portions of the Carolinas, the magnitude of the piecewise moisture flux at the 750-hPa level is greater than or equal to the total moisture flux, suggesting that the flow associated with the lower-tropospheric PV maximum was the major contributor to the onshore moisture flux into the region. Farther north and west the piecewise moisture flux is greater than the total moisture flux because the balanced flow from other portions of the PV distribution, especially the background PV field, is acting to cancel the southeasterly flow from the cyclonic anomaly (not shown).
In addition to providing the moisture necessary for heavy precipitation, the PV anomaly also provided forcing for ascent in the lower troposphere through warm advection associated with the onshore flow. The 850-hPa piecewise Q vectors computed from the balanced flow at 24/21 show a large area of convergence over the coastal Carolinas, as the balanced geostrophic flow leads to warm advection in this area (Fig. 19). This is consistent with Q-vector convergence and forcing for quasigeostrophic ascent over the Carolinas and Virginia.
5. Conclusions
Prior to the rapid cyclogenesis that occurred off the southeast coast of the United States late on 24 January 2000, a large area of IP developed over Alabama and Georgia. The IP was not well forecasted by the operational Eta Model run initialized at 24/00, and this model run also failed to properly forecast the inland extent of heavy precipitation associated with the cyclone over the Carolinas and Virginia. We hypothesized that the lower-tropospheric PV maximum initially generated by the IP increased the onshore component of the flow in the lower troposphere, which strengthened the inland moisture transport into the Carolinas and Virginia. The increased horizontal moisture flux and resulting increase in moisture flux convergence is consistent with the inland precipitation maximum over the region in question. When MM5 was initialized at 24/09, after the initial development of the IP, the model was able to generate a realistic precipitation shield over Georgia in close agreement with radar imagery, as well as a lower-tropospheric PV maximum. Subsequently, the MM5 simulation produced heavy precipitation in the Carolinas and Virginia, consistent with the above hypothesis.
Results from a PV budget showed that the initial development of the lower-tropospheric PV maximum was directly related to strong latent heating in the midtroposphere associated with the IP. The PV maximum formed in the lower troposphere upstream along the absolute vorticity vector from the level of maximum latent heating. Piecewise PV inversion was performed on all anomalous cyclonic PV from 950 to 600 hPa to quantify the impact of this diabatic PV anomaly on the moisture transport into the Carolinas and Virginia. Results from the inversion showed a strong onshore component to the balanced flow in the midtroposphere, which produced a strong onshore moisture flux in the Carolinas that extended well inland in the lower to midtroposphere. The region of heavy precipitation in the simulation was coincident with strong convergence of the piecewise moisture flux vectors. The total moisture flux at the same time was similar in magnitude and direction, particularly over the eastern Carolinas, suggesting that the cyclonic PV anomaly associated with the IP was the dominant factor behind the moisture transport into the region. Additionally, the balanced flow from the PV inversion produced Q-vector convergence in the lower troposphere, providing forcing for ascent in this same area.
These results strongly suggest that the evolving lower-tropospheric PV maximum initially generated by the IP was a critical factor in driving the heavy precipitation that occurred over the Carolinas and Virginia later in the event by increasing the magnitude of the lower-tropospheric moisture flux, and resulting in a surface cyclone that was deeper and tracked farther west than forecasted by the operational models. It thus appears that the inability of the operational Eta Model and other models to properly predict the development of the IP was a major factor in their poor precipitation forecast during this event. These findings are consistent with those of Zupanski et al. (2002), who documented large sensitivity in the Eta Model to the analysis of moisture and surface convergence over Georgia at 24/12, in the region of the IP. It is likely that any numerical model would need to accurately forecast the IP and generate the PV anomaly in order to properly forecast the westward extent of the heavy precipitation over the Carolinas and Virginia, as well as the evolution of the cyclone itself. Future work is also underway to investigate the source of the forcing mechanism(s), instability, and moisture for the IP and why the Eta Model was unable to properly forecast it, even though it occurred in a region of relatively high data density.
Also, these results suggest that other examples of instant-occlusion-type cyclogenesis would be associated with a diabatic PV maximum produced by precipitation occurring in the cold-air cluster upstream of the main baroclinic zone. Further study is needed to determine the role that this type of PV anomaly might generally play in the instant occlusion process. In the PV framework, the relative strength of the cold-air PV maximum may play a role in linking the cold-air cluster to the downstream baroclinic zone. Further research is also necessary to determine how often this type of lower-tropospheric PV anomaly governs the westward extent of the precipitation shield in East Coast cyclones. However, results from this case suggest that if operational forecasters observe a large area of precipitation not forecasted by the operational models over the southeastern United States, they should be aware of the diabatic feedbacks of this precipitation on cyclone evolution and moisture transport. Along the East Coast of the United States, this feedback of the latent heating could enhance the westward transport of moisture, which may, in some cases, produce heavier precipitation farther inland than model forecasts would indicate.
Acknowledgments
This research was supported by NOAA Collaborative Science, Technology, and Applied Research (CSTAR) Grants NA-07WA0206 and NA03NWS4680007 and NSF Grant ATM-0079425, all awarded to North Carolina State University. Thanks to Dr. Chris Davis of NCAR for providing the piecewise PV inversion code. Thanks also to Dr. Jimy Dudhia of NCAR and Dr. Jack Kain of NSSL for help with outputting the temperature tendency terms in MM5. The MM5 was made available through NCAR, which is sponsored by the NSF. Prof. Jim Steenburgh and Mr. Daryl Onton of the University of Utah provided the gridder and GEMPAK conversion software used with MM5. The State Climate Office of North Carolina at North Carolina State University provided the cooperative observer liquid-equivalent precipitation reports. NCEP provided much of the meteorological data used in this study, which were delivered to North Carolina State University through the Unidata program. Thanks to Ms. Kelly Mahoney of NCSU who provided helpful comments on previous versions of this manuscript. Finally, thanks to the three anonymous reviewers who provided comments and suggestions to improve the manuscript.
REFERENCES
Benjamin, S G., J M. Brown, K J. Brundage, B E. Schwartz, T G. Smirnova, T L. Smith, and L. Morone, 1998: RUC-2—The Rapid Update Cycle Version 2. NWS Technical Procedure Bulletin 448, NOAA/NWS, 18 pp. [Available online at http://205.156.54.206/om/tpb/448.pdf.].
Bosart, L F., 1981: The Presidents’ Day snowstorm of 18–19 February 1979: A sub-synoptic scale event. Mon. Wea. Rev., 109 , 1542–1566.
Bosart, L F., and S C. Lin, 1984: A diagnostic analysis of the Presidents’ Day storm of February 1979. Mon. Wea. Rev., 112 , 2148–2177.
Cammas, J-P., D. Keyser, G M. Lackmann, and J. Molinari, 1994: Diabatic redistribution of potential vorticity accompanying the development of an outflow jet within a strong extratropical cyclone. Proc. Int. Symp. on the Life Cycles of Extratropical Cyclones, Vol. II, Bergen, Norway, Geophysical Institute, University of Bergen, 403–409.
Charney, J G., 1955: The use of primitive equations of motion in numerical prediction. Tellus, 7 , 22–26.
Davis, C A., 1992: A potential-vorticity diagnosis of the importance of initial structure and condensational heating in observed extratropical cyclogenesis. Mon. Wea. Rev., 120 , 2409–2428.
Davis, C A., and K A. Emanuel, 1991: Potential vorticity diagnostics of cyclogenesis. Mon. Wea. Rev., 119 , 1929–1953.
Davis, C A., M T. Stoelinga, and Y-H. Kuo, 1993: The integrated effect of condensation in numerical simulations of extratropical cyclogenesis. Mon. Wea. Rev., 121 , 2309–2330.
Dudhia, J., 1989: Numerical study of convection observed during the Winter Monsoon Experiment using a mesoscale two-dimensional model. J. Atmos. Sci., 46 , 3077–3107.
Dudhia, J., 1993: A nonhydrostatic version of the Penn State–NCAR mesoscale model: Validation tests and simulation of an Atlantic cyclone and cold front. Mon. Wea. Rev., 121 , 1493–1513.
Durran, D R., and J B. Klemp, 1982: On the effects of moisture on the Brunt–Väisälä frequency. J. Atmos. Sci., 39 , 2152–2158.
Ertel, H., 1942: Ein Neuer hydrodynamischer Wirbelsatz. Meteor. Z., 59 , 271–281.
Evans, M S., D. Keyser, L F. Bosart, and G M. Lackmann, 1994: A satellite-derived classification scheme for rapid maritime cyclogenesis. Mon. Wea. Rev., 122 , 1381–1416.
Grell, G A., J. Dudhia, and D S. Stauffer, 1994: A description of the fifth-generation Penn State/NCAR mesoscale model (MM5). NCAR Tech. Note NCAR/TN-398+STR, 122 pp.
Gyakum, J G., 1983a: On the evolution of the QE II storm. I: Synoptic aspects. Mon. Wea. Rev., 111 , 1137–1155.
Gyakum, J G., 1983b: On the evolution of the QE II storm. II: Dynamic and thermodynamic structure. Mon. Wea. Rev., 111 , 1156–1173.
Hoskins, B J., 1990: Theory of extratropical cyclones. Extratropical Cyclones: The Erik Palmén Memorial Volume, C. W. Newton and E. O. Holopainen, Eds., Amer. Meteor. Soc., 63–80.
Hoskins, B J., M E. McIntyre, and A W. Robertson, 1985: On the use and significance of isentropic potential vorticity maps. Quart. J. Roy. Meteor. Soc., 111 , 877–946.
Jang, K-I., X. Zou, M. S. F. V. De Pondeca, M. Shapiro, C. Davis, and A. Krueger, 2003: Incorporating TOMS ozone measurements into the prediction of the Washington, D.C., winter storm during 24–25 January 2000. J. Appl. Meteor., 42 , 797–812.
Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77 , 437–471.
Koch, S E., M. desJardins, and P. Kocin, 1983: An interactive Barnes objective map analysis scheme for use with satellite and conventional data. J. Climate Appl. Meteor., 22 , 1487–1503.
Lackmann, G M., 2002: Cold-frontal potential vorticity maxima, the low-level jet, and moisture transport in extratropical cyclones. Mon. Wea. Rev., 130 , 59–74.
Lackmann, G M., and J R. Gyakum, 1999: Heavy cold-season precipitation in the northwestern United States: Synoptic climatology and an analysis of the flood of 17–18 January 1986. Wea. Forecasting, 14 , 687–700.
Lackmann, G M., J R. Gyakum, and R. Benoit, 1998: Moisture transport diagnosis of a wintertime precipitation event in the Mackenzie River Basin. Mon. Wea. Rev., 126 , 668–691.
Langland, R H., M A. Shapiro, and R. Gelaro, 2002: Initial condition sensitivity and error growth in forecasts of the 25 January 2000 East Coast snowstorm. Mon. Wea. Rev., 130 , 957–974.
Manobianco, J., 1989: Explosive east coast cyclogenesis over the west-central North Atlantic Ocean: A composite study derived from ECMWF operational analyses. Mon. Wea. Rev., 117 , 2365–2383.
Morgan, M C., and J W. Nielsen-Gammon, 1998: Using tropopause maps to diagnose midlatitude weather systems. Mon. Wea. Rev., 126 , 2555–2579.
NCDC, 2000a: Storm Data. Vol. 42, No. 1, 171 pp.
NCDC, 2000b: Local Climatological Data. January 2000, Raleigh, NC, 8 pp.
NWS, cited. 2005: The tipping bucket rain gauge. [Available online at http://www.nws.noaa.gov/asos/tipbuck.html.].
Pettersen, S., 1956: Weather Analysis and Forecasting. 2d ed. McGraw-Hill, 428 pp.
Raymond, D J., 1992: Nonlinear balance and potential-vorticity thinking at large Rossby number. Quart. J. Roy. Meteor. Soc., 118 , 987–1015.
Reed, R J., and Y-H. Kuo, 1988: Numerical simulations of an explosively deepening cyclone in the eastern Pacific. Mon. Wea. Rev., 116 , 2081–2105.
Reed, R J., A J. Simmons, M D. Albright, and P. Undén, 1988: The role of latent heat release in explosive cyclogenesis: Three examples based on ECMWF operational forecasts. Wea. Forecasting, 3 , 217–229.
Reed, R J., M T. Stoelinga, and Y-H. Kuo, 1992: A model-based study of the origin and evolution of the anomalously high potential vorticity in the inner region of a rapidly deepening marine cyclone. Mon. Wea. Rev., 120 , 893–913.
Reisner, J., R M. Rasmussen, and R T. Bruintjes, 1998: Explicit forecasting of supercooled liquid water in winter storms using the MM5 mesoscale model. Quart. J. Roy. Meteor. Soc., 124 , 1071–1107.
Rossby, C-G., 1940: Planetary flow patterns in the atmosphere. Quart. J. Roy. Meteor. Soc., 66 , (Suppl.),. 68–87.
Schultz, D M., and P N. Schumacher, 1999: The use and misuse of conditional symmetric instability. Mon. Wea. Rev., 127 , 2709–2732. Corrigendum. 128 , 1573.
Stoelinga, M T., 1996: A potential-vorticity based study of the role of diabatic heating and friction in a numerically simulated baroclinic cyclone. Mon. Wea. Rev., 124 , 849–874.
Sutcliffe, R C., and A G. Forsdyke, 1950: The theory and use of upper air thickness patterns in forecasting. Quart. J. Roy. Meteor. Soc., 76 , 189–217.
Tracton, M S., 1973: The role of cumulus convection in the development of extratropical cyclones. Mon. Wea. Rev., 101 , 573–592.
Uccellini, L W., R A. Petersen, K F. Brill, P J. Kocin, and J J. Tuccillo, 1987: Synergistic interactions between an upper-level jet streak and diabatic processes that influence the development of a low-level jet and a secondary coastal cyclone. Mon. Wea. Rev., 115 , 2227–2261.
Whitaker, J S., L W. Uccellini, and K F. Brill, 1988: A model-based diagnostic study of the rapid development phase of the Presidents’ Day cyclone. Mon. Wea. Rev., 116 , 2337–2365.
Zhang, D-L., and R A. Anthes, 1982: A high-resolution model of the planetary boundary layer—Sensitivity tests and comparisons with SESAME-79 data. J. Appl. Meteor., 21 , 1594–1609.
Zhang, F., C. Snyder, and R. Rotunno, 2002: Mesoscale predictability of the “surprise” snowstorm of 24–25 January 2000. Mon. Wea. Rev., 130 , 1617–1632.
Zhang, F., C. Snyder, and R. Rotunno, 2003: Effects of moist convection on mesoscale predictability. J. Atmos. Sci., 60 , 1173–1185.
Zupanski, M., D. Zupanski, D F. Parrish, E. Rogers, and G. DiMego, 2002: Four-dimensional variational data assimilation for the blizzard of 2000. Mon. Wea. Rev., 130 , 1967–1988.
(a) Objective analysis of cooperative observer liquid-equivalent precipitation reports (mm) from 24 to 26 Jan 2000. (b) Eta Model 48-h forecast of liquid-equivalent precipitation (mm) ending at 0000 UTC 26 Jan 2000 from the 0000 UTC 24 Jan 2000 run.
Citation: Monthly Weather Review 133, 7; 10.1175/MWR2959.1
Radar mosaic reflectivity at 0900 UTC 24 Jan with 6-h observed precipitation totals ending at 1200 UTC (italicized numerals, mm) and Eta Model 6-h precipitation forecast ending at 1200 UTC (solid contours every 5 mm) from the 0000 UTC 24 Jan run. The black arrow indicates the region of incipient precipitation described in the text.
Citation: Monthly Weather Review 133, 7; 10.1175/MWR2959.1
RUC analysis of 900–700-hPa PV (black solid contours every 0.25 PVU starting at 0.5 PVU) and radar mosaic reflectivity imagery (see color scale in Fig. 2) at (a) 0600 UTC 24 Jan, (b) 0900 UTC 24 Jan, (c) 1200 UTC 24 Jan, and (d) 1800 UTC 24 Jan.
Citation: Monthly Weather Review 133, 7; 10.1175/MWR2959.1
The 700-hPa moisture flux (shading, g kg−1 m s−1), wind barbs (kt), and 900–700-hPa PV (solid contours every 0.25 PVU starting at 0.5 PVU) from (a) Eta Model 24-h forecast valid at 0000 UTC 25 Jan and (b) RUC analysis at 0000 UTC 25 Jan.
Citation: Monthly Weather Review 133, 7; 10.1175/MWR2959.1
MM5 domain used in simulation of 24–25 Jan 2000 case.
Citation: Monthly Weather Review 133, 7; 10.1175/MWR2959.1
Mean-state 875–825-hPa layer PV from 14-day MM5 simulation contoured and shaded every 0.2 PVU starting at 0.4 PVU.
Citation: Monthly Weather Review 133, 7; 10.1175/MWR2959.1
RUC model analysis of 250-hPa isotachs (shaded above 50 m s−1), wind barbs (m s−1), divergence (dashed contours, × 10−5 s−1), and geopotential height (solid contours, dam) valid at (a) 0000 UTC 24 Jan, (b) 1200 UTC 24 Jan, (c) 0000 UTC 25 Jan, and (d) 1200 UTC 25 Jan.
Citation: Monthly Weather Review 133, 7; 10.1175/MWR2959.1
As in Fig. 7, except showing 500-hPa absolute vorticity (shaded above 12 × 10−5 s−1) and geopotential height (solid contours, dam).
Citation: Monthly Weather Review 133, 7; 10.1175/MWR2959.1
Manual surface analysis of sea level pressure (solid contours every 4 hPa) and temperature (dashed contours every 5°C). Station reports and surface fronts are indicated by standard convention, with numerous observations omitted from the figure to make those shown more legible. Analyses are valid at (a) 0000 UTC 24 Jan, (b) 1200 UTC 24 Jan, (c) 0000 UTC 25 Jan, and (d) 1200 UTC 25 Jan.
Citation: Monthly Weather Review 133, 7; 10.1175/MWR2959.1
(a) Geostationary Operational Environmental Satellite-8 (GOES-8) infrared satellite image and RUC analysis of 300-hPa isotachs (solid contours, m s−1) at 1200 UTC 24 Jan; (b) as in (a) except for 0000 UTC 25 Jan. Analyzed surface low position indicated by “L.”
Citation: Monthly Weather Review 133, 7; 10.1175/MWR2959.1
Radar mosaic reflectivity imagery from (a) 0000 UTC 24 Jan, (b) 1200 UTC 24 Jan, (c) 0000 UTC 25 Jan, and (d) 1200 UTC 25 Jan.
Citation: Monthly Weather Review 133, 7; 10.1175/MWR2959.1
RUC analysis at 0600 UTC 24 Jan of (a) 900-hPa frontogenesis [solid contours every 0.5 K 100 km−1 (3 h)−1] and radar mosaic reflectivity imagery (shading, for this same time is shown in Fig. 3a); (b) cross section from 33°N, 89.5°W to 29°N, 88°W showing frontogenesis [solid contours every 0.5 K 100 km−1 (3 h)−1], potential temperature (dotted contours every 3 K), ageostrophic circulation (arrows), and mixing ratio (g kg−1, shaded); (c) 850–500-hPa Q vectors (arrows, × 10−9 K m−1 s−1) and Q-vector convergence (shaded, × 10−14 K m−2 s−1); and (d) as in (b) except equivalent-potential temperature (contours every 4 K) and negative moist geostrophic potential vorticity (shaded, PVU). The thick solid line in (a) and (c) represents the cross section displayed in (b) and (d).
Citation: Monthly Weather Review 133, 7; 10.1175/MWR2959.1
Results from MM5 simulation showing (a) sea level pressure (solid contours every 4 hPa), 2-m temperature (dashed contours every 5°C), 10-m winds (barbs, kt), and simulated grid-scale radar reflectivity (shaded, dBZ) valid at 1200 UTC 24 Jan; (b) as in (a) except at 0000 UTC 25 Jan; (c) as in (a) except at 1200 UTC 25 Jan; and (d) total precipitation (shaded, mm) ending at 0000 UTC 26 Jan.
Citation: Monthly Weather Review 133, 7; 10.1175/MWR2959.1
The 900–700-hPa potential vorticity (PVU, contoured every 0.25 PVU starting at 0.75 PVU) and 700-hPa temperature tendency (shaded, ×10−5 K s−1) from grid-scale precipitation and cumulus parameterization schemes with heating (cooling) shaded light to dark (dark to light) valid at (a) 1200 UTC 24 Jan, (b) 1500 UTC 24 Jan, (c) 1800 UTC 24 Jan, and (d) 0000 UTC 25 Jan from MM5 simulation. Solid black line in (b) shows cross section displayed in Fig. 15.
Citation: Monthly Weather Review 133, 7; 10.1175/MWR2959.1
Cross section from 32.5°N, 86°W to 30.5°N, 79°W from MM5 simulation at 1500 UTC 24 Jan, showing (a) potential vorticity (shaded, PVU), temperature tendency (contours, × 10−5 K s−1) from explicit moisture and cumulus parameterization schemes with heating (cooling) solid (dashed) and absolute vorticity vectors (b) As in (a) except arrows are nonadvective PV flux vectors, and solid (dashed) contours are positive (negative) nonadvective PV tendency in units of PVU day−1. Cross-sectional line is shown in Fig. 14b.
Citation: Monthly Weather Review 133, 7; 10.1175/MWR2959.1
Nondimensionalized cyclonic potential vorticity anomaly in the 875–825-hPa layer (shaded above 0.1), resultant 850-hPa balanced flow (barbs, kt), and geopotential height (solid contours, m) from the inversion of all cyclonic perturbation PV between 950 and 600 hPa at (a) 1500 UTC 24 Jan, (b) 1800 UTC 24 Jan, (c) 2100 UTC 24 Jan, and (d) 0000 UTC 25 Jan.
Citation: Monthly Weather Review 133, 7; 10.1175/MWR2959.1
(a) Magnitude of 750-hPa piecewise horizontal moisture flux (shaded, g kg−1 m s−1) and piecewise moisture flux vectors (arrows, g kg−1 m s−1) from the inversion of all cyclonic perturbation PV between 950 and 600 hPa, and nondimensionalized cyclonic 875–825-hPa PV anomaly (solid contours every 0.1), valid at 2100 UTC 24 Jan; (b) cross section from 37°N, 81.5°W to 32°N, 76°W showing piecewise horizontal moisture flux (shaded) and piecewise moisture flux vectors (arrows). Solid line in (a) shows cross section displayed in (b).
Citation: Monthly Weather Review 133, 7; 10.1175/MWR2959.1
The 750-hPa total horizontal moisture flux (shaded, g kg−1 m s−1), total horizontal moisture flux vectors (arrows, g kg−1 m s−1), and 900–700-hPa PV (solid contours every 0.5 PVU) from the MM5 simulation valid at 2100 UTC 24 Jan.
Citation: Monthly Weather Review 133, 7; 10.1175/MWR2959.1
The 850-hPa piecewise Q vectors (arrows, × 10−9 K m−1 s−1) and Q-vector convergence (shaded, × 10−15 K m−2 s−1), 850-hPa height from PV inversion (solid contours, m), and 850-hPa temperature (dotted contours, °C) from the MM5 simulation valid at 2100 UTC 24 Jan.
Citation: Monthly Weather Review 133, 7; 10.1175/MWR2959.1
