• Cortinas, J., 2000: A climatology of freezing rain in the Great Lakes region of North America. Mon. Wea. Rev, 128 , 35743588.

  • DeGaetano, A. T., 2000: Climatic perspective and impacts of the 1998 northern New York and New England ice storm. Bull. Amer. Meteor. Soc, 81 , 237254.

    • Search Google Scholar
    • Export Citation
  • Dudhia, J., 1989: Numerical study of convection observed during the Winter Monsoon Experiment using a mesoscale two-dimensional model. J. Atmos. Sci, 46 , 30773107.

    • Search Google Scholar
    • Export Citation
  • Jones, K. F., , and N. D. Mulherin, 1998: An evaluation of the severity of the January 1998 ice storm in northern New England. U.S. Army Corps of Engineers Cold Regions Research and Engineering Laboratory Report, 67 pp. [Available from CRREL, 72 Lyme Road, Hanover, NH 03755.].

    • Search Google Scholar
    • Export Citation
  • Kain, J. S., , and J. M. Fritsch, 1993: Convective parameterization for mesoscale models: The Kain–Fritsch scheme. The Representation of Cumulus Convection in Numerical Models, Meteor. Monogr., No. 46, Amer. Meteor. Soc., 165–170.

    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc, 77 , 437471.

  • NCDC, cited 1998: Service assessment: The ice storm and flood of January 1998. [Available online at http://www.nws.noaa.gov/om/iceflood.pdf.].

    • Search Google Scholar
    • Export Citation
  • NCDC, 1999: Billion dollar U.S. weather disasters 1980–1998. National Climatic Data Center World Wide Web document. [Available online at http://www.ncdc.noaa.gov/ol/reports/billionz.html.].

    • Search Google Scholar
    • Export Citation
  • 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, 124B , 10711108.

    • Search Google Scholar
    • Export Citation
  • Roebber, P. J., , and L. F. Bosart, 1998: The sensitivity of precipitation to circulation details. Part I: An analysis of regional analogs. Mon. Wea. Rev, 126 , 437455.

    • Search Google Scholar
    • Export Citation
  • Stauffer, D. R., , and N. L. Seaman, 1990: Use of four-dimensional data assimilation in a limited-area mesoscale model. Part I: Experiments with synoptic-scale data. Mon. Wea. Rev, 118 , 12501277.

    • Search Google Scholar
    • Export Citation
  • Trenberth, K. E., , and C. J. Guillemot, 1995: Evaluation of the global atmospheric moisture budget as seen from analyses. J. Climate, 8 , 22552272.

    • Search Google Scholar
    • Export Citation
  • 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 , 15941609.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Precipitation analysis for the 6-day period ending at 1200 UTC 10 Jan 1998, with solid contour interval of 10 mm, beginning with 50 mm. Heavy solid (dashed) contour is that of the 0°C isotherm at 850 hPa (surface) for 1200 UTC 5 Jan 1998. Very heavy solid (dashed) contour is that of the 0°C isotherm at 850 hPa (surface) for 0000 UTC 9 Jan 1998. The location of Montreal is indicated by the open circle

  • View in gallery

    (a) Time-mean sea level pressure (heavy solid, interval of 4 hPa) and 1000–500-hPa thickness (dashed, interval of 60 m) for the 5–9 Jan 1998 period. Time-mean anomalies of (b) sea level pressure [heavy contours with interval of 4 hPa and solid (dashed) for positive (negative) values], (c) 1000–500-hPa thickness [heavy contours with interval of 60 m with solid (dashed) for positive (negative) values], and (d) 1000–925 hPa thickness [heavy contours with interval of 7 m with solid (dashed) for positive (negative) values]. Thickness anomaly contour intervals in (c) and (d) correspond to approximately 3°C

  • View in gallery

    Sea level pressure (heavy solid, interval of 8 hPa) and 1000–500-hPa thickness (dashed, interval of 60 m) at 1200 UTC for (a) 5, (b) 6, (c) 7, and (d) 8 Jan 1998

  • View in gallery

    Anomaly correlations of the designated fields, plotted each 12 h, from 0000 UTC 5 Jan through 0000 UTC 9 Jan 1998, for the region 20°–60°N and 10°–130°W

  • View in gallery

    Eulerian moisture budget in the ice storm precipitation zone, 6 hourly, from 0000 UTC 4 Jan through 0000 UTC 9 Jan 1998, with units of mm (24 h)−1. Each term is as shown in Eq. (2), except for “−divQ,” which is the first term on the right-hand side of (2) and computed as a residual from the other terms

  • View in gallery

    Representative trajectories for air parcels arriving in the precipitation region at (a) 1200 UTC 5 Jan, and (b) 0000 UTC 9 Jan, 1998. All beginning points occur at 0000 UTC 26 Dec 1997. The pressure levels (hPa) correspond to the shading convention indicated in the inset. Eight trajectories are displayed in (a) and 10 in (b)

  • View in gallery

    (a) Percentage of all segments of trajectories, whose terminus is in the precipitation region at 1200 UTC 5 Jan, which are contained within each 360 km × 360 km region below and at 2 km. (b) As in (a) except for trajectories ending at 0000 UTC 9 Jan. (c) Percentage of all segments of trajectories, whose terminus is in the precipitation region at 1200 UTC 5 Jan, which are contained within each 360 km × 360 km region above 2 km. (d) As in (c) except at 0000 UTC 9 Jan

  • View in gallery

    Paths of three trajectories shown in Fig. 6, whose characteristics are analyzed in Figs. 9 (I), 10 (II), and 11 (III). The 194 400 km2 boxed region is that used for the precipitation zone's moisture budget in Fig. 5. Tick marks indicate 12-hourly positions, beginning with the initial time

  • View in gallery

    Diagnoses of parcel I (shown in Figs. 6a, 8) each 6 h for (a) pressure (hPa) and equivalent potential temperature (K), (b) mixing ratio q (g kg−1), and the summation of evaporation minus precipitation (mm day−1), and (c) instantaneous EP (mm day−1) and latent heat flux (W m−2). Beginning (end) point of the trajectory are indicated with A (B) or A′ (B′), as appropriate. The beginning and end points correspond to those shown for parcel I in Fig. 8. The abscissa (km) shows the distance of each parcel from Montreal (location shown in Fig. 1)

  • View in gallery

    As in Fig. 9 except for parcel II (shown in Figs. 6b, 8)

  • View in gallery

    As in Fig. 9 except for parcel III (shown in Figs. 6b, 8)

  • View in gallery

    Anomaly correlations of (a) sea level pressure, (b) 1000–925-hPa thickness, and (c) 1000–500-hPa thickness, plotted each 12 h, from 0000 UTC 5 Jan through 0000 UTC 9 Jan 1998, for the region 20°–60°N and 10°–130°W, as in Fig. 4. The five best analog anomaly correlations are plotted (year after 1900 indicated in the legend) on the same timescale as in the 1998 ice storm case

  • View in gallery

    Time-mean anomalies of 1000–500-hPa thickness [heavy contours with interval of 60 m with solid (dashed) for positive (negative) values] for the (a) 1966, (b) 1967, (c) 1974, and (d) 1991 analog cases

  • View in gallery

    Time-mean anomalies of sea level pressure [heavy contours with interval of 4 hPa and solid (dashed) for positive (negative) values] for the (a) 1966, (b) 1967, (c) 1974, and (d) 1991 analog cases

  • View in gallery

    Time-mean anomalies of 1000–925-hPa thickness (heavy contours with interval of 7 m with solid (dashed) for positive negative values] for the (a) 1966, (b) 1967, (c) 1974, and (d) 1991 analog cases

  • View in gallery

    Precipitation analysis for the 5-day period ending at 1200 UTC 27 Jan 1967, with solid contour interval of 10 mm, beginning with 50 mm. Dotted contours show the 20-, 30-, and 40-mm contours. Heavy solid (dashed) contour is that of the 0°C isotherm at 850 hPa (surface) for 1200 UTC 23 Jan 1967. Very heavy solid (dashed) contour is that of the 0°C isotherm at 850 hPa (surface) for 1200 UTC 25 Jan 1967

  • View in gallery

    As in Fig. 5 except for the period from 0000 UTC 22 Jan through 1200 UTC 27 Jan 1967

  • View in gallery

    Paths of three trajectories (IV, V, and VI) discussed in the text for the 1967 analog case, whose characteristics are analyzed in Figs. 19 (IV), 20 (V), and 21 (VI). Beginning times are shown as the Jan 1967 date/time. Ending times are shown as 23/12 for the arrowhead at the end of the westernmost trajectory, and as 25/18 for the two arrowheads at the ends of the two other trajectories. The 194 400 km2 boxed region is that used for the precipitation zone's moisture budget in Fig. 17. Tick marks indicate 12-hourly positions, beginning with the initial time

  • View in gallery

    Diagnoses of parcel IV (shown in Fig. 18) each 6 h for (a) pressure (hPa) and equivalent potential temperature (K); (b) mixing ratio q (g kg−1), and the summation of evaporation minus precipitation (mm day−1); and (c) instantaneous EP (mm day−1) and latent heat flux (W m−2). Beginning (end) point of the trajectory is indicated with A (B) or A′ (B′), as appropriate. The beginning and end points correspond to those shown for parcel IV in Fig. 18. The abscissa (km) shows the distance of each parcel from Montreal (location shown in Fig. 1)

  • View in gallery

    As in Fig. 19 except for parcel V (shown in Fig. 18)

  • View in gallery

    As in Fig. 19 except for parcel VI (shown in Fig. 18)

  • View in gallery

    Precipitable water (mm) for (a) 0000 UTC 6 Jan 1998, (b) 0000 UTC 9 Jan 1998, (c) 1200 UTC 23 Jan 1967, and (d) 1200 UTC 25 Jan 1967

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 82 82 49
PDF Downloads 43 43 27

The 1998 Ice Storm—Analysis of a Planetary-Scale Event

View More View Less
  • 1 Department of Atmospheric and Oceanic Sciences, McGill University, Montreal, Quebec, Canada
  • 2 Atmospheric Science Group, Department of Mathematical Sciences, University of Wisconsin–Milwaukee, Milwaukee, Wisconsin
© Get Permissions
Full access

Abstract

The ice storm of 5–9 January 1998, affecting the northeastern United States and the eastern Canadian provinces, was characterized by freezing rain amounts greater than 100 mm in some areas. The event was associated with a 1000–500-hPa positive (warm) thickness anomaly, whose 5-day mean exceeded +30 dam (+15°C) over much of New York and Pennsylvania. The region of maximum precipitation occurred in a deformation zone between an anomalously cold surface anticyclone to the north and a surface trough axis extending from the Gulf of Mexico into the Great Lakes. The thermodynamic impact of this unprecedented event was studied with the use of a four-dimensional data assimilation spanning an 18-day period ending at 0000 UTC 9 January 1998. A moisture budget for the precipitation region reveals the bulk of the precipitation to be associated with the convergence of water vapor transport throughout the precipitation period. The ice storm consisted of two primary synoptic-scale cyclonic events. The first event was characterized by trajectories arriving in the precipitation zone that had been warmed and moistened by fluxes over the Gulf Stream Current and the Gulf of Mexico. The second and more significant event was associated with air parcels arriving in the precipitation zone that had been warmed and moistened for a period of several days in the planetary boundary layer (PBL) of the subtropical Atlantic Ocean. These parcels had equivalent potential temperatures of approximately 330 K at 800 hPa as they traveled into the ice storm's precipitation zone.

Analogs to this unprecedented meteorological event were sought by examining anomaly correlations (ACs) of sea level pressure, and 1000–925 and 1000–500-hPa thicknesses. Five analogs to the ice storm were found, four of which are characterized by extensive freezing rain. The best analog, that of 22–27 January 1967, is characterized by freezing rain extending from the northeastern United States into central Ontario. However, the maximum amounts are less than 50% of the 1998 case. An examination of air parcel trajectories for the 1967 case reveals a similar-appearing horizontal spatial structure of trajectories, with several traveling anticyclonically from the subtropical regions of the eastern Atlantic. However, a crucial distinguishing characteristic of these trajectories in the 1967 case is that the air parcels arriving in the precipitation zone had equivalent potential temperature values of only 310 K, as compared with 330 K for the 1998 ice storm trajectories. It was found that these air parcels had traveled above the PBL and, therefore, had not been warmed and moistened by fluxes from the subtropical oceans.

Corresponding author address: Dr. John R. Gyakum, Department of Atmospheric and Oceanic Sciences, McGill University, 805 Sherbrooke Street West, Montreal, QC H3A 2K6, Canada. Email: gyakum@zephyr.meteo.mcgill.ca

Abstract

The ice storm of 5–9 January 1998, affecting the northeastern United States and the eastern Canadian provinces, was characterized by freezing rain amounts greater than 100 mm in some areas. The event was associated with a 1000–500-hPa positive (warm) thickness anomaly, whose 5-day mean exceeded +30 dam (+15°C) over much of New York and Pennsylvania. The region of maximum precipitation occurred in a deformation zone between an anomalously cold surface anticyclone to the north and a surface trough axis extending from the Gulf of Mexico into the Great Lakes. The thermodynamic impact of this unprecedented event was studied with the use of a four-dimensional data assimilation spanning an 18-day period ending at 0000 UTC 9 January 1998. A moisture budget for the precipitation region reveals the bulk of the precipitation to be associated with the convergence of water vapor transport throughout the precipitation period. The ice storm consisted of two primary synoptic-scale cyclonic events. The first event was characterized by trajectories arriving in the precipitation zone that had been warmed and moistened by fluxes over the Gulf Stream Current and the Gulf of Mexico. The second and more significant event was associated with air parcels arriving in the precipitation zone that had been warmed and moistened for a period of several days in the planetary boundary layer (PBL) of the subtropical Atlantic Ocean. These parcels had equivalent potential temperatures of approximately 330 K at 800 hPa as they traveled into the ice storm's precipitation zone.

Analogs to this unprecedented meteorological event were sought by examining anomaly correlations (ACs) of sea level pressure, and 1000–925 and 1000–500-hPa thicknesses. Five analogs to the ice storm were found, four of which are characterized by extensive freezing rain. The best analog, that of 22–27 January 1967, is characterized by freezing rain extending from the northeastern United States into central Ontario. However, the maximum amounts are less than 50% of the 1998 case. An examination of air parcel trajectories for the 1967 case reveals a similar-appearing horizontal spatial structure of trajectories, with several traveling anticyclonically from the subtropical regions of the eastern Atlantic. However, a crucial distinguishing characteristic of these trajectories in the 1967 case is that the air parcels arriving in the precipitation zone had equivalent potential temperature values of only 310 K, as compared with 330 K for the 1998 ice storm trajectories. It was found that these air parcels had traveled above the PBL and, therefore, had not been warmed and moistened by fluxes from the subtropical oceans.

Corresponding author address: Dr. John R. Gyakum, Department of Atmospheric and Oceanic Sciences, McGill University, 805 Sherbrooke Street West, Montreal, QC H3A 2K6, Canada. Email: gyakum@zephyr.meteo.mcgill.ca

1. Introduction

The freezing rain event of 5–9 January 1998 devastated regions of northern New York and New England in the United States, and southern regions of the Canadian provinces of Quebec, Ontario, and New Brunswick. This meteorological event was unprecedented during the past two decades for this region of North America (DeGaetano 2000), both in terms of the quantity of the precipitation in that phase and its persistence, and in terms of economic damage [approximately $4.4 billion, with $3 billion in Canada alone; NCDC (1999)]. Between 80 and 100 mm of freezing rain fell in the Montreal region, leading to accreted radial ice loads of 45 mm. Radial ice loads as high as 125 mm were reported in some areas (Jones and Mulherin 1998). More than 3 million utility customers lost power in Canada and 1 million in the northeastern United States, including 80% of Maine's population (NCDC 1998). There were 56 fatalities attributed to the event (12 of these were the result of flooding rains in the southeastern United States, where up to 400 mm accumulated in a 2-day period).

Figure 1 presents an objective analysis (based upon approximately 450 station reports) of the total precipitation for the 5–9 January 1998 period in the northeastern United States and southeastern Canada. The key signature of this event is the accumulated precipitation in excess of 90 mm across a wide region of subfreezing surface temperatures coincident with 850-hPa temperatures well above 0°C. The intensity, persistence, and geographic focus of the precipitation, in combination with the persistence of the surface-based inversion, led to the widespread devastation highlighted above.

Given this unprecedented meteorological character, we intend to define and establish the thermodynamic impact of the event's persistent planetary-scale dynamical signature. Our next goal is to search for large-scale analogs to this event. A more detailed understanding of the thermodynamic processes associated with one analog will provide us with insight as to what specific processes are responsible for distinguishing the significant meteorological event from a benign case. We will focus on the planetary-scale circulation features that made moist, high equivalent potential temperature (θe) air available in the ice storm region for an extended period.

The structure of the paper is as follows: section 2 presents analyses of the planetary-scale signature of the event, documents its persistence, and elucidates the role of this structure in the moisture history of the event, through a combination of moisture budget and trajectory analyses. We document five analogs to the 1998 ice storm in section 3 and compare the impacts. We also study the large-scale thermodynamics of the best analog case and compare these results with those found in section 2. The conclusions, including a discussion of the implications of these results for the predictability of such events, are presented in section 4.

2. Planetary-scale features and attendant moisture history

a. Planetary-scale features

To investigate the planetary-scale structures associated with this event, we utilize height fields on a 2.5° latitude–longitude grid at 500, 925 and 1000 hPa obtained from the National Centers for Environmental Prediction (NCEP) reanalysis dataset (Kalnay et al. 1996). We produce anomaly (the difference between the actual value and the 30-year January climatology) fields every 12 h during the 5–9 January 1998 period. The period of 5–9 January 1998 is marked by a persistent cold anticyclone located in eastern Canada, an anomalously warm inverted trough extending from the Gulf of Mexico into the Great Lakes, and an anticyclone in the southwestern North Atlantic (Figs. 2, 3). The placement of the freezing rain in the northeastern quadrant of a surface cyclone, with a relatively strong surface anticyclone to the north, has also been documented by Cortinas (2000) for a composite of eastern Great Lakes freezing rain events. The anomalously strong southerly geostrophic flow occurring during the 5–9 January 1998 period (Fig. 2b) is associated with anomalously warm temperatures in a deep layer, centered in the northeastern United States (Fig. 2c). However, despite this deep warmth, surface-based air temperatures remained below 0°C (Fig. 1) as a result of cold northeasterly surface flow from the shallow cold air in an anticyclone to the north (Fig. 2d). Figures 3a–d indicate that these structures persisted throughout the period, with synoptic-scale lift provided by the passage of individual cyclones on 6 January (Fig. 3b) and 8 January (Fig. 3d). To quantify the degree of this persistence, we compute anomaly correlations (ACs) between composite fields and individual, 12-hourly maps in the region 130°–10°W and 20°–60°N using
i1520-0493-129-12-2983-e1
where the a(1)ij and a(2)ij denote the anomalies at each grid point (i, j) for the period composite and the individual map, respectively. This analysis (Fig. 4) reveals extraordinarily high AC between the 5–9 January 1998 composite fields and individual maps during that entire period. The 1000–925-hPa thickness stands out with values greater than 0.9 during 3 out of 4 days. More typically, ACs for events in areas of comparable or smaller scale fluctuate rapidly in association with mobile weather systems and the associated rapid evolution of the mass field (e.g., Roebber and Bosart 1998).

b. Moisture budgets and trajectory analysis

We relate the persistent planetary-scale features discussed in section 2a to the moisture budget of the precipitation region (Fig. 5) and to the moisture history of individual air parcels (Figs. 6 and 7) constituting the atmospheric structure at key times and locations during the event. To perform these calculations, a diagnostic dataset is generated using the Pennsylvania State University–National Center for Atmospheric Research (PSU–NCAR) fifth-generation Mesoscale Model (MM5). A single domain, employing 90-km grid spacing, is constructed for the region depicted in Fig. 6. Model initialization and lateral boundary conditions are determined as follows: first-guess fields of atmospheric variables (wind, temperature, and moisture) are obtained from the NCEP historical archives of global (2.5° latitude by 2.5° longitude) mandatory-level analyses and are then adjusted using a Cressman-type objective analysis of surface and rawinsonde data for all stations within or near the grid domain. These analyses also provide boundary conditions on the outermost grid domain throughout the course of the integrations. Vertical sigma levels are arranged such that the model output is available on a total of 23 levels (0.995, 0.985, 0.970, 0.945, 0.910, 0.870, 0.825, 0.775, 0.725, 0.675, 0.625, 0.575, 0.525, 0.475, 0.425, 0.375, 0.325, 0.275, 0.225, 0.175, 0.125, 0.075, 0.025), with a relatively large concentration at the lowest levels in order to resolve planetary boundary layer (PBL) structure. Precipitation processes are simulated using the Kain–Fritsch (1993) convective parameterization and an explicit moisture scheme that includes prognostic equations for cloud water, rainwater, ice, and supercooled water (Reisner et al. 1998). This latter scheme does not include graupel or riming processes, but does allow for time-dependent melting (freezing) of snow/ice (water) below (above) the freezing level. Radiative processes are computed using a cloud–radiation scheme in which diurnally varying shortwave and longwave radiative fluxes interact with explicitly cloudy and clear air, while the surface fluxes are used in the ground energy budget calculations (Dudhia 1989). The PBL is modeled using the high-resolution Blackadar scheme (Zhang and Anthes 1982). Physiographic and land use patterns are back-interpolated from a global high-resolution topographic and land use dataset (with resolution of approximately 19 km) to the model grids.

To generate an analysis dataset that provides time continuity and dynamic coupling among the various model fields, a continuous four-dimensional data assimilation (FDDA) known as Newtonian relaxation is employed within MM5 (see Stauffer and Seaman 1990). The model solution is nudged toward three-dimensional gridded analyses of wind, temperature, and water vapor mixing ratio. The integration commences at 0000 UTC 21 December 1997 and ends at 0000 UTC 9 January 1998. A subjective comparison of the model output with the synoptic analyses indicates excellent agreement throughout the period of integration. Furthermore, we compare the mean of 15 precipitation reports for the 96-h period ending at 1200 UTC 8 January with those of the MM5 for the same locations interpolated from the grid. The observed 15-station mean is 50.5 mm and the MM5 mean is 47.4 mm. These results suggest that the MM5 FDDA succeeds in reproducing the coarse-mesh character of the ice storm precipitation.

An Eulerian moisture budget is conducted in the 194 400 km2 precipitation zone (shown as the boxed region in Fig. 8) of the event, using the equation (Trenberth and Guillemot 1995)
i1520-0493-129-12-2983-e2
where w is the precipitable water (PW) defined by
i1520-0493-129-12-2983-e3
q the specific humidity, vh the horizontal wind velocity, Psfc (Ptop) the pressures at the surface (top) of the atmosphere, E the evaporation from the surface, and P the precipitation. The first term on the right-hand side of (2) represents the divergence of the total integrated moisture transport, which can be partitioned into contributions that include mass divergence in the presence of water vapor (−q∇·vh) and moisture advection (−vh·∇q). The second term on the right-hand side of (2) represents the local time rate of change of atmospheric PW and represents storage in the column. Precipitation and evaporation are explicit model outputs.

The results of this analysis are presented in Fig. 5. The role of the cyclones on 5–6 and 8–9 January 1998 in the production of precipitation is indicated by the peaks in the precipitation at these times (Figs. 3a, 3d, and 5). Clearly, local sources of moisture make no substantial contribution to the overall moisture budget in the region during this period. Rather, moisture flux convergence (primarily through convergence in the presence of water vapor but with non-negligible contribution from horizontal advection) provides the means for the extended period of somewhat lighter freezing precipitation rates (10–20 mm day−1) on 5–6 January and the more intense freezing rain episode (20–40 mm day−1) of 8–9 January 1998.

To find the source of this moisture, we use the diagnostic dataset to compute all trajectories that entered the precipitation zone at 1200 UTC 5 January and at 0000 UTC 9 January 1998. The precipitation zone (using this criterion, 194 400 km2 for 1200 UTC 5 Jan and 550 800 km2 for 0000 UTC 9 Jan is defined at the surface by all model grid points at which precipitation is occurring. The vertical extent of the precipitation zone at each of these grid points is defined by relative humidity values with respect to water in excess of 70%. The trajectory calculation employs a simple forward difference using gridded data of eastward (u), northward (υ), and upward (w) wind components at 180-km horizontal spacing (subsampled from the 90-km data), 500-m vertical spacing (interpolated from the model sigma coordinate), and a 1-h time step. An analysis of these trajectories entering the precipitation zone at these two times is conducted to provide additional details concerning the moisture history (Figs. 6–11).

Air parcels arriving in the precipitation zone (outlined by the box in Fig. 8) at 1200 UTC 5 January 1998 follow two distinct paths (Fig. 6a). The first path is a looplike trajectory in which upper-tropospheric parcels above northern Canada travel southward and descend to the PBL over the Gulf of Mexico. The parcels then return northward and eastward after having been warmed and moistened in the PBL over the Gulf of Mexico. The second is an east-to-northeast ascending trajectory of relatively dry air from the Pacific coast. Parcels arriving in the ice storm precipitation zone at 0000 UTC 9 January 1998 (Fig. 6b) follow two trajectories similar to those of 5 January 1998, but prominent additional paths are evident in Fig. 6b, namely, large, anticyclonic gyres traversing the North Atlantic. Although the origins of these latter parcels are upper tropospheric, they descend to the PBL as they arrive in the subtropical Atlantic (Fig. 6b).

Figure 7 illustrates a more complete statistical summary of the air parcels traveling into the precipitation regions at 1200 UTC 5 January and 0000 UTC 9 January. The former time is characterized by trajectories whose contacts in the PBL (below and at 2 km) in the marine regions are confined to the Gulf of Mexico and the Gulf Stream region eastward of the southeastern United States (Fig. 7a). The majority of these trajectories above 2 km are confined to the Great Lakes region (Fig. 7c). Figures 7b,d, illustrate the statistics for trajectories terminating in the precipitation zone at 0000 UTC 9 January. The more extensive contact in the marine PBL throughout the subtropical Atlantic is seen in Figs. 7b. Furthermore, the effects of the larger-scale anticyclonic gyre previously discussed for the trajectories ending at 0000 UTC 9 January are seen at upper levels in Fig. 7d with slightly more extensive residence in the subtropical Atlantic.

Since the moisture content for the looplike trajectory of Fig. 6a is substantially higher than that for the more zonal trajectories of Fig. 6a, the subsequent analysis focuses on a representative trajectory of the former type (denoted as trajectory I in Fig. 8). Figure 9 shows parcel I's vertical position and thermodynamic characteristics, as a function of horizontal position (the distance from Montreal). After the parcel's descent into the PBL (below 850 hPa), θe increases to approximately 320 K over the Gulf of Mexico (Fig. 9a). Rapid accession of moisture occurs during the 42-h period in which this parcel is contained within the PBL of the Gulf of Mexico, with mixing ratios rising from 2.6 to 10.5 g kg−1 (Fig. 9b). This moisture accession occurs as would be expected in a region of positive EP (Fig. 9c) in association with surface latent heat fluxes peaking at (averaging) 619 (330) W m−2. The surface latent heat fluxes (Fig. 9c) are relevant to the thermodynamic characteristics of the air parcel only when that air parcel is located in the PBL. The loss of moisture occurs as the parcel enters the precipitation region near the end of its trajectory, behavior typical of all the trajectories studied in this case.

Figures 6b, 7, 8 (trajectory II), and 10 show that the looplike trajectory has shifted eastward by 9 January 1998, such that extended contact with the PBL occurs in the subtropical Atlantic rather than the Gulf of Mexico. Moisture accession in this case occurs over an approximately 102-h period, with mixing ratios increasing from 4.6 to 13.7 g kg−1 (Fig. 10b). This moisture accession occurs in a region of positive EP (Fig. 10c) in conjunction with surface latent heat fluxes peaking at (averaging) 236 (151) W m−2. Hence, the extended residence within the subtropical Atlantic provides more heating and moistening of the air parcels than is the case with the shorter-term contact with the Gulf of Mexico (θe of 330 vs 315 K in the latter case).

However, parcel II's equivalent potential temperature decreases from 327 to 307 K (Fig. 10a) during the 12-h period ending at 1800 UTC 8 January. It is likely that this decrease occurs as a result of downward sensible heat transfer toward the cold ground of the northeast United States (Fig. 8). The FDDA sensible heat flux averages −60 W m−2 (downward; not shown) during this 12-h period. The corresponding temperature change for a PBL depth of 50 hPa (390 m) yields a cooling of 7 K (12 h)−1. A reduction of the PBL depth to 130 m corresponds to a cooling of 21 K (12 h)−1. Additional cooling could occur from parcel mixing with ambient cooler air.

The heating and moistening process is especially important in the case of the anticyclonic gyre trajectory (Fig. 6b, trajectory III in Fig. 8, and Fig. 11), whose air resides for approximately 120 h within the subtropical Atlantic in its path to the freezing rain zone. During this time, mixing ratios increased from 1.7 to 12.2 g kg−1 (Fig. 11b) in conjunction with surface latent heat fluxes peaking at (averaging) 218 (133) W m−2 and θe of approximately 330 K. Hence, the subtropical Atlantic is the primary source of the warmth, the moisture, and the resulting high equivalent potential temperature that ultimately is delivered to northern New York, New England, southeastern Ontario, and southern Quebec. The severity of this second synoptic-scale event is exemplified by Montreal receiving a record 24-h precipitation amount for January of 35.8 mm on 8 January.

3. Analogs

The unprecedented nature of this event does not necessarily imply that similar persistent, planetary-scale structures have not occurred in the historical record. For a particularly prominent large-scale signature, details concerning near-surface frontal structure, convective processes, and mesoscale forcing of ascent through frontal circulations may all strongly modulate the amount and phase of precipitation to this region. To examine this question, we perform a search for analogs to the anomaly structures, depicted in Fig. 2, using the NCEP reanalysis gridded data (Kalnay et al. 1996) for the period 1963–96. This search reveals five periods in which extended correspondence (greater than 3 days) with the planetary-scale anomaly correlation structures of sea level pressure, and 1000–500 and 1000–925 hPa-thickness, occurs (Fig. 12). The criterion for this extended correspondence is that the mean AC among the three fields be at least 0.6 for the 3-day period. These analog cases are listed in Table 1. A comparison of the 1000–500-hPa thickness anomalies (Fig. 13) for the four best cases (1966, 1967, 1974, 1991) with the corresponding anomaly field of 5–9 January 1998 (Fig. 2c) reveals similar structures. The anomalously warm air, focused over eastern North America, is flanked by cold anomalies to the east and west. However, only the 1967 case has a positive anomaly amplitude that is comparably strong to that of the 1998 ice storm. Structures in the sea level pressure anomalies (Fig. 14) are also generally similar to those seen in the ice storm (Fig. 2b). Although the 1000–925-hPa thickness anomaly structure for each analog case (Fig. 15) is generally similar to those observed in the 1998 ice storm (Fig. 2d), only the 1967 case has a comparable amplitude in both the warm and cold anomalies of eastern North America. Records from the publication Storm Data reveal that freezing rain occurs in four of the five analogs. Economic damage resulted from high winds in the fifth case, occurring in 1974. The best analog, in terms of the anomaly correlation (Fig. 12), is the 1967 case. This particular case is characterized by extensive freezing rain in New York state with widespread downed trees and power lines. Given the notable correspondence between the 1967 analog case and the 1998 ice storm, on both the regional and planetary scales, the 5-day precipitation analysis for the analog case nonetheless shows significant differences from that of the 1998 case (cf. Figs. 16, 1). Unlike the 1998 case, the analog is characterized by amounts only in the range of 20–35 mm in the Saint Lawrence River valley. The only amounts greater than 50 mm are displaced 450 km to the northwest of the 1998 ice storm amounts. While the analog case is characterized by subfreezing surface air and above-freezing 850-hPa temperatures, the orientation of this inversion region is northwest–southeast and is not as persistent as that of the 1998 ice storm.

To consider further comparisons, we compute the moisture budget of the 1967 analog case with the same procedure that we use for the 1998 ice storm. Figure 17 shows the results of this budget for the precipitation region of approximately 194 400 km2 (shown as the boxed region in Fig. 18). A comparison of the moisture budget in the analog case with that of a same-sized area in the 1998 ice storm (Fig. 5) reveals the analog case had only three periods in which rates as much as 10 mm (24 h)−1 occurred, as compared with seven such periods for the ice storm. Furthermore, among these seven periods in 1998, four occurring during the later stages of the ice storm are characterized by rates of 20 mm (24 h)−1 or greater. The 1967 analog case has no periods in which as much as 20 mm (24 h)−1, averaged over the region, occurred. As for the 1998 case, the analog case precipitation peaks are associated with substantial contributions from moisture flux convergence. Unlike the 1998 case, the early phases of the precipitation were associated with a positive contribution from the storage term. Nevertheless, we conclude that the lack of the significant precipitation events in the 1967 case is associated primarily with the lack of similarly strong events of convergence of the moisture transport.

Representative trajectories are computed from a four-dimensional data assimilation that begins at 1200 UTC 13 January and ends 1200 UTC 27 January 1967. An analysis of representative trajectories ending in the precipitation zone (shown as the 194 400 km2 boxed region in Fig. 18) reveals the geographical positions and timescales of their paths to be similar to those found in the 1998 case (cf. Figs. 18 and 8). Parcel IV (Fig. 18) begins in the western United States, travels into the region of the Gulf of Mexico, and curves northeastward into the precipitation region. Parcel V, originating along the western shores of the Hudson Bay, travels equatorward, and loops anticyclonically over the Gulf Stream current and the Gulf of Mexico before traveling into the Saint Lawrence River valley. The subtropical Atlantic is the region for much of parcel VI's path, before passing over the Gulf of Mexico and flowing poleward into the precipitation region.

Though the parcel trajectories associated with the precipitation appear similar in each of the 1998 and 1967 cases, it is necessary to compare the thermodynamic and vertical structures of these parcels. The details of parcel IV (Fig. 19) reveal a path of descent from 700 hPa into the PBL of the Gulf of Mexico, where its water vapor mixing ratio increases from 4.6 to 9.9 g kg−1. During this maritime passage of approximately 66 h, the latent heat flux peaks at (averages) 249 (112) W m−2 with θe increasing to 317 K. Parcel V's (Fig. 20) path from the Hudson Bay region is characterized by descent from 360 hPa to the PBL over the Gulf Stream where the latent heat flux peaks at 429 W m−2. During the 66-h period over this region, the latent heat fluxes average 194 W m−2, and the water vapor mixing ratio increases from 1.5 to 11.6 g kg−1. The most substantive contrast to that of the 1998 case (e.g., Fig. 11) occurs in the case of trajectory VI (Fig. 21) in which its passage over the subtropical Atlantic is characterized by only a 6-h period in the PBL. The water vapor mixing ratio increased from 3.9 to 4.6 g kg−1 and the latent heat flux peaks (averages) at only 7 (7) W m−2. The θe value of 310 K is substantially less than that of the 330-K value seen in the 1998 ice storm's parcel III (Fig. 11), which traveled along a similar path (cf. Figs. 8, 18).

A comparison of PW fields (Fig. 22) during the first precipitation events in each case (cf. Fig. 22a at 0000 UTC 6 Jan 1998, and Fig. 22c at 1200 UTC 23 Jan 1967) reveals a similar structure and closely corresponding area-averaged PW values. These similar-appearing fields correspond to the similar results from our moisture budget calculations for the first phase of precipitation in each case (e.g., Figs. 5 and 17). However, the PW fields at the approximate ending times for the trajectories III (0000 UTC 9 Jan 1998; Fig. 22b) and VI (1200 UTC 25 Jan 1967; Fig. 22d) reveal several differences. The PW values in the region of strongest precipitation for the 1998 ice storm (Fig. 1) through the Saint Lawrence River valley (approximately along the U.S.–Canadian border) exceeds that of the 1967 analog case by 8–12 mm. Additionally, the structure of the PW fields differs. The gradient of PW is oriented toward the west in the 1967 analog case, and to the south in the 1998 ice storm. There are corresponding differences in the spatial structures of the precipitation (cf. Figs. 1 and 16). Clearly, there are crucial details relating to water vapor that may account for the very different precipitation fields occurring in each of the 1998 and 1967 analog cases. These differences, corresponding to the different upstream trajectory characteristics (cf. Figs. 11 and 21), occur despite the excellent mass field analog characteristics of the 1967 case to that of the 1998 ice storm (Fig. 12; cf. Figs. 2b, 14b; cf. Figs. 2c, 13b; cf. Figs. 2d, 15b).

4. Concluding discussion

The 1998 ice storm of 5–9 January 1998 is an unprecedented meteorological combination of heavy precipitation, occurring primarily in the form of freezing rain, which persisted during this 5-day period. We show the affected regions to be located in a region of persistent easterly surface geostrophic flow in a deformation region between a warm cyclonic circulation to the south and a cold anticyclonic flow to the north (Figs. 2a, 3). The anomalies of sea level pressure, 1000–500-hPa thickness, and 1000–925-hPa thickness, as defined for the North American and western Atlantic regions (Figs. 2b–d), are used in an anomaly correlation calculation to demonstrate objectively the persistence of the event (Fig. 4).

A moisture budget, calculated from the output of a credible MM5 FDDA of the ice storm, reveals the event to be composed of two significant synoptic-scale precipitation occurrences, in which the convergence of the moisture transport dominated the contribution to the moisture budget (Fig. 5). An analysis of representative trajectories arriving in the precipitation zone during each of these two occurrences (at 1200 UTC 5 Jan and at 0000 UTC 9 Jan) reveals a distinct character to each. The earlier pulse is associated with parcels that descended from 300 to 400 hPa in the Hudson Bay region as they traveled equatorward into the Gulf Stream and Gulf of Mexico PBL, before ascending on an anticyclonic path back to the Saint Lawrence River valley (Fig. 6a). The later pulse is associated with 9-day passages of air parcels from the vicinity of Greenland near the tropopause anticyclonically toward the PBL of the subtropical Atlantic, through the Caribbean, and subsequently upward into the precipitation region (Fig. 6b). These results are confirmed by the statistics of the parcels shown in Fig. 7, in which maritime PBL contact is confined to the Gulf Stream and Gulf of Mexico for the first event (Fig. 7a) and a much more extensive maritime PBL contact encompasses the subtropical Atlantic (Fig. 7b). A thermodynamic analysis of three individual trajectories for the 1998 ice storm (Fig. 8) reveals the crucial importance of the persistent contact with the subtropical maritime PBL in which water vapor and equivalent potential temperature increases occur as a result of strong surface latent heat fluxes. These fluxes are especially important in the extended subtropical Atlantic trajectory III (Figs. 8, 11) in which the equivalent potential temperature values reach 330 K prior to reaching the ice storm precipitation region.

Despite the unprecedented nature of this ice storm, a search through the cold seasons from 1963 through 1996 reveals the presence of five analog cases in which an extended correspondence with the planetary-scale anomaly correlation structures of sea level pressure, and 1000–500 hPa and 1000–925-hPa thickness, occurred (Fig. 12, Table 1). Considering that the best correspondence occurs in the January 1967 case (Fig. 12), we analyze this case in the same manner as we did for the 1998 case. This analysis reveals both similarities and crucial differences between the cases. First, the 1967 case is associated with only about 50% of the maximum precipitation seen in the 1998 case (cf. Figs. 1, 16). Though there are reports of significant freezing rain and downed trees, the core of the freezing precipitation is displaced 450 km to the northwest of its occurrence in 1998. To understand why such an excellent analog as the 1967 case fails to produce a severe ice storm, we produce an MM5 FDDA to compare the thermodynamic structures of the two cases. A comparison of the moisture budgets for areas of the same 194 400 km2 dimension reveals comparably small precipitation rates and convergence of moisture transport for the 1967 case (cf. Figs. 17, 5). To understand possible reasons for this difference, we study back-trajectory analyses for air parcels arriving in the precipitation region, and find similar geographical structures to those found for the 1998 case (cf. Figs. 18, 8). However, a further analysis of three representative trajectories (Figs. 19, 20, 21) shows that these parcels have considerably less contact with the maritime PBL, particularly parcel VI in its passage over the subtropical Atlantic. Though the geographic structures of parcels III and VI appear similar (Figs. 8, 18), the crucial difference is that parcel III in 1998 benefits from an extensive increase in its equivalent potential temperature from latent heat fluxes. This effect is absent in parcel VI of the 1967 case (cf. Figs. 21, 11) because of its passage well above the PBL. The implication of this effect is that the parcel arrives in the precipitation region with an equivalent potential temperature that is 20 K less, and a mixing ratio of 6 g kg−1 less, than that seen in the 1998 ice storm.

Our results have potentially significant implications for the analysis and prediction of such an extreme event as the 1998 ice storm. Though the planetary-scale circulation structure has been repeated five times during a 34-yr period, the ice storm of 1998 is the only case of the six that produces such extraordinary amounts of freezing rain. Mesoscale details, such as frontal structure and details of the Saint Lawrence River valley orography, likely modulate the location and magnitude of the precipitation. However, the moistening and warming of the preferentially high equivalent potential temperature and moisture-laden air (Fig. 22), characteristic of heavy precipitation, is produced by remote processes. These processes include surface latent heat fluxes over the subtropical Atlantic Ocean. Not only is it crucial to correctly forecast these processes for a medium-range forecast, but properly initialized maritime data are necessary for both short- and medium-range forecasts. The air that transported the extremely large values of equivalent potential temperature into the ice storm region traveled northward from the South Carolina–Georgia coast in only 24 h (see trajectory III in Figs. 8, 11). Incorrectly initialized moisture and/or temperature fields for a case such as this has the potential to degrade a short-range forecast. Our research reporting on the mesoscale and short-range forecasting aspects of the ice storm 1998 will be reported upon in future work.

Acknowledgments

This work has been supported by a research grant from the National Science Foundation (NSF OCE-9726679), the Natural Sciences and Engineering Council of Canada, and a subvention from the Meteorological Service of Canada (MSC). We would like to express our deepest appreciation to Paul Sisson of the Burlington, Vermont, National Weather Service Forecast Office, and to Jim Abraham and William Richards of the MSC for providing precipitation data. We would like to thank Lance Bosart of the University at Albany and the three anonymous reviewers for their constructive comments. The Climate Diagnostics Center is acknowledged for providing the NCEP reanalysis data.

REFERENCES

  • Cortinas, J., 2000: A climatology of freezing rain in the Great Lakes region of North America. Mon. Wea. Rev, 128 , 35743588.

  • DeGaetano, A. T., 2000: Climatic perspective and impacts of the 1998 northern New York and New England ice storm. Bull. Amer. Meteor. Soc, 81 , 237254.

    • Search Google Scholar
    • Export Citation
  • Dudhia, J., 1989: Numerical study of convection observed during the Winter Monsoon Experiment using a mesoscale two-dimensional model. J. Atmos. Sci, 46 , 30773107.

    • Search Google Scholar
    • Export Citation
  • Jones, K. F., , and N. D. Mulherin, 1998: An evaluation of the severity of the January 1998 ice storm in northern New England. U.S. Army Corps of Engineers Cold Regions Research and Engineering Laboratory Report, 67 pp. [Available from CRREL, 72 Lyme Road, Hanover, NH 03755.].

    • Search Google Scholar
    • Export Citation
  • Kain, J. S., , and J. M. Fritsch, 1993: Convective parameterization for mesoscale models: The Kain–Fritsch scheme. The Representation of Cumulus Convection in Numerical Models, Meteor. Monogr., No. 46, Amer. Meteor. Soc., 165–170.

    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc, 77 , 437471.

  • NCDC, cited 1998: Service assessment: The ice storm and flood of January 1998. [Available online at http://www.nws.noaa.gov/om/iceflood.pdf.].

    • Search Google Scholar
    • Export Citation
  • NCDC, 1999: Billion dollar U.S. weather disasters 1980–1998. National Climatic Data Center World Wide Web document. [Available online at http://www.ncdc.noaa.gov/ol/reports/billionz.html.].

    • Search Google Scholar
    • Export Citation
  • 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, 124B , 10711108.

    • Search Google Scholar
    • Export Citation
  • Roebber, P. J., , and L. F. Bosart, 1998: The sensitivity of precipitation to circulation details. Part I: An analysis of regional analogs. Mon. Wea. Rev, 126 , 437455.

    • Search Google Scholar
    • Export Citation
  • Stauffer, D. R., , and N. L. Seaman, 1990: Use of four-dimensional data assimilation in a limited-area mesoscale model. Part I: Experiments with synoptic-scale data. Mon. Wea. Rev, 118 , 12501277.

    • Search Google Scholar
    • Export Citation
  • Trenberth, K. E., , and C. J. Guillemot, 1995: Evaluation of the global atmospheric moisture budget as seen from analyses. J. Climate, 8 , 22552272.

    • Search Google Scholar
    • Export Citation
  • 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 , 15941609.

    • Search Google Scholar
    • Export Citation
Fig. 1.
Fig. 1.

Precipitation analysis for the 6-day period ending at 1200 UTC 10 Jan 1998, with solid contour interval of 10 mm, beginning with 50 mm. Heavy solid (dashed) contour is that of the 0°C isotherm at 850 hPa (surface) for 1200 UTC 5 Jan 1998. Very heavy solid (dashed) contour is that of the 0°C isotherm at 850 hPa (surface) for 0000 UTC 9 Jan 1998. The location of Montreal is indicated by the open circle

Citation: Monthly Weather Review 129, 12; 10.1175/1520-0493(2001)129<2983:TISAOA>2.0.CO;2

Fig. 2.
Fig. 2.

(a) Time-mean sea level pressure (heavy solid, interval of 4 hPa) and 1000–500-hPa thickness (dashed, interval of 60 m) for the 5–9 Jan 1998 period. Time-mean anomalies of (b) sea level pressure [heavy contours with interval of 4 hPa and solid (dashed) for positive (negative) values], (c) 1000–500-hPa thickness [heavy contours with interval of 60 m with solid (dashed) for positive (negative) values], and (d) 1000–925 hPa thickness [heavy contours with interval of 7 m with solid (dashed) for positive (negative) values]. Thickness anomaly contour intervals in (c) and (d) correspond to approximately 3°C

Citation: Monthly Weather Review 129, 12; 10.1175/1520-0493(2001)129<2983:TISAOA>2.0.CO;2

Fig. 3.
Fig. 3.

Sea level pressure (heavy solid, interval of 8 hPa) and 1000–500-hPa thickness (dashed, interval of 60 m) at 1200 UTC for (a) 5, (b) 6, (c) 7, and (d) 8 Jan 1998

Citation: Monthly Weather Review 129, 12; 10.1175/1520-0493(2001)129<2983:TISAOA>2.0.CO;2

Fig. 4.
Fig. 4.

Anomaly correlations of the designated fields, plotted each 12 h, from 0000 UTC 5 Jan through 0000 UTC 9 Jan 1998, for the region 20°–60°N and 10°–130°W

Citation: Monthly Weather Review 129, 12; 10.1175/1520-0493(2001)129<2983:TISAOA>2.0.CO;2

Fig. 5.
Fig. 5.

Eulerian moisture budget in the ice storm precipitation zone, 6 hourly, from 0000 UTC 4 Jan through 0000 UTC 9 Jan 1998, with units of mm (24 h)−1. Each term is as shown in Eq. (2), except for “−divQ,” which is the first term on the right-hand side of (2) and computed as a residual from the other terms

Citation: Monthly Weather Review 129, 12; 10.1175/1520-0493(2001)129<2983:TISAOA>2.0.CO;2

Fig. 6.
Fig. 6.

Representative trajectories for air parcels arriving in the precipitation region at (a) 1200 UTC 5 Jan, and (b) 0000 UTC 9 Jan, 1998. All beginning points occur at 0000 UTC 26 Dec 1997. The pressure levels (hPa) correspond to the shading convention indicated in the inset. Eight trajectories are displayed in (a) and 10 in (b)

Citation: Monthly Weather Review 129, 12; 10.1175/1520-0493(2001)129<2983:TISAOA>2.0.CO;2

Fig. 7.
Fig. 7.

(a) Percentage of all segments of trajectories, whose terminus is in the precipitation region at 1200 UTC 5 Jan, which are contained within each 360 km × 360 km region below and at 2 km. (b) As in (a) except for trajectories ending at 0000 UTC 9 Jan. (c) Percentage of all segments of trajectories, whose terminus is in the precipitation region at 1200 UTC 5 Jan, which are contained within each 360 km × 360 km region above 2 km. (d) As in (c) except at 0000 UTC 9 Jan

Citation: Monthly Weather Review 129, 12; 10.1175/1520-0493(2001)129<2983:TISAOA>2.0.CO;2

Fig. 8.
Fig. 8.

Paths of three trajectories shown in Fig. 6, whose characteristics are analyzed in Figs. 9 (I), 10 (II), and 11 (III). The 194 400 km2 boxed region is that used for the precipitation zone's moisture budget in Fig. 5. Tick marks indicate 12-hourly positions, beginning with the initial time

Citation: Monthly Weather Review 129, 12; 10.1175/1520-0493(2001)129<2983:TISAOA>2.0.CO;2

Fig. 9.
Fig. 9.

Diagnoses of parcel I (shown in Figs. 6a, 8) each 6 h for (a) pressure (hPa) and equivalent potential temperature (K), (b) mixing ratio q (g kg−1), and the summation of evaporation minus precipitation (mm day−1), and (c) instantaneous EP (mm day−1) and latent heat flux (W m−2). Beginning (end) point of the trajectory are indicated with A (B) or A′ (B′), as appropriate. The beginning and end points correspond to those shown for parcel I in Fig. 8. The abscissa (km) shows the distance of each parcel from Montreal (location shown in Fig. 1)

Citation: Monthly Weather Review 129, 12; 10.1175/1520-0493(2001)129<2983:TISAOA>2.0.CO;2

Fig. 10.
Fig. 10.

As in Fig. 9 except for parcel II (shown in Figs. 6b, 8)

Citation: Monthly Weather Review 129, 12; 10.1175/1520-0493(2001)129<2983:TISAOA>2.0.CO;2

Fig. 11.
Fig. 11.

As in Fig. 9 except for parcel III (shown in Figs. 6b, 8)

Citation: Monthly Weather Review 129, 12; 10.1175/1520-0493(2001)129<2983:TISAOA>2.0.CO;2

Fig. 12.
Fig. 12.

Anomaly correlations of (a) sea level pressure, (b) 1000–925-hPa thickness, and (c) 1000–500-hPa thickness, plotted each 12 h, from 0000 UTC 5 Jan through 0000 UTC 9 Jan 1998, for the region 20°–60°N and 10°–130°W, as in Fig. 4. The five best analog anomaly correlations are plotted (year after 1900 indicated in the legend) on the same timescale as in the 1998 ice storm case

Citation: Monthly Weather Review 129, 12; 10.1175/1520-0493(2001)129<2983:TISAOA>2.0.CO;2

Fig. 13.
Fig. 13.

Time-mean anomalies of 1000–500-hPa thickness [heavy contours with interval of 60 m with solid (dashed) for positive (negative) values] for the (a) 1966, (b) 1967, (c) 1974, and (d) 1991 analog cases

Citation: Monthly Weather Review 129, 12; 10.1175/1520-0493(2001)129<2983:TISAOA>2.0.CO;2

Fig. 14.
Fig. 14.

Time-mean anomalies of sea level pressure [heavy contours with interval of 4 hPa and solid (dashed) for positive (negative) values] for the (a) 1966, (b) 1967, (c) 1974, and (d) 1991 analog cases

Citation: Monthly Weather Review 129, 12; 10.1175/1520-0493(2001)129<2983:TISAOA>2.0.CO;2

Fig. 15.
Fig. 15.

Time-mean anomalies of 1000–925-hPa thickness (heavy contours with interval of 7 m with solid (dashed) for positive negative values] for the (a) 1966, (b) 1967, (c) 1974, and (d) 1991 analog cases

Citation: Monthly Weather Review 129, 12; 10.1175/1520-0493(2001)129<2983:TISAOA>2.0.CO;2

Fig. 16.
Fig. 16.

Precipitation analysis for the 5-day period ending at 1200 UTC 27 Jan 1967, with solid contour interval of 10 mm, beginning with 50 mm. Dotted contours show the 20-, 30-, and 40-mm contours. Heavy solid (dashed) contour is that of the 0°C isotherm at 850 hPa (surface) for 1200 UTC 23 Jan 1967. Very heavy solid (dashed) contour is that of the 0°C isotherm at 850 hPa (surface) for 1200 UTC 25 Jan 1967

Citation: Monthly Weather Review 129, 12; 10.1175/1520-0493(2001)129<2983:TISAOA>2.0.CO;2

Fig. 17.
Fig. 17.

As in Fig. 5 except for the period from 0000 UTC 22 Jan through 1200 UTC 27 Jan 1967

Citation: Monthly Weather Review 129, 12; 10.1175/1520-0493(2001)129<2983:TISAOA>2.0.CO;2

Fig. 18.
Fig. 18.

Paths of three trajectories (IV, V, and VI) discussed in the text for the 1967 analog case, whose characteristics are analyzed in Figs. 19 (IV), 20 (V), and 21 (VI). Beginning times are shown as the Jan 1967 date/time. Ending times are shown as 23/12 for the arrowhead at the end of the westernmost trajectory, and as 25/18 for the two arrowheads at the ends of the two other trajectories. The 194 400 km2 boxed region is that used for the precipitation zone's moisture budget in Fig. 17. Tick marks indicate 12-hourly positions, beginning with the initial time

Citation: Monthly Weather Review 129, 12; 10.1175/1520-0493(2001)129<2983:TISAOA>2.0.CO;2

Fig. 19.
Fig. 19.

Diagnoses of parcel IV (shown in Fig. 18) each 6 h for (a) pressure (hPa) and equivalent potential temperature (K); (b) mixing ratio q (g kg−1), and the summation of evaporation minus precipitation (mm day−1); and (c) instantaneous EP (mm day−1) and latent heat flux (W m−2). Beginning (end) point of the trajectory is indicated with A (B) or A′ (B′), as appropriate. The beginning and end points correspond to those shown for parcel IV in Fig. 18. The abscissa (km) shows the distance of each parcel from Montreal (location shown in Fig. 1)

Citation: Monthly Weather Review 129, 12; 10.1175/1520-0493(2001)129<2983:TISAOA>2.0.CO;2

Fig. 20.
Fig. 20.

As in Fig. 19 except for parcel V (shown in Fig. 18)

Citation: Monthly Weather Review 129, 12; 10.1175/1520-0493(2001)129<2983:TISAOA>2.0.CO;2

Fig. 21.
Fig. 21.

As in Fig. 19 except for parcel VI (shown in Fig. 18)

Citation: Monthly Weather Review 129, 12; 10.1175/1520-0493(2001)129<2983:TISAOA>2.0.CO;2

Fig. 22.
Fig. 22.

Precipitable water (mm) for (a) 0000 UTC 6 Jan 1998, (b) 0000 UTC 9 Jan 1998, (c) 1200 UTC 23 Jan 1967, and (d) 1200 UTC 25 Jan 1967

Citation: Monthly Weather Review 129, 12; 10.1175/1520-0493(2001)129<2983:TISAOA>2.0.CO;2

Table 1.

List of analog periods discussed in the text

Table 1.
Save