Airborne Doppler Observations of a Landfalling Cold Front Upstream of Steep Coastal Orography

Cheng-Ku Yu Department of Atmospheric Sciences, University of Washington, Seattle, Washington

Search for other papers by Cheng-Ku Yu in
Current site
Google Scholar
PubMed
Close
and
Bradley F. Smull NOAA/National Severe Storms Laboratory, and Department of Atmospheric Sciences, University of Washington, Seattle, Washington

Search for other papers by Bradley F. Smull in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

This study uses airborne Doppler radar observations to describe the mesoscale structure and evolution of a cold frontal system as it made landfall on the mountainous coast of Oregon and northern California on 1 December 1995 during the Coastal Observations and Simulations with Topography experiment. This section of coastline constitutes a steep, approximately two-dimensional north–south-oriented orographic barrier. The front exhibited a northeast–southwest orientation and thus intersected the axis of high terrain at an acute angle. The along-barrier pressure gradient and low-level winds increased with time along the coastal zone and reached a maximum as the front made landfall. Stably stratified prefrontal flow was strongly blocked by the orography, resulting in a confluent transition from pervasive southwesterly winds offshore to a narrow zone of accelerated south-southwesterly flow near the coast, where wind speeds approached 30 m s−1 at a height of 750 m above mean sea level. Postfrontal flow was much less affected by the topography, probably because of its weaker static stability. Upstream blocking by the steep coastal terrain also evidently led to modifications of precipitation in the vicinity of the front, including the rapid genesis of a narrow cold-frontal rainband (NCFR) and nearshore enhancement of two prefrontal precipitation bands. This evolution of the NCFR is interpreted in conjunction with changes in prefrontal vertical wind shear, which favored more upright convective ascent as the front neared shore and encountered accelerated along-barrier flow adjacent to the steep terrain. In addition, a statistical examination of observed radar reflectivity patterns shows that the intensity of frontal precipitation systematically decreased with upstream distance away from the orographic barrier.

Corresponding author address: Prof. B. F. Smull, University of Washington, JISAO, Box 354235, Seattle WA 98195-4235.

Email: smull@atmos.washington.edu

Abstract

This study uses airborne Doppler radar observations to describe the mesoscale structure and evolution of a cold frontal system as it made landfall on the mountainous coast of Oregon and northern California on 1 December 1995 during the Coastal Observations and Simulations with Topography experiment. This section of coastline constitutes a steep, approximately two-dimensional north–south-oriented orographic barrier. The front exhibited a northeast–southwest orientation and thus intersected the axis of high terrain at an acute angle. The along-barrier pressure gradient and low-level winds increased with time along the coastal zone and reached a maximum as the front made landfall. Stably stratified prefrontal flow was strongly blocked by the orography, resulting in a confluent transition from pervasive southwesterly winds offshore to a narrow zone of accelerated south-southwesterly flow near the coast, where wind speeds approached 30 m s−1 at a height of 750 m above mean sea level. Postfrontal flow was much less affected by the topography, probably because of its weaker static stability. Upstream blocking by the steep coastal terrain also evidently led to modifications of precipitation in the vicinity of the front, including the rapid genesis of a narrow cold-frontal rainband (NCFR) and nearshore enhancement of two prefrontal precipitation bands. This evolution of the NCFR is interpreted in conjunction with changes in prefrontal vertical wind shear, which favored more upright convective ascent as the front neared shore and encountered accelerated along-barrier flow adjacent to the steep terrain. In addition, a statistical examination of observed radar reflectivity patterns shows that the intensity of frontal precipitation systematically decreased with upstream distance away from the orographic barrier.

Corresponding author address: Prof. B. F. Smull, University of Washington, JISAO, Box 354235, Seattle WA 98195-4235.

Email: smull@atmos.washington.edu

1. Introduction

The west coast of the United States is frequently the target of cool-season extratropical cyclones originating over the Pacific Ocean. Interaction of these synoptic-scale disturbances with steeply rising coastal terrain such as that found in the Pacific Northwest can result in the development of high winds and heavy rainfall within the coastal zone. Nevertheless, the lack of observations in this region has led to considerable gaps in our knowledge of the severe weather conditions and underlying dynamical processes accompanying the orographic modification of landfalling frontal systems. To better document the mesoscale structure of landfalling frontal systems and improve understanding of their underlying dynamics, the COAST experiment [Coastal Observations and Simulations with Topography; Bond et al. (1997)] was conducted in late November–December of both 1993 and 1995. In this paper we present results from a detailed observational analysis of a landfalling cold front observed on 1 December 1995. We will show evidence that coastal terrain acted to significantly modify both the winds and the precipitation associated with this landfalling frontal system.

It has long been recognized that one of the most striking influences of mountains on the lower atmosphere is the “upstream blocking” effect. Deflection of low-level flow may occur upstream of mountain ranges having widths of a few tens to hundreds of kilometers (e.g., Chen and Smith 1987; Bell and Bosart 1988). The fundamental dynamics of upstream blocking have been explored in previous observational and modeling studies for the special case of relatively steady, horizontally uniform onshore flow approaching idealized terrain. For a steep, approximately two-dimensional barrier, one important result of upstream blocking is the development of a “barrier jet” marked by strong low-level winds parallel to the axis of high terrain (e.g., Parish 1982). Generation of orographically enhanced along-barrier flow has been described as resulting from downgradient acceleration caused by ageostrophic forcing in the along-barrier direction (Overland 1984; Lackmann and Overland 1989). Another important characteristic associated with the occurrence of orographic blocking is the upstream deceleration of the incident flow component normal to the barrier. Under suitably unstable environmental conditions, lifting caused by terrain-induced deceleration of low-level wind can play a role in the initiation of upstream deep convection (Grossman and Durran 1984).

Recent work has extended these idealized studies of orographic blocking in the presence of steady onshore flow to address terrain–atmosphere interactions relevant to landfalling frontal systems. Analyses of synoptic observations suggest that airflow in the vicinity of landfalling fronts may be altered substantially so as to produce strong surface winds immediately upstream of coastal mountains. For instance, through analysis of routine synoptic observations Mass and Ferber (1990) found that strengthening of coastal winds in advance of fronts approaching the Olympic Mountains can result from enhancement of the alongshore pressure gradient in conjunction with mesoscale pressure ridging upstream of the topography. Overland and Bond (1993) documented strong surface winds at the leading edge of a postfrontal pressure surge in an environment characterized by low Froude number along the mountainous coastline of southeastern Alaska. Doyle (1997) reported the existence of a prefrontal low-level wind maximum adjacent to steep terrain along the central California coast. Trier et al. (1990) investigated a subtropical cold front observed during the Taiwan Area Mesoscale Experiment (Kuo and Chen 1990) and noted blocking, as evinced by the deceleration of prefrontal flow and the splitting of the prefrontal and postfrontal flow around the mountainous island of Taiwan. Li and Chen (1998) conducted a more general investigation of conditions along the northwestern coast of Taiwan, and their results suggest that barrier jets can also occur at subtropical latitudes. Nevertheless, owing to the relatively coarse temporal and spatial resolution of the available observations, these previous studies have provided only a gross view of orographically modified winds in the vicinity of landfalling fronts.

Special measurements collected during COAST provide highly detailed views of cool-season mesoscale atmospheric structures near a variety of coastal terrain configurations in the Pacific Northwest (Fig. 1). Two early COAST studies described such structures in the presence of relatively steady onshore flow. Overland and Bond (1995) used flight-level observations from the National Oceanic and Atmospheric Administration’s (NOAA) P-3 aircraft to describe conditions upstream of the quasi-two-dimensional barrier presented by Vancouver Island, while Colle and Mass (1996) used the P-3’s scanning Doppler radar to construct a comprehensive view of flow splitting around the more three-dimensional Olympic Mountains. More recently, two additional COAST studies have considered the more complex situation in which inherently unsteady flow accompanying landfalling baroclinic systems interacts with coastal terrain. Braun et al. (1997) studied an intense cold frontal system oriented parallel to the mountain barrier as it advanced from a position ∼400 km offshore to within 20 km of the southern Oregon coast near Cape Blanco during the third intensive operations period (IOP 3) of COAST on 8 December 1993. Their analysis focused on the front’s offshore structure, but limited observations nearer shore showed that the low-level prefrontal airflow parallel to the coastal barrier increased while that perpendicular to the barrier decreased as the front neared the coast. Braun et al. provided theoretical evidence showing this trend to be consistent with upstream blocking by the coastal terrain. In another study utilizing airborne Doppler observations and simulations from version five of the Pennsylvania State University–National Center for Atmospheric Research Mesoscale Model, Colle et al. (1999) investigated complexities associated with the interaction of a cold front with the Olympic Mountains during COAST IOP 5. Based upon these simulations, which successfully reproduced many attributes of the airborne Doppler observations, deflection of the prefrontal southerly flow was shown to have enhanced stretching deformation and resulting mesoscale frontogenesis upstream of this relatively isolated, quasi-circular barrier.

The chief objective of the present study is to document the mesoscale structure and evolution of a landfalling cold frontal system as it encountered a relatively two-dimensional orographic barrier near the Oregon–California border on 1 December 1995, during COAST IOP 8. A distinct aspect of this event is that this segment of coastline is characterized by an exceptionally steep north–south-oriented barrier (the Coastal Range), along which terrain height rises to >1600 m above mean sea level (MSL) within 120 km of the shore (cf. Fig. 1). Furthermore, the cold front studied herein was oriented at an acute angle (∼45°) to the orographic barrier during landfall. On the basis of numerical simulations, Doyle (1997) posited that such a configuration favors the channeling of prefrontal airflow to form a barrier jet as the front nears shore. Nevertheless, this phenomenon has never been documented using detailed three-dimensional airflow observations. The sequence of 14 volumetric airborne dual-Doppler radar analyses collected in COAST IOP 8 provides a unique description of a landfalling cold front and its interaction with steep coastal terrain.

While the potential for terrain to influence frontal circulations and precipitation over the Pacific Northwest was recognized prior to COAST (e.g., Hobbs et al. 1975;Parsons and Hobbs 1983a), earlier studies lacked dual-Doppler radar observations capable of addressing conditions offshore, along the coast, and over the mountain slopes in a comprehensive fashion. The advent of airborne Doppler radar as employed in COAST allows observation of precipitation and three-dimensional flow over extensive regions so that terrain influences may be clearly revealed. This study is guided by several fundamental questions: Did a barrier jet develop on 1 December 1995, and if so, how did its strength vary with height and distance from the orographic barrier and approaching front? How did frontal precipitation change as a function of upstream distance from the barrier? Were these changes related to alterations of the frontal circulation near the coast, and if so, to what extent were they the result of upstream blocking or other as yet undocumented dynamical processes?

2. Data sources and analysis methods

The primary datasets used in this study were provided by the NOAA P-3 aircraft and include volumetric distributions of reflectivity and radial velocity from helically scanning X-band (3.2 cm) Doppler radar, horizontal maps of radar reflectivity derived from the horizontally scanning conventional C-band (5 cm) lower fuselage radar, and a full complement of in situ flight-level measurements (Jorgensen 1984). Other data sources used in this study (summarized in Fig. 1) include routine surface/buoy observations along the Pacific Northwest coast, the 915-MHz research wind profiler operated by NOAA’s Environmental Technology Laboratory (ETL) at Crescent City, California, and archived volume scans from the National Weather Service (NWS) operational Weather Surveillance Radar-1988 Doppler (WSR-88D) located on Cape Mendocino near Eureka, California.

During COAST, the tail-mounted Doppler radar employed an efficient scanning referred to as the Fore/Aft Scanning Technique (Hildebrand 1989; Jorgensen and Smull 1993), which allows multiple Doppler measurements to be collected continuously on both sides of the aircraft as it moves along a straight flight path. As a result, the time required to gather two independent radial velocity measurements at a given point is considerably reduced. In this study, three-dimensional airflow in the frontal region was obtained by use of a pseudo-dual-Doppler synthesis technique described by Jorgensen and Smull (1993) and Jorgensen et al. (1994). Prior to interpolation to Cartesian coordinates, folded radial velocities and contamination due to sea/mountain clutter and receiver noise were corrected interactively. Radial velocities were also adjusted to remove contributions due to the aircraft’s motion. During COAST’s 1995 field phase, availability of Global Positioning System navigation aboard the P-3 allowed for optimal determination of the aircraft’s position and velocity following Matejka and Lewis (1997). The horizontal and vertical analysis grid spacing were set to 1.5 km and 0.25 km, respectively, over multiple volumes encompassing 135 × 135 km2 in the horizontal and 8.5 km in the vertical, with the lowest analysis level located at 0.25 km MSL. Vertical air motions were obtained through an iterative procedure by assuming anelastic continuity through downward integration from an upper-boundary condition of zero vertical motion at echo top. A variational adjustment scheme described by O’Brien (1970) was applied to compensate for divergence errors that produce nonzero vertical motions at the lower boundary. To mitigate the effects of X-band attenuation, radar reflectivity was taken to be the maximum observed value measured from two independent radar beams for each point in the Cartesian analysis grid. Additionally, prior to calculation and vertical integration of divergence, the horizontal wind field was smoothed with a two-pass Leise filter (Leise 1981) to reduce noise in the horizontal divergence and derived vertical velocities. As a result, scales of motion having horizontal wavelengths less than ∼6 km are strongly damped (Jorgensen et al. 1997).

3. Case overview

Intensive observations by the NOAA P-3 were conducted in the coastal zone near the Oregon–California border between 0500 and 1000 UTC on 1 December 1995. The 850-mb analysis from the National Centers for Environmental Prediction (NCEP) at 0000 UTC 1 December (Fig. 2) shows a baroclinic trough over the northeast Pacific Ocean near 135°W. Wind speeds at 850 mb along the coast near the Oregon–California border were ∼10 m s−1, and the geopotential gradient in this region was nearly parallel to the coast. Overland (1984) determined that this type of large-scale pattern, in which the pressure gradient force is directed primarily parallel to the barrier, is favorable for the development of strong winds adjacent to steep coastal terrain.

The sea level pressure analysis valid at about the time that the aircraft began to collect dual-Doppler observations along the southern Oregon coast is shown in Fig. 3. The northern portion of the surface front was located immediately off the southern Oregon coast and moved southeastward at a speed of ∼11 m s−1 (based on dual-Doppler observations). An elongated pressure trough extended north-northwest of the front off the coast of Washington and Oregon in association with a midlevel short-wave disturbance. Infrared satellite imagery indicated a sharpening southwest–northeast-oriented baroclinic cloud band associated with the frontal system (cf. Fig. 12 of Bond et al. 1997). Mesoscale pressure ridging analogous to that shown by previous observational (e.g., Mass and Ferber 1990) and numerical (e.g., Doyle 1997) studies developed in the warm sector adjacent to the zone of steepest coastal terrain. Surface winds within this zone were approximately parallel to the coast and highly ageostrophic, in contrast to more geostrophic prefrontal southwesterlies farther offshore.

The horizontal structure of precipitation observed from the P-3’s lower fuselage radar during the aircraft observations is summarized in Fig. 4. Reflectivity values over land are not shown because of pervasive ground clutter contamination. The position of the surface front corresponding to the wind shift zone seen in dual-Doppler observations (to be shown) is also indicated in Fig. 4. Prior to its landfall at 0525 UTC 1 December (Fig. 4a), banded precipitation exhibiting a southwest–northeast orientation was present ahead of the surface front and associated radar reflectivities were generally less than 35–40 dBZ. About 1 h later, around 0640 UTC (Fig. 4b), the composite radar reflectivity showed a rapid intensification of precipitation as a narrow precipitation band characterized by strong reflectivity values (>40 dBZ) formed adjacent to the coast along the leading edge of the surface front. A separate north–south-oriented narrow prefrontal precipitation band (hereafter rainband A) developed 20–30 km ahead of the front. With time, postfrontal precipitation became more evident in the form of a broad band that approximately paralleled the surface front (Fig. 4c). While precipitation along the surface front generally weakened during this later period, another mesoscale precipitation area (rainband B) developed ahead of the front and exhibited an orientation nearly parallel to the coastline.

The impact of nearby steep terrain on coastal weather conditions can be readily seen from a time section of objectively analyzed surface conditions encompassing the period of frontal landfall (Fig. 5). This analysis clearly shows passage of a surface pressure trough, extending from Astoria, Oregon (AST), to near Cape Mendocino, California (buoy 46030). Little if any surface baroclinity was detected poleward of Northbend, Oregon (OTH), and hence no frontal boundary is analyzed in that region. Farther south, pressures rose sharply and temperatures dropped by 2°–3°C in the wake of the analyzed cold front. The strongest sustained surface winds (>10 m s−1, denoted by shading in Fig. 5), which coincided with an enhanced along-barrier pressure gradient, were observed within the coastal zone between buoys 46027 and 46030 just prior to frontal passage. While these observations are of limited resolution, the correspondence of this wind maximum and associated pressure perturbation to the zone of most steeply rising terrain (cf. right panel of Fig. 5) is strongly suggestive of upstream blocking by the coastal orography.

Special datasets provided by COAST fill critical gaps in this synoptic analysis. For example, hourly averaged observations from the 915-MHz wind profiler at Crescent City (CRC, located adjacent to the highest coastal terrain and ∼10 km southeast of buoy 46027; cf. Fig. 1) shown in Fig. 6 resolve the rapid frontal wind shift from southerlies to westerlies that occurred at ∼0800 UTC. The low-level along-barrier (i.e., southerly) flow increased with the approach of the frontal system, and reached its maximum value (∼25 m s−1) in the lowest 600 m prior to frontal passage, between 0600 and 0700 UTC. A second, distinct prefrontal wind maximum was detected at a height of ∼1500 m and likely represents a baroclinically induced low-level jet, a feature commonly observed ahead of fronts within the 900–850-mb layer (e.g., Browning and Pardoe 1973; Hobbs et al. 1980). In the front’s wake, low-level winds rapidly weakened to ∼10 m s−1 and veered to a more northerly direction.

Thermodynamic profiles located ∼30 km to either side of the surface boundary were constructed using a combination of dropsonde, flight-level, and buoy data along the coastal zone near the Oregon–California border (Fig. 7). The prefrontal airmass exhibited comparatively strong stratification in the lowest kilometer, with convectively stable-to-neutral conditions (equivalent potential temperature increasing or remaining constant with height) below 1.5 km MSL. In the postfrontal zone, a stronger degree of stratification in the 1–2-km layer likely corresponds to the elevated frontal zone, with convectively unstable conditions confined to a shallow layer immediately above it.

4. Airborne Doppler radar observations

a. Coevolving airflow and precipitation in the coastal zone

A sequence of P-3 dual-Doppler radar analyses constructed over the period 0500–1000 UTC 1 December 1995 provides a uniquely detailed description of three-dimensional kinematic and precipitation fields as the cold front encountered the zone of most steeply rising coastal terrain near the Oregon–California border. Figure 8 shows winds and precipitation at a height of 0.75 km (all heights MSL) at 0515 UTC when the surface front was ∼70 km north of the border (i.e., located near the northwest corner of the analysis domain). This level was close to the height of the maximum along-barrier flow determined by the Doppler radar. As evident in Fig. 4a, the heaviest precipitation at this early stage was prefrontal, with some suggestion of bands oriented roughly parallel to the front (Fig. 8a). No evidence of the rapid shift to west-northwesterly postfrontal winds identified at later times in the surface and wind profiler observations (cf. Figs. 5, 6) is seen at this early stage. Winds immediately ahead of the front were southwesterly (Fig. 8a) but markedly strengthened and backed with distance toward the south. The strongest Doppler-observed wind speeds at this time were 22–24 m s−1 at a location 40–50 km south of the front (Fig. 8b), where the flow was predominantly parallel to the coastal barrier (Fig. 8d). The position of the confluent flow transition from moderate southwesterlies to strong south-southwesterlies seen in Fig. 8a is consistent with the northernmost extent of the strong surface winds that developed as the frontal system approached the coast (cf. Fig. 5).

As the cold front progressed southeastward, its associated convergence strengthened and supported the rapid development of a narrow but intense rainband coincident with the low-level frontal wind shift zone, as seen in the 0642 UTC airborne Doppler analysis (Fig. 9). Wind speeds in the prefrontal region had strengthened compared to those found 90 min earlier, but retained the confluent transition from southwesterly flow offshore to south-southwesterlies near the coast. Maximum wind speeds of ∼26–28 m s−1 (cf. isotachs in Fig. 9b) were confined to a narrow band ∼20 km in width, whose long axis was nearly coincident with the coastline. Once again this zone of strongest winds was dominated by along-barrier flow as incident southwesterlies were deflected so as to reduce the cross-barrier component in a manner consistent with upstream blocking (Figs. 9c,d). A vertical section of cross-barrier (U) and along-barrier (V) velocities in advance of the front (Fig. 10) shows along-barrier flow below 1 km MSL increasing from ∼20 m s−1 at a position 20 km offshore to ∼28 m s−1 at the coast, with a concomitant reduction in cross-barrier flow. Above the height of the coastal mountains (which extend to ∼1500 m in this region), the wind pattern was relatively unperturbed and along-barrier velocities were weaker. These nearshore airflow structures resemble those of barrier jets observed along other approximately two-dimensional mountain ranges (e.g., Overland and Bond 1995; Parish 1982).

At this stage the mesoscale precipitation pattern exhibited two modes of linear organization. One was oriented approximately parallel to the coast in the prefrontal region, as represented by rainband A in Fig. 9b. It was apparently associated with an orographically induced deflection of the prefrontal flow, and coincided with a pronounced deceleration of cross-barrier flow within ∼15 km of the shore (Fig. 9c). The vertical section in Fig. 10 indicates this band (near X = 28 km) was relatively deep, with the 30-dBZ contour reaching above 3 km. This band extended southward to a position about 60 km south of the Oregon–California border (cf. Fig. 4b). The second distinct mode of organization evident in Fig. 9 is a narrow but intense precipitation band coincident with the frontal wind shift zone (i.e., collocated with tightly packed isotachs in Fig. 9b) and is thus interpreted to be a narrow cold-frontal rainband [NCFR; Houze (1993, chapter 11)]. The NCFR was marked by multiple centers of very heavy precipitation (maximum radar reflectivities >45 dBZ). These discrete cores of heavy precipitation and intervening regions of lighter precipitation along the NCFR (often referred to as precipitation cores and gap regions) are similar to those previously observed in midlatitude cyclones (Hobbs and Biswas 1979; James and Browning 1979; Hobbs and Persson 1982). This characteristic structure persisted over the next hour. The lower fuselage radar also showed decreasing precipitation intensity along the NCFR with distance offshore (Fig. 4b).

The sequence of vertical sections in Fig. 11 captures the development of the NCFR. The movement of the front in the plane of the section has been subtracted to show airflow relative to the front. Corresponding divergence fields are shown in Fig. 12. At 0515 UTC, when the surface front was ∼35 km offshore as viewed in the plane of the section, most of the enhanced precipitation was located ahead of the front (Fig. 11a). Though limited in their near-surface coverage by range effects, Doppler measurements showed precipitation and convergence at the front to be weak at this time (Figs. 11a, 12a). About 35 min later (Fig. 11b), as the surface front approached the coast, the strength of low-level convergence associated with the front increased (Fig. 12b). Stratiform precipitation ahead of the front appeared to be enhanced at this time as well, with reflectivity fallstreaks reaching 35 dBZ extending from melting level (∼2.5 km MSL) to the surface. By 0642 UTC (Fig. 11c), a well-defined NCFR was observed coincident with the surface front, accompanied by strong low-level convergence reaching a magnitude of ∼2.4 × 10−3 s−1 (Fig. 12c). The vertical extent of the NCFR was limited (∼2 km MSL); however, its maximum radar reflectivity was >45 dBZ and Doppler-derived updraft velocities >2 m s−1 were observed.

As the frontal system continued its southeastward advance, dual-Doppler observations at 0843 UTC (Fig. 13) showed the same general patterns of organization but with several important differences. The strength of prefrontal southwesterlies (∼22 m s−1) and along-barrier flow (Figs. 13b,d) had lessened, and the deceleration of cross-barrier flow near the coast was not as marked as that observed at 0642 UTC (cf. Fig. 9c). Perhaps the most significant change was the absence of the NCFR at this time, marking a ⩽2 h lifetime for this feature. Following the weakening of rainband A, a second prefrontal rainband (B in Figs. 13a and 4c) had developed approximately parallel to the coast, coincident with the zone of enhanced along-barrier velocities and the decelerating cross-barrier flow (Figs. 13c,d).

b. Comparison with idealized theory

Past theoretical studies have shown that low-level flow encountering terrain is generally blocked when the upstream Froude number (Fr)1 is less than unity (Smith 1979). Nevertheless, the flow response is complex and may not be uniquely determined by the ratio of the kinetic to the potential energy as expressed in Fr (e.g., Schumacher et al. 1996). In the rotating limit (i.e., when Coriolis effects are considered), Pierrehumbert (1984) and Pierrehumbert and Wyman (1985) have shown that for an idealized two-dimensional barrier the Burger number (B)2 is also an important parameter in determining the potential for upstream blocking. For situations having comparable latitude and stratification, the Burger number is effectively proportional to the steepness of the mountain slope (i.e., H/L, where H is the mountain height and L is the mountain half-width). When B is less than unity, the flow is quasigeostrophic/semigeostrophic as the flow proceeds over the mountain;however, when B exceeds unity, the mountain is said to be “hydrodynamically steep” and flow is diverted by the barrier (Overland and Bond 1995). In the latter situation (B > 1), Pierrehumbert and Wyman (1985) have shown that the terrain influence attains a maximum upstream extent on the order of the Rossby radius of deformation (lR = NH/f, where N is the Brunt–Väisälä frequency, H is the mountain height, and f is Coriolis parameter).

For the case studied here, where the volume of interest was generally filled with precipitation below mountaintop level, static stability is appropriately calculated using the expression for the saturated Brunt–Väisälä frequency (Nm) derived by Durran and Klemp (1982). Using the soundings in Fig. 7, this gives Nm = 1.2 × 10−2 s−1 and 4.8 × 10−3 s−1 in the prefrontal and postfrontal regions, respectively. If we then take H ≈ 1400 m (averaged along the coastal barrier); U ≈ 14 m s−1 and 12 m s−1 in the prefrontal and postfrontal regions, respectively; and L ≈ 120 km (mountain half-width approximated by the extent of windward slope; cf. Fig. 1), we obtain Fr ≈ 0.8, B ≈ 1.5, and lR ≈ 170 km in the prefrontal region, with Fr ≈ 1.8, B ≈ 0.6, and lR ≈ 65 km in the postfrontal air mass. These values appear consistent with dual-Doppler observations showing considerable upstream blocking in the prefrontal region (e.g., Fig. 9), yet little evidence of terrain-modified flow in the front’s wake, consistent with the lower static stability in the cool air mass (cf. Fig. 7).

Moreover, dual-Doppler radar observations indicate that a profound wind shift suggestive of the terrain-blocked flow was located ∼20 km upstream of the coastline (e.g., Fig. 9), a distance much less than the estimated Rossby radius (∼170 km). This apparent inconsistency may imply limitations of the theory, or at least its application in realistic situations. For example, classical theory does not allow for modification of upstream static stability by moist diabatic processes (i.e., latent heat release and/or evaporation cooling in association with precipitation), as almost certainly occurred in this case. In addition, the topography in the COAST study region (cf. Fig. 1) is not characterized by an isolated two-dimensional coastal barrier, but rather includes even higher terrain (viz., the Cascade Range and Sierra Nevada, and ultimately the Rocky Mountains) farther inland. Although it is difficult to deduce the impact of this more distant orography using idealized theory and observations alone, recent numerical work by Braun et al. (1999) points to the potential significance of continent-scale terrain in modifying flow within the coastal zone. Finally, previous studies have quantified the influence of terrain under highly idealized conditions of steady, horizontally uniform incident flow. The inherently unsteady and nonuniform nature of onshore flow associated with a landfalling front represents a significant complicating factor capable of influencing the degree and upstream extent of observed flow blocking.

5. Ground-based Doppler radar observations

The National Weather Service WSR-88D S-band Doppler radar on Cape Mendocino (EKA in Fig. 1) provides a complementary view of the landfalling frontal system, offering better temporal continuity than can be achieved using airborne Doppler analyses alone. However, the altitude of the Eureka radar antenna (766 m MSL) and the lowest elevation angle (0.5°) limit the degree to which near-surface radial velocity and precipitation patterns can be observed, particularly at far ranges where the P-3 began its investigation of the landfalling front. The height of the lowest available radar beam reaches ∼3.5 km MSL near the Oregon–California border. To compensate, we apply a procedure to each radar volume in which the maximum value of observed radar reflectivity at any height in a vertical column is projected onto a horizontal map.3 Though susceptible to other problems including brightband contamination, these low-level composite reflectivity maps allow continuous tracking of organized precipitation features over long periods. Moreover, by summing these individual maps, it is possible to obtain a qualitative view of trends and extremes in precipitation intensity that occurred as the front made landfall.

Over the period of intensive P-3 observations (∼0500–1100 UTC), the front advanced from the Oregon–California border to a point near the Eureka radar site. During this interval, the nearby steep coastal terrain evidently affected the upstream distribution of total precipitation (Fig. 14). To compensate for range-induced bias in recorded reflectivity values as the frontal precipitation shield progressed toward the radar, reflectivity values at each grid point within a composite reflectivity map have been normalized by the domain-mean value before summing each map in the series. As such, the scaling of “cumulative normalized reflectivity” magnitudes derived in this manner is arbitrary; it is their horizontal pattern that conveys important information. Figure 14 shows that the heaviest and/or most persistent precipitation occurred within ∼50 km of the coast, with a marked reduction of precipitation farther offshore. Peak cumulative reflectivity values were generally found on the windward slopes of the coastal range (some 20–50 km inland) along a segment of the coastline between ∼20 and ∼120 km south of the Oregon–California border, adjacent to the highest/steepest mountains within range of the Eureka radar (cf. Fig. 5). Local maxima in the cumulative reflectivity pattern were found over hills as small as 500 m MSL in peak elevation [reminiscent of findings by Browning et al. (1974)], although the centers of heaviest radar-indicated precipitation accumulation occurred at higher elevations, being more nearly collocated with mountain crests in the range 1000–1500 m MSL. These enhanced reflectivities were not the result of ground-clutter contamination, since mountain heights over this zone were generally well below the lowest radar beam. Ground clutter contamination did produce a narrow band of intense reflectivities within a narrow northwest–southeast-oriented zone ∼50 km east-northeast of the radar site, where the beam was blocked by locally high terrain as evidenced by attenuated values at longer ranges in this sector.

The sequence of individual radar reflectivity maps in Fig. 15 shows that rainband A was initially observed ∼80 km offshore and then moved toward the coast at ∼16 m s−1. Its alignment was ∼10°–190°, and thus distinct from the northeast–southwest orientation of the NCFR. Moreover, precipitation intensity within rainband A strengthened as the band approached the zone of blocked flow near the coast (Figs. 15c,d; Fig. 9). During the following hour, similar evolution was observed for rainband B (not shown). Owing to the absence of dual-Doppler observations far offshore, the formative mechanism of these prefrontal bands remains uncertain. Nonetheless, their distinct alignment and intensification as they neared shore suggest that these prefrontal rainbands were at least partly the result of orographic forcing.

Figure 16 shows the 0929 UTC color display of radar reflectivity and radial velocity at 0.5° elevation from the Eureka radar. The position of the front at this time was ∼20 km northwest of the radar site, as shown by an ill-defined discontinuity in the Doppler velocity field (highlighted in Fig. 16b). The low-level radar reflectivity associated with the surface front was relatively weak (<25–30 dBZ). Somewhat heavier precipitation was detected both behind and ahead of the front. A series of maps encompassing the period when the front passed over the Eureka radar shows no evidence of any narrow, intense precipitation feature or sharp wind shift near Cape Mendocino. It thus appears that the NCFR and associated sharpening of the frontal wind shift existed only during a limited interval as the front made landfall along the segment of particularly steep coastal terrain near the Oregon–California border.

6. Upstream influence of the steep coastal terrain

a. Modulation of airflow

Influences of coastal orography upon the 1 December 1995 landfalling frontal system may be identified and succinctly summarized in a statistical sense by constructing and interpreting CFAD analyses [contoured frequency by altitude diagram; Yuter and Houze (1995)] for selected dual-Doppler analysis times. Each CFAD represents a joint probability distribution plot showing the frequency of occurrence of a selected parameter as a function of some independent variable (generally height) within a radar-observed volume. These frequencies are normalized by the number of observations at each level. When processed in this manner, the data more readily evince preferred modalities and subtle yet systematic shifts in observed conditions that shed light on underlying dynamic processes.

Attention is focused upon distributions of kinematic quantities during the period when the circulation and precipitation patterns accompanying the landfalling front were undergoing rapid change. This is accomplished by processing the sequence of airborne Doppler analyses valid at 0515, 0550, and 0642 UTC, corresponding to the set of vertical cross sections shown in Figs. 11 and 12. Though well positioned in time, CFADs derived from this sequence must be interpreted carefully owing to rapidly evolving radar echo coverage relative to the front. Sampled conditions within the 0515 UTC radar volume are weighted strongly toward the prefrontal regime owing to limited radar returns along/behind the front (cf. Fig. 8a), while the 0550 and 0642 UTC volumes encompass both prefrontal and postfrontal conditions (Fig. 9a).

Sequences of CFADs for the cross- and along-front components of horizontal flow are shown in Figs. 17 and 18, respectively. In this rotated coordinate system, positive values of u′ denote cross-front flow toward the warm side of the front. At 0515 UTC (Fig. 17a), the distribution of cross-front velocities was broad at low levels but appreciably narrowed near the 4-km level. Apart from the lowest 2 km of the sampled (mainly prefrontal) volume, cross-front velocities generally increased with height. At later times, significant changes occurred at low levels as the frontal circulation/precipitation pattern developed and advanced into the analysis domain. A bimodal structure developed within the lowest 1.5 km (Figs. 17b,c) indicating a growing contrast between the prefrontal south-southwesterly flow (negative u′) and postfrontal westerlies (positive u′). In particular, the modal value of cross-front velocities in the prefrontal zone shifted from near 0 to about −8 m s−1, signifying an increase in low-level inflow toward the front. Above ∼1.5 km (corresponding to the mean height of the coastal terrain) the sequence shows relatively steady conditions. An analogous transition was apparent in CFADs of along-front velocities within the 0–1.5-km layer occupied by the coastal barrier (Fig. 18). In particular, the modal value of prefrontal along-front velocities shifted from ∼18 m s−1 at 0515 UTC to ∼25 m s−1 at 0642 UTC. The general increase of along-front flow with height (particularly at higher levels) is consistent with the presence of large-scale baroclinity in the frontal zone. In contrast to the marked acceleration of the prefrontal airflow at low levels, the portion of the velocity distribution (either cross- or along-frontal) corresponding to the postfrontal regime changed little during landfall (Figs. 17 and 18).

Of particular dynamical significance is that the observed strengthening of low-level prefrontal winds was focused neither in the cross- nor along-front component, but rather resulted from the acceleration of flow parallel to the coastal orographic barrier (cf. along-barrier speeds of 18–20 m s−1 in Fig. 8d vs 24–26 m s−1 in Fig. 9d). This is further evidence that upstream blocking of stably stratified prefrontal flow was responsible for the observed rapid increase in low-level wind speeds adjacent to the steep orographic barrier.

b. Enhancement of precipitation

Another important question concerns how frontal precipitation may have been modified as a result of upstream blocking. Modulation of low-level precipitation intensity may be examined by means of CFAD-type joint probability plots in which statistical distributions of radar reflectivity are displayed not according to altitude, but rather as a function of offshore distance (Fig. 19). This analysis is confined to reflectivity measurements in the layer ⩽1.5 km MSL, where airborne Doppler kinematic measurements indicated the effects of upstream blocking were most pronounced.

Figure 19 highlights several interesting aspects of the low-level radar reflectivity pattern that persisted over a ∼3 h period spanned by a sequence of three Doppler analysis volumes (panels a–c) collected adjacent to steep coastal terrain as the front made landfall. The largest radar reflectivities within the analysis volume were consistently located near the coast, with a systematic decrease of reflectivity with distance offshore. Mean radar reflectivity values fell 30%–40% between the coast and a distance 40 km offshore, consistent with the pattern seen by the Eureka radar (cf. Fig. 14). None of the airborne Doppler radar volumes summarized in Fig. 19 extended inland over the steep terrain slopes, where the Eureka radar indicated heaviest accumulations of precipitation (as discussed in section 5). Since the aircraft’s position was ∼15 km offshore for each of these volumes, the indicated decrease of reflectivity with distance inland may be at least partly due to attenuation by intervening heavy precipitation near the coast. At heights above 1.5 km (not shown), the spectrum of observed reflectivities as a function of offshore distance broadened, consistent with the presence of elevated convective elements, and the tendency for a systematic offshore decrease in echo intensity was not so pronounced.

The observed decrease in precipitation rates away from the coast is consistent with variations in Doppler-derived divergence in this layer averaged in the along-coast direction and plotted as a function of offshore distance (Fig. 20). The strongest mean low-level convergence was consistently located near the coast, decreasing to a mere 20% of its coastal value just 20 km offshore. Evidently this enhanced low-level convergence contributed to production of precipitation in the coastal zone. To gain further insight into these systematic changes, the relationship of low-level wind and divergence were further examined via detailed horizontal maps. Figure 21 shows the spatial distribution of divergence (corresponding to winds plotted in Fig. 13a) that contributed to this coastal convergence maximum. Observed convergence was concentrated in an elongated zone within ∼10 km of the coast (Fig. 21a). There were two distinct kinematic sources contributing to convergence along this zone. One was the decreased speed of cross-barrier flow in the cross-barrier direction (∂u/∂x) in the prefrontal region (Fig. 21b). This convergence was distributed rather uniformly along the stretch of coastline south of the front, as would be expected if upstream blocking of the cross-barrier prefrontal flow component was the responsible mechanism. A second, more localized contribution to the coastal convergence maximum noted in Fig. 20 occurred near the intersection of the frontal wind shift zone and the coastline. This center owed its existence to rapid deceleration of the along-barrier component (∂υ/∂y; Fig. 21c) as northward-deflected prefrontal flow collided with the cool postfrontal air mass. Thus, spatially distinct variations of both along-barrier and cross-barrier flow that occurred in response to terrain-induced blocking of the prefrontal flow apparently contributed to enhanced convergence and precipitation near the coast.

c. Formation of the NCFR and associated vertical circulation

In this section we seek a possible dynamical connection between the formation of the NCFR on 1 December 1995 and the upstream influence of steep coastal terrain. While a large number of past studies have investigated the finescale structure of the NCFR, relatively few have accessed detailed observations from which topographic influences upon cold-frontal structure might be discerned. Using radar reflectivity measurements obtained during the Cyclonic Extratropical Storms project Hobbs et al. (1980) and Parsons and Hobbs (1983a) identified and tracked various types of rainbands in four Pacific cyclones as they approached the relatively complex, three-dimensional terrain of the Olympic Mountains in western Washington (cf. Fig. 1). Among these cases were two cold fronts, both of which were oriented nearly parallel to the western Washington coast during landfall. Parsons and Hobbs noted a slight decrease in the intensity of both NCFRs following landfall and speculated this might be due to land–sea contrasts involving disruption of boundary layer flow over rough terrain; more significant changes in NCFR behavior (mainly pertaining to rainband orientation) were observed downstream of the Olympics over Puget Sound. More recently, Colle et al. (1999) exploited airborne Doppler observations of a landfalling cold front on 11 December 1993 during COAST, but once again for the case of a north–south-oriented front encountering the Olympics. Their analysis necessarily focused on changes in frontal intensity as the cold front ascended (and ultimately split around) this quasi-circular barrier. These previous authors could thus not address changes in NCFR structure that might occur as a cold front originating over the open ocean encounters a barrier jet adjacent to more two-dimensional steep terrain.

To our knowledge, only one other study has considered the interaction of a cold front with a relatively steep, quasi-two-dimensional barrier in such detail. Braun et al. (1997) also used airborne Doppler radar to investigate an intense frontal system approaching southern Oregon on 8 December 1993 during the first phase of COAST, and documented a marked decrease in the depth of the radar echo and magnitude of the Doppler-derived vertical velocities within the NCFR as it neared shore. However, the orientation of that front, which was more nearly parallel to the coast, apparently combined with a lack of significant onshore flow to preclude formation of a strong barrier jet in their case. Owing to the timing of the P-3 flight upon which their study was based, Braun et al.’s analysis ended at a point when the front was still ∼20 km offshore and, thus, did not encompass the front’s actual landfall.

Many previous observational studies (e.g., Carbone 1982; Hobbs and Persson 1982; Parsons and Hobbs 1983b; Shapiro 1984; Roux et al. 1993) have noted similarities between the leading edge of some cold fronts and gravity currents (Simpson 1969). Individual scan-plane plots of Doppler radial velocities from this case (not shown) clearly depict gravity current–like structure as the leading edge of the front came ashore. Provided that gravity current dynamics are indeed applicable, the convection at the leading edge of the cold front can be viewed as a response to low-level convergence generated as advancing cold air forces the ascent of relatively warm, moist air originating ahead of the front. This mechanism provides a reasonable explanation for the development of NCRF structure even in the absence of appreciable potential instability in the prefrontal environment (Carbone 1982; Parsons 1992). When retardation by opposing prefrontal flow is considered, Simpson and Britter (1980) showed that the front’s velocity can be approximated by the expression
i1520-0493-128-6-1577-e1
where g is the gravitational acceleration, h is the depth of the cold air; Tυ1 and Tυ2 are the warm- and cool-airmass virtual temperatures, respectively; u0 is the prefrontal cross-front wind component (negative for a wind toward the cold air); and k and b are constants equal to 0.9 and 0.6, respectively. For the present case, Tυ1 = 288.7 K and Tυ2 = 286.7 K (average values observed at buoy 46027; cf. Fig. 1), and Doppler-derived flow fields indicate the depth of the cold air to be ∼2000 m (cf. Fig. 11c). The mean u0 over a depth corresponding to that of the cool air mass was −2.4 m s−1. Substituting these values into (1) yields u′ = 9.1 m s−1. The observed absolute forward motion of the front was ∼11 m s−1, though slowing to ∼8 m s−1 as the front encountered the strongest along-barrier flow during landfall. The adjusted gravity current velocity thus reasonably approximates the front’s observed motion in this case.
The time series of dual-Doppler results in Fig. 11 indicates that the formation of the NCFR was related to the rapid enhancement of the low-level frontal lifting during the period of landfall. Gravity current dynamics relate strong lifting at the surface front to the sharp temperature gradient between the warm and cold air masses. On the basis of quasi-steady dynamics, Parsons (1992) argued that the prefrontal ambient vertical shear (in the cross-front direction) constrains the intensity of the lifting at the leading edge of cold air. He performed a series of numerical experiments and demonstrated that an optimal vertical shear exists such that the frontal updraft is upright and maximized. A complementary view of this process involves horizontal vorticity balance arguments, as advanced by Rotunno et al. (1988). Following this approach, horizontal vorticity induced by density changes across the leading edge of the cold pool may be represented by
i1520-0493-128-6-1577-e2
where Δθmin is the cold-pool temperature deficit relative to ambient conditions, θ0 is the base-state potential temperature, and H is the cold-pool depth. The ambient shear-induced vorticity can be represented by Δu, which is the vertical shear over the depth of the cold pool (H). Rotunno et al. (1988) maintain that the situation in which C = Δu represents an optimal condition in which convection at the leading edge of the cold pool is upright and most intense.

If we use surface observations at the buoy 46027 to estimate thermodynamic values in (2), and the depth of the cold air mass associated with the front is again assumed to be 2 km, a corresponding value of ∼17.1 m s−1 is obtained for C. The magnitude of Δu may be readily estimated from the dual-Doppler synthesis winds. Figure 22 shows that the strongest cross-front vertical shear was confined to the lowest 1–2 km, with weaker shear aloft. The mean Δu over the depth of the cold air during the interval over which the NCFR was observed to rapidly form (encompassed by analyses at 0515, 0550, and 0642 UTC) was 8.6, 13.0, and 15.7 m s−1, respectively. The significant increase in Δu between 0515 and 0550 UTC was exclusively caused by the enhancement of cross-front velocities at low levels (Fig. 22), which was in turn due to intensification of southerly along-barrier flow over time in association with upstream blocking of prefrontal southwesterly flow (as described in section 6a).

The above calculations suggest that the vertical shear in advance of the front was generally suboptimal (i.e., Δu < C) during landfall, consistent with dual-Doppler observations showing frontal updrafts sloping over the cold air (i.e., exhibiting upshear tilt) over the entire period (cf. Fig. 11). Nonetheless, a progression toward increasingly upright and intense frontal ascent is evident in Fig. 11. Provided that the temporal variation of C was comparatively small,4 this observed Δu represents an evolution toward increasingly optimal conditions during the interval 0515–0642 UTC. It follows that upstream blocking within the 0–2-km layer was likely a crucial effect contributing to the observed rapid development and intensification of the NCFR. Further support for this hypothesis is that the subsequent reduction in the strength of observed along-barrier flow between 0642 and 0843 UTC (viz., from a maximum value of ∼26 m s−1 in Fig. 9d to 18 m s−1 in Fig. 13d) was accompanied by dissipation of the NCFR. Other factors that are not implicit in gravity current theory, such as frontogenesis and large-scale advection, represent possible complications to this interpretation. Nonetheless, the particular relationship of the coevolving fields of wind and precipitation to the steep coastal terrain are highly suggestive of a causal relationship between upstream blocking and the NCFR’s rapid formation and decay.

7. Conclusions

Through analysis of a comprehensive set of observations from the COAST experiment the mesoscale structure and evolution of a cold frontal system that made landfall near the Oregon–California border on 1 December 1995 has been documented in detail. This section of coastline, extending southward from Cape Blanco to Cape Mendocino, is notable for its steepness and presents a profound and relatively two-dimensional obstacle to landfalling storms. As the front neared shore, hourly averaged wind profiler and surface observations showed the along-barrier pressure gradient and low-level winds to increase markedly, reaching their greatest magnitudes immediately prior to frontal landfall.

The front was oriented northeast–southwest, at an acute angle to the steep coastal barrier, and thus advanced southward along the shore with time. As the front moved southward along the zone of highest coastal terrain, airborne dual-Doppler radar observations revealed a rapid increase in the intensity of precipitation and low-level convergence along the front. These changes occurred in conjunction with the formation of a narrow cold-frontal rainband. During the ∼1.5 h period over which an intense NCFR was present, prefrontal flow at low levels was strongly deflected by the orography, resulting in a sharp transition of wind direction from southwesterly (offshore) to south-southwesterly within ∼20 km of the coast, where maximum observed wind speeds defining the core of the “barrier jet” reached 26–28 m s−1 at 0.75 km. A marked deceleration of cross-barrier flow was also evident within this zone. Strengthening of the prefrontal winds thus resulted primarily from the intensification of the along-barrier flow component. In contrast to the strong prefrontal winds, flow in the cool postfrontal air mass was considerably weaker (∼10 m s−1) and appeared relatively unaffected by the orography. Frequency distributions of kinematic quantities during this landfalling period showed a rapid enhancement of prefrontal cross- and along-front velocities within the layer below ∼1.5 km occupied by coastal mountains. This contrasting response of the prefrontal and postfrontal flows to the identical steep coastal terrain likely resulted from the greater static stability observed in the prefrontal zone.

This observed modification of prefrontal flow adjacent to the coastal barrier is qualitatively consistent with upstream blocking as assessed using traditional Froude/Burger number analysis. The most profound influence of coastal orography was, however, confined to a zone within ∼20 km of the coast—a distance appreciably less than that predicted by theory (cf. estimated Rossby radius of deformation of ∼170 km). This result implies that effects not fully accounted for by idealized theory (e.g., diabatic processes, baroclinic modulations of upstream flow, and stability) likely altered the orographically forced response in this case. The asymmetric and convoluted nature of the actual topography (i.e., which comprises not only an isolated coastal barrier but multiple ranges of even higher mountains farther inland) represents yet another complication not readily accounted for by existing theory (Braun et al. 1999). Moreover, other potential influences on the dynamics of flow in the coastal zone including deformation of flow around Cape Mendocino (cf. Fig. 1), finite length of the steep coastline, and surface roughness/friction may also have played a role. Though not definitively addressed here, these factors are deserving of future examination through sensitivity tests using a mesoscale numerical forecast model.

Another important result of this study concerns the observed evolution of the NCFR and related flow dynamics in zones of steep coastal orography. Our findings indicate that the interaction between the advancing cool air mass and prefrontal cross-front vertical shear [reminiscent of processes occurring along gust fronts in warm-season convective systems as described by Rotunno et al. (1988)] may influence the development of the NCFR in ways that cannot be explained in terms of synoptic-scale processes such as secondary circulations predicted by semigeostrophic theory (e.g., chapter 2 of Bluestein 1993). Observations presented herein indicate that blocking of the low-level prefrontal flow upstream of steep coastal terrain significantly intensified the prefrontal low-level along-barrier flow, which in turn altered the cross-front vertical wind shear immediately ahead of the advancing cool air mass. The front’s acute orientation relative to the coastal barrier on 1 December 1995 was thus an important factor contributing to the NCFR’s evolution, for if the front were oriented parallel to the coastal barrier, the degree of upstream blocking would be greatly reduced and hence the cross-front vertical shear would not be modulated in this manner. As the front advanced southward toward Cape Mendocino, the flow pattern appeared to relax, the along-barrier flow decreased, and the NCFR dissipated, consistent with a steady decrease in the length of steep coastline to the south that was available to block and divert warm-sector flow northward toward the frontal boundary.

During the period of frontal landfall, two prominent prefrontal rainbands were observed within 80–100 km of the shore. These bands approximately paralleled the coastal barrier, and both intensified as they entered the zone of strongest low-level upstream blocking and associated low-level confluence. Cumulative radar reflectivity statistics during this period show evidence of a systematic increase in precipitation intensity toward shore, with mean radar reflectivities ∼40 km offshore being only 60%–70% of those found at the coast. This trend was closely related to cross-shore variations of low-level convergence, which was strongly influenced by the blocking of prefrontal flow. The influence of orographic forcing upon the organization and intensity of offshore precipitation is thus suggested. Observations from the operational WSR-88D radar on Cape Mendocino also provide evidence of a more direct link between coastal orography and precipitation. Rainfall inferred from maps of cumulative reflectivity was generally enhanced over windward slopes within ∼20–50 km of the coast, though local rainfall maxima were analyzed near the summits of coastal hills/mountains having peak elevations of 500–1500 m MSL.

Schumacher et al. (1996) demonstrate that large-scale upper-tropospheric conditions can significantly influence frontal behavior in the vicinity of mountains. In the absence of detailed upper-level measurements (as might be provided by an additional aircraft releasing dropsondes over the open Pacific), such effects could not be definitively evaluated in the present case. Moreover, because our detailed mesoscale analyses of three-dimensional airflow and precipitation were largely confined to a zone extending from the coast to at most 50 km offshore, changes in lower-tropospheric frontal structure and associated orographic precipitation farther upstream and directly over the inland sloped terrain could not be thoroughly addressed, and deserve more attention in the future. Appropriately, processes active directly over steep orography are also a major focus of the upcoming joint U.S.–European Mesoscale Alpine Programme (Houze et al. 1998). While analysis of data collected during COAST strongly suggests that orographic forcing is capable of inducing rapid evolution of frontal structure within the coastal zone, the extent to which these changes relate to frontogenesis is difficult to assess owing to a lack of detailed thermodynamic information. Future observational programs should concentrate on obtaining a more complete description of offshore conditions, including open-ocean stratification and frontal structure, in tandem with detailed observations near the coast. Only then will a wholly adequate description of dynamic processes contributing to rapidly evolving coastal weather conditions be possible.

Acknowledgments

We are indebted to Prof. Robert Houze for generously sharing his experience in the observation and analysis of cool-season frontal systems in the Pacific Northwest, as well as for his detailed comments that significantly improved the manuscript. Dr. Brad Colman (NOAA/NWS) kindly assisted us in obtaining data from the Eureka WSR-88D radar. We also extend thanks to Prof. Sandra Yuter for her expert assistance in processing of CFAD analyses, to Stacy Brodzik and Curtis James for assistance in processing data from the Eureka radar, to Dan Gottas and Paul Neiman (NOAA/ETL) for providing processed wind profiler data, and to David M. Johnson (NOAA/NSSL) for assistance in editing of P-3 airborne Doppler radar data. We thank Dr. Nicholas Bond, Dr. Brian Colle and two anonymous reviewers for their constructive and insightful comments on the manuscript, as well as Candace Gudmundson and Kay Dewar for their expert technical assistance in preparing the text and illustrations. We also wish to recognize the staff of NOAA’s Aircraft Operations Center, particularly Aircraft Commander Phil Kennedy and Flight Director Barry Damiano, whose dedication made this study possible. This research was supported by ONR Grants N00014-97-1-0717 and N00014-95-F-0045.

REFERENCES

  • Bell, G. D., and L. F. Bosart, 1988: Appalachian cold-air damming. Mon. Wea. Rev.,116, 137–161.

  • Bluestein, H. B., 1993: Synoptic-Dynamic Meteorology in Midlatitudes. Vol. II, Observations and Theory of Weather Systems, Oxford Press, 594 pp.

  • Bond, N. A., and Coauthors, 1997: The Coastal Observations and Simulations with Topography (COAST) experiment. Bull. Amer. Meteor. Soc.,78, 1941–1955.

  • Braun, S. A., R. A. Houze Jr., and B. F. Smull, 1997: Airborne dual-Doppler observations of an intense frontal system approaching the Pacific Northwest coast. Mon. Wea. Rev.,125, 3131–3156.

  • ——, R. Rotunno, and J. B. Klemp, 1999: Effects of coastal orography on landfalling cold fronts. Part I: Dry, inviscid dynamics. J. Atmos. Sci.,56, 517–533.

  • Browning, K. A., and C. W. Pardoe, 1973: Structure of low-level jet streams ahead of midlatitude cold fronts. Quart. J. Roy. Meteor. Soc.,99, 619–638.

  • ——, F. F. Hill, and C. W. Pardoe, 1974: Structure and mechanism of precipitation and the effect orography in a wintertime warm sector. Quart. J. Roy. Meteor. Soc.,100, 309–330.

  • Carbone, R., 1982: A severe frontal rainband. Part I: Stormwide hydrodynamic structure. J. Atmos. Sci.,39, 258–279.

  • Chen, W.-D., and R. B. Smith, 1987: Blocking and deflection of airflow by the Alps. Mon. Wea. Rev.,115, 2578–2597.

  • Colle, B. A., and C. F. Mass, 1996: An observational and modeling study of the interaction of low-level southwesterly flow with the Olympic Mountains during COAST IOP 4. Mon. Wea. Rev.,124, 2152–2175.

  • ——, ——, and B. F. Smull, 1999: An observational and numerical study of a cold front interacting with the Olympic Mountains during COAST IOP 5. Mon. Wea. Rev.,127, 1310–1334.

  • Doyle, J. D., 1997: The influence of mesoscale orography on a coastal jet and rainband. Mon. Wea. Rev.,125, 1465–1488.

  • Durran, D. R., and J. B. Klemp, 1982: On the effects of moisture on the Brunt–Väisälä frequency. J. Atmos. Sci.,39, 2152–2158.

  • Grossman, R. L., and D. R. Durran, 1984: Interaction of low-level flow with the western Ghat Mountains and offshore convection in the summer monsoon. Mon. Wea. Rev.,112, 652–672.

  • Hildebrand, P. H., 1989: Airborne Doppler radar accuracy. Preprints, 24th Conf. on Radar Meteorology, Tallahassee, FL, Amer. Meteor. Soc., 585–588.

  • Hobbs, P. V., and K. R. Biswas, 1979: The cellular structure of narrow cold-frontal rainbands. Quart. J. Roy. Meteor. Soc.,105, 723–727.

  • ——, and P. O. G. Persson, 1982: The mesoscale and microscale structure of clouds and precipitation in midlatitude cyclones. V: The substructure of narrow cold frontal rainbands. J. Atmos. Sci.,39, 280–295.

  • ——, R. A. Houze Jr., and T. J. Matejka, 1975: The dynamical and microphysical structure of an occluded frontal system and its modification by orography. J. Atmos. Sci.,32, 1542–1562.

  • ——, T. J. Matejka, P. H. Herzegh, J. D. Locatelli, and R. A. Houze Jr., 1980: The mesoscale and microscale structure and organization of clouds and precipitation in midlatitude cyclones. I: A case study of a cold front. J. Atmos. Sci.,37, 568–596.

  • Houze, R. A., Jr., 1993: Cloud Dynamics. Academic Press, 573 pp.

  • ——, J. Kuettner, and R. Smith, Eds., 1998: Mesoscale Alpine Programme U.S. overview document and experiment design. UCAR, Boulder, CO, 30 pp. [Available via e-mail request to rjm@ucar.edu.].

  • James, P. K., and K. A. Browning, 1979: Mesoscale structure of line convection at surface cold fronts. Quart. J. Roy. Meteor. Soc.,105, 371–382.

  • Jorgensen, D. P., 1984: Mesoscale and convective-scale characteristics of mature hurricanes. Part I: General observations by research aircraft. J. Atmos. Sci.,41, 1268–1285.

  • ——, and B. F. Smull, 1993: Mesovortex circulations seen by airborne Doppler radar within a bow-echo mesoscale convective system. Bull. Amer. Meteor. Soc.,74, 2146–2157.

  • ——, T. Matejka, and J. D. DuGranrut, 1994: Multi-beam techniques for deriving wind fields from airborne Doppler radars. Meteor. Atmos. Phys.,59, 83–104.

  • ——, M. A. LeMone, and S. B. Trier, 1997: Structure and evolution of the 22 February 1993 TOGA COARE squall line: Aircraft observations of precipitation, circulation, and surface energy fluxes. J. Atmos. Sci.,54, 1961–1985.

  • Joss, J., and A. Waldvogel, 1990: Precipitation measurements and hydrology. Radar in Meteorology, D. Atlas, Ed., Amer. Meteor. Soc., 577–606.

  • Kuo, Y.-H., and G. T.-J. Chen, 1990: The Taiwan Area Mesoscale Experiment (TAMEX): An overview. Bull. Amer. Meteor. Soc.,71, 488–503.

  • Lackmann, G. M., and J. E. Overland, 1989: Atmospheric structure and momentum balance during a gap-wind event in Shelikof strait, Alaska. Mon. Wea. Rev.,117, 1817–1833.

  • Leise, J. A., 1981: A multidimensional scale-telescoped filter and data extension package. NOAA Tech. Memo. ERL WPL-82, 18 pp. [NTIS PB82-164104.].

  • Li, J., and Y.-L. Chen, 1998: Barrier jets during TAMEX. Mon. Wea. Rev.,126, 959–971.

  • Mass, C. F., and G. K. Ferber, 1990: Surface pressure perturbations produced by an isolated mesoscale topographic barrier. Part I: General characteristics and dynamics. Mon. Wea. Rev.,118, 2579–2596.

  • Matejka, T., and S. A. Lewis, 1997: Improving research aircraft navigation by incorporating INS and GPS information in a variational solution. J. Atmos. Oceanic Technol.,14, 495–511.

  • O’Brien, J. J., 1970: Alternative solutions to the classical vertical velocity problem. J. Appl. Meteor.,9, 197–203.

  • Overland, J. E., 1984: Scale analysis of marine winds in straits and along mountainous coasts. Mon. Wea. Rev.,112, 2530–2534.

  • ——, and N. A. Bond, 1993: The influence of coastal orography: The Yakutat storm. Mon. Wea. Rev.,121, 1388–1397.

  • ——, and ——, 1995: Observations and scale analysis of coastal wind jets. Mon. Wea. Rev.,123, 2934–2941.

  • Parish, T. R., 1982: Barrier winds along the Sierra Nevada mountains. J. Appl. Meteor.,21, 925–930.

  • Parsons, D. B., 1992: An explanation of intense frontal updrafts and narrow cold-frontal rainbands. J. Atmos. Sci.,49, 1810–1825.

  • ——, and P. V. Hobbs, 1983a: The mesoscale and microscale structure and organization of clouds and precipitation in midlatitude cyclones. IX: Some effects of orography on rainbands. J. Atmos. Sci.,40, 1930–1949.

  • ——, and ——, 1983b: The mesoscale and microscale structure and organization of clouds and precipitation in midlatitude cyclones. XI: Comparisons between observational and theoretical aspects of rainbands. J. Atmos. Sci.,40, 2377–2397.

  • Pierrehumbert, R. T., 1984: Linear results on the barrier effects of mesoscale mountains. J. Atmos. Sci.,41, 1356–1367.

  • ——, and B. Wyman, 1985: Upstream effects of mesoscale mountains. J. Atmos. Sci.,42, 977–1003.

  • Rotunno, R., J. B. Klemp, and M. L. Weisman, 1988: A theory for strong, long-lived squall lines. J. Atmos. Sci.,45, 463–485.

  • Roux, F., V. Marécal, and D. Hauser, 1993: The 12/13 January 1988 narrow cold-frontal rainband observed during MFDP/FRONTS 87. Part I: Kinematics and thermodynamics. J. Atmos. Sci.,50, 951–974.

  • Schumacher, P. N., D. J. Knight, and L. F. Bosart, 1996: Frontal interaction with the Appalachian Mountains. Part I: A climatology. Mon. Wea. Rev.,124, 2453–2468.

  • Shapiro, M. A., 1984: Meteorological tower measurements of a surface cold front. Mon. Wea. Rev.,112, 1634–1639.

  • Simpson, J. E., 1969: A comparison between atmospheric and laboratory density currents. Quart. J. Roy. Meteor. Soc.,95, 758–765.

  • ——, and R. E. Britter, 1980: A laboratory model of an atmospheric mesofront. Quart. J. Roy. Meteor. Soc.,106, 485–500.

  • Smith, R. B., 1979: The influence of mountains on the atmosphere. Advances in Geophysics, Vol. 21, Academic Press, 87–230.

  • Trier, S. B., D. B. Parsons, and T. J. Matejka, 1990: Observations of a subtropical cold front in a region of complex terrain. Mon. Wea. Rev.,118, 2449–2470.

  • Yuter, S. E., and R. A. Houze Jr., 1995: Three-dimensional kinematic and microphysical evolution of Florida cumulonimbus. Part II: Frequency distributions of vertical velocity, reflectivity, and differential reflectivity. Mon. Wea. Rev.,123, 1941–1963.

Fig. 1.
Fig. 1.

Topography along the Pacific Northwest coast of the United States. Terrain height (m MSL) is indicated by shading at 200-m intervals (key at right), and major orographic barriers are noted. Locations of select surface observing stations and marine buoys are indicated by dots along with their three-letter/five-digit identifiers, respectively. Locations of the 915-MHz research wind profiler site at Crescent City (CRC) and operational WSR-88D radar site near Eureka (EKA) are indicated by the square and triangle, respectively.

Citation: Monthly Weather Review 128, 6; 10.1175/1520-0493(2000)128<1577:ADOOAL>2.0.CO;2

Fig. 2.
Fig. 2.

The 850-mb analysis from NCEP valid at 0000 UTC 1 Dec 1995. Full wind barbs correspond to 5 m s−1; half barbs to 2.5 m s−1.

Citation: Monthly Weather Review 128, 6; 10.1175/1520-0493(2000)128<1577:ADOOAL>2.0.CO;2

Fig. 3.
Fig. 3.

Sea level pressure and frontal analysis valid at 0600 UTC on 1 Dec 1995. Isobars are plotted at 2-mb intervals. Full wind barbs correspond to 5 m s−1; half barbs to 2.5 m s−1. The heavy dashed line denotes the pressure trough.

Citation: Monthly Weather Review 128, 6; 10.1175/1520-0493(2000)128<1577:ADOOAL>2.0.CO;2

Fig. 4.
Fig. 4.

Composite radar reflectivity maps from the P-3 lower fuselage radar at (a) 0525–0530, (b) 0640–0645, and (c) 0846–0850 UTC 1 Dec 1995. Shading at 5-dBZ intervals (key at upper left) denotes reflectivities exceeding 25 dBZ. The aircraft track during each composite interval is indicated by small arrows. The aircraft altitude corresponding to these composite intervals is (a) ∼1550 m, (b) ∼1850 m, and (c) ∼1250 m MSL. Inset boxes locate 135 × 135 km2 dual-Doppler analysis domains. In each panel, terrain height contours at 600, 1000, and 1600 m MSL are indicated by the dashed–dotted, thin-solid, and thick-solid contours, respectively. The heavy dashed line in each panel indicates the estimated location of the surface cold front (based on dual-Doppler observations). Prefrontal rainbands A and B (see text) are denoted in (b) and (c), respectively.

Citation: Monthly Weather Review 128, 6; 10.1175/1520-0493(2000)128<1577:ADOOAL>2.0.CO;2

Fig. 5.
Fig. 5.

Objectively analyzed time section of hourly surface winds (wind barbs as in Fig. 3), sea level pressure (solid contours at 1-mb interval), and surface temperature (dashed contours at 1°C interval) along a 1200-km north–south segment of the Pacific Northwest coast for the interval 0100–2000 UTC 1 Dec 1995. Locations of selected surface stations (identifiers along left margin) used to construct this analysis are indicated in Fig. 1. Shading indicates surface winds >10 m s−1. Heavy solid and dashed lines denote subjectively analyzed positions of the surface cold front and pressure trough, respectively. (right) Corresponding meridional profile of maximum terrain height within ∼120 km of the coast.

Citation: Monthly Weather Review 128, 6; 10.1175/1520-0493(2000)128<1577:ADOOAL>2.0.CO;2

Fig. 6.
Fig. 6.

Time–height cross section of hourly averaged winds observed by the 915-MHz profiler at Crescent City from 0000 to 2300 UTC on 1 Dec 1995. Full barbs and flags represent wind speeds of 5 and 25 m s−1, respectively.

Citation: Monthly Weather Review 128, 6; 10.1175/1520-0493(2000)128<1577:ADOOAL>2.0.CO;2

Fig. 7.
Fig. 7.

Vertical profiles of potential temperature and equivalent potential temperature in the prefrontal region (dashed lines) and the postfrontal region (solid lines). The prefrontal profiles were obtained from an aircraft sounding conducted ∼30 km southwest of buoy 46027 (cf. Fig. 1) between 0543 and 0546 UTC, with values below 948 m MSL linearly interpolated from observed values at buoy 46027. The postfrontal profiles were constructed from the dropsonde observation at a point ∼30 km behind the front near 0800 UTC.

Citation: Monthly Weather Review 128, 6; 10.1175/1520-0493(2000)128<1577:ADOOAL>2.0.CO;2

Fig. 8.
Fig. 8.

Ground-relative winds and radar reflectivity (dBZ, key to shading at upper left) at 0.75 km MSL derived from the airborne dual-Doppler analysis at 0515 UTC. (a) Wind vectors (key at upper right) and surface cold-front position (heavy dashed line); (b) corresponding wind speeds with isotach contour interval of 2 m s−1; (c) and (d) corresponding cross-barrier (east–west) and along-barrier (north–south) flow components, respectively; with 2 m s−1 contour interval. In each panel, terrain height thresholds of 600 and 1000 m MSL are indicated by dashed–dotted and thick-solid contours, respectively.

Citation: Monthly Weather Review 128, 6; 10.1175/1520-0493(2000)128<1577:ADOOAL>2.0.CO;2

Fig. 9.
Fig. 9.

As in Fig. 8 except at 0642 UTC. In (a), thick line segments A–B and C–D mark locations of vertical cross sections shown in Figs. 10 and 11, respectively. Flight track as shown in Fig. 4b. In (b), “A” marks location of prefrontal rainband A (see text).

Citation: Monthly Weather Review 128, 6; 10.1175/1520-0493(2000)128<1577:ADOOAL>2.0.CO;2

Fig. 10.
Fig. 10.

Vertical cross section along A–B in Fig. 9a showing radar reflectivity (dBZ, shading key at upper left) in conjunction with (a) cross-barrier and (b) along-barrier component flow (2 m s−1 contour interval) at 0642 UTC. Heavy solid line in lower-right portion of each panel indicates height of coastal topography along the section.

Citation: Monthly Weather Review 128, 6; 10.1175/1520-0493(2000)128<1577:ADOOAL>2.0.CO;2

Fig. 11.
Fig. 11.

Vertical cross section along C–D in Fig. 9a showing time sequence of radar reflectivity (dBZ, shading key at upper left) and front-relative wind vectors (key at upper right) at (a) 0515, (b) 0550, and (c) 0642 UTC. Heavy solid line in lower-right portion of each panel indicates height of coastal topography along the section, while arrow denotes estimated location of surface front.

Citation: Monthly Weather Review 128, 6; 10.1175/1520-0493(2000)128<1577:ADOOAL>2.0.CO;2

Fig. 12.
Fig. 12.

As in Fig. 11 except showing horizontal divergence (contour interval 4 × 10−4 s−1). Solid (dashed) lines denote positive (negative) divergence values, respectively.

Citation: Monthly Weather Review 128, 6; 10.1175/1520-0493(2000)128<1577:ADOOAL>2.0.CO;2

Fig. 13.
Fig. 13.

As in Fig. 8 except for 0843 UTC. In (a), “B” marks location of prefrontal rainband B (see text).

Citation: Monthly Weather Review 128, 6; 10.1175/1520-0493(2000)128<1577:ADOOAL>2.0.CO;2

Fig. 14.
Fig. 14.

Horizontal pattern of cumulative normalized reflectivity (shading according to key at upper left; see text) derived from the Eureka WSR-88D radar during the period of intensive P-3 observations (0500–1100 UTC 1 Dec 1995). Terrain height thresholds of 600, 1000, and 1600 m MSL are indicated by the dashed–dotted, thin-solid, and thick-solid contours, respectively.

Citation: Monthly Weather Review 128, 6; 10.1175/1520-0493(2000)128<1577:ADOOAL>2.0.CO;2

Fig. 15.
Fig. 15.

Low-level composite radar reflectivity maps (dBZ, key to shading at upper left) from the Eureka WSR-88D radar (+ symbol) at (a) 0610, (b) 0622, (c) 0633, and (d) 0645 UTC. The “A” marks location of prefrontal rainband A. Inset box in (d) indicates the airborne dual-Doppler analysis domain.

Citation: Monthly Weather Review 128, 6; 10.1175/1520-0493(2000)128<1577:ADOOAL>2.0.CO;2

Fig. 16.
Fig. 16.

Base scan (0.5° elevation) color display of (a) radar reflectivity (dBZ) and (b) radial velocity (kt) from the Eureka WSR-88D radar at 0929 UTC. Arrow in (b) indicates ill-defined discontinuity marking the front.

Citation: Monthly Weather Review 128, 6; 10.1175/1520-0493(2000)128<1577:ADOOAL>2.0.CO;2

Fig. 17.
Fig. 17.

CFADs (see text) of ground-relative cross-front velocity at (a) 0515, (b) 0550, and (c) 0642 UTC. The bin size is 2.5 m s−1 and the vertical grid increment is 250 m. The contour interval is 10% (m s−1)−1 km−1 and the bold contour represents a threshold value of 10% (m s−1)−1 km−1.

Citation: Monthly Weather Review 128, 6; 10.1175/1520-0493(2000)128<1577:ADOOAL>2.0.CO;2

Fig. 18.
Fig. 18.

As in Fig. 17 except for ground-relative alongfront velocity.

Citation: Monthly Weather Review 128, 6; 10.1175/1520-0493(2000)128<1577:ADOOAL>2.0.CO;2

Fig. 19.
Fig. 19.

Contoured frequency distribution of radar reflectivity for heights ⩽1.5 km as a function of offshore distance (see text) at (a) 0550, (b) 0642, and (c) 0843 UTC. The bin size is 2.5 dBZ and the horizontal increment is 4.5 km. The contour interval is 0.5% (dBZ)−1 km−1 and the bold contour represents a threshold value of 1% (dBZ)−1 km−1.

Citation: Monthly Weather Review 128, 6; 10.1175/1520-0493(2000)128<1577:ADOOAL>2.0.CO;2

Fig. 20.
Fig. 20.

Low-level (below 1.5 km MSL) mean horizontal divergence as a function of offshore distance (i.e., averaged in the alongshore direction) derived from airborne dual-Doppler analyses at 0550, 0642, and 0843 UTC (solid, dotted, and dashed lines, respectively).

Citation: Monthly Weather Review 128, 6; 10.1175/1520-0493(2000)128<1577:ADOOAL>2.0.CO;2

Fig. 21.
Fig. 21.

Horizontal divergence (contour interval 4 × 10−4 s−1) at 0.75 km MSL derived from the dual-Doppler analysis at 0843 UTC. (a) Total divergence, (b) contribution by the variation of zonal (∼cross-barrier) flow in the east–west (∼cross barrier) direction (∂u/∂x), and (c) corresponding contribution by the variation of along-barrier flow in the north–south direction (∂υ/∂y). Solid and dashed lines indicating positive and negative values, respectively; regions of convergence value greater than 4 × 10−4 s−1 are shaded. (a) Heavy dashed line marks estimated surface front position. Terrain height thresholds of 600 and 1000 m MSL are indicated by dashed–dotted and the thick-solid contours, respectively.

Citation: Monthly Weather Review 128, 6; 10.1175/1520-0493(2000)128<1577:ADOOAL>2.0.CO;2

Fig. 22.
Fig. 22.

Vertical profiles of mean cross-front velocity within the prefrontal region (ahead of the surface frontal shear line) derived from dual-Doppler observations at 0515, 0550, and 0642 UTC (solid, dotted, and dashed lines, respectively).

Citation: Monthly Weather Review 128, 6; 10.1175/1520-0493(2000)128<1577:ADOOAL>2.0.CO;2

1

The Froude number is defined as Fr = U/HN, where U is the speed of incident flow, H is the mountain height, and N is the Brunt–Väisälä frequency.

2

The Burger number is defined as B = HN/Lf, where H is the mountain height, N is the Brunt–Väisälä frequency, L is the mountain half-width, and f is the Coriolis parameter.

3

The Swiss Meteorological Agency uses a similar procedure to produce horizontal maps of radar reflectivity over regions adjacent to the Alps (Joss and Waldvogel 1990).

4

The influence of diabatic cooling in the cold air mass due to hydrometeor evaporation and melting should not have been significant during this period because the largest precipitation rates were confined to the prefrontal region (cf. Fig. 11) and conditions in the vicinity of the front were quasi-saturated.

Save
  • Bell, G. D., and L. F. Bosart, 1988: Appalachian cold-air damming. Mon. Wea. Rev.,116, 137–161.

  • Bluestein, H. B., 1993: Synoptic-Dynamic Meteorology in Midlatitudes. Vol. II, Observations and Theory of Weather Systems, Oxford Press, 594 pp.

  • Bond, N. A., and Coauthors, 1997: The Coastal Observations and Simulations with Topography (COAST) experiment. Bull. Amer. Meteor. Soc.,78, 1941–1955.

  • Braun, S. A., R. A. Houze Jr., and B. F. Smull, 1997: Airborne dual-Doppler observations of an intense frontal system approaching the Pacific Northwest coast. Mon. Wea. Rev.,125, 3131–3156.

  • ——, R. Rotunno, and J. B. Klemp, 1999: Effects of coastal orography on landfalling cold fronts. Part I: Dry, inviscid dynamics. J. Atmos. Sci.,56, 517–533.

  • Browning, K. A., and C. W. Pardoe, 1973: Structure of low-level jet streams ahead of midlatitude cold fronts. Quart. J. Roy. Meteor. Soc.,99, 619–638.

  • ——, F. F. Hill, and C. W. Pardoe, 1974: Structure and mechanism of precipitation and the effect orography in a wintertime warm sector. Quart. J. Roy. Meteor. Soc.,100, 309–330.

  • Carbone, R., 1982: A severe frontal rainband. Part I: Stormwide hydrodynamic structure. J. Atmos. Sci.,39, 258–279.

  • Chen, W.-D., and R. B. Smith, 1987: Blocking and deflection of airflow by the Alps. Mon. Wea. Rev.,115, 2578–2597.

  • Colle, B. A., and C. F. Mass, 1996: An observational and modeling study of the interaction of low-level southwesterly flow with the Olympic Mountains during COAST IOP 4. Mon. Wea. Rev.,124, 2152–2175.

  • ——, ——, and B. F. Smull, 1999: An observational and numerical study of a cold front interacting with the Olympic Mountains during COAST IOP 5. Mon. Wea. Rev.,127, 1310–1334.

  • Doyle, J. D., 1997: The influence of mesoscale orography on a coastal jet and rainband. Mon. Wea. Rev.,125, 1465–1488.

  • Durran, D. R., and J. B. Klemp, 1982: On the effects of moisture on the Brunt–Väisälä frequency. J. Atmos. Sci.,39, 2152–2158.

  • Grossman, R. L., and D. R. Durran, 1984: Interaction of low-level flow with the western Ghat Mountains and offshore convection in the summer monsoon. Mon. Wea. Rev.,112, 652–672.

  • Hildebrand, P. H., 1989: Airborne Doppler radar accuracy. Preprints, 24th Conf. on Radar Meteorology, Tallahassee, FL, Amer. Meteor. Soc., 585–588.

  • Hobbs, P. V., and K. R. Biswas, 1979: The cellular structure of narrow cold-frontal rainbands. Quart. J. Roy. Meteor. Soc.,105, 723–727.

  • ——, and P. O. G. Persson, 1982: The mesoscale and microscale structure of clouds and precipitation in midlatitude cyclones. V: The substructure of narrow cold frontal rainbands. J. Atmos. Sci.,39, 280–295.

  • ——, R. A. Houze Jr., and T. J. Matejka, 1975: The dynamical and microphysical structure of an occluded frontal system and its modification by orography. J. Atmos. Sci.,32, 1542–1562.

  • ——, T. J. Matejka, P. H. Herzegh, J. D. Locatelli, and R. A. Houze Jr., 1980: The mesoscale and microscale structure and organization of clouds and precipitation in midlatitude cyclones. I: A case study of a cold front. J. Atmos. Sci.,37, 568–596.

  • Houze, R. A., Jr., 1993: Cloud Dynamics. Academic Press, 573 pp.

  • ——, J. Kuettner, and R. Smith, Eds., 1998: Mesoscale Alpine Programme U.S. overview document and experiment design. UCAR, Boulder, CO, 30 pp. [Available via e-mail request to rjm@ucar.edu.].

  • James, P. K., and K. A. Browning, 1979: Mesoscale structure of line convection at surface cold fronts. Quart. J. Roy. Meteor. Soc.,105, 371–382.

  • Jorgensen, D. P., 1984: Mesoscale and convective-scale characteristics of mature hurricanes. Part I: General observations by research aircraft. J. Atmos. Sci.,41, 1268–1285.

  • ——, and B. F. Smull, 1993: Mesovortex circulations seen by airborne Doppler radar within a bow-echo mesoscale convective system. Bull. Amer. Meteor. Soc.,74, 2146–2157.

  • ——, T. Matejka, and J. D. DuGranrut, 1994: Multi-beam techniques for deriving wind fields from airborne Doppler radars. Meteor. Atmos. Phys.,59, 83–104.

  • ——, M. A. LeMone, and S. B. Trier, 1997: Structure and evolution of the 22 February 1993 TOGA COARE squall line: Aircraft observations of precipitation, circulation, and surface energy fluxes. J. Atmos. Sci.,54, 1961–1985.

  • Joss, J., and A. Waldvogel, 1990: Precipitation measurements and hydrology. Radar in Meteorology, D. Atlas, Ed., Amer. Meteor. Soc., 577–606.

  • Kuo, Y.-H., and G. T.-J. Chen, 1990: The Taiwan Area Mesoscale Experiment (TAMEX): An overview. Bull. Amer. Meteor. Soc.,71, 488–503.

  • Lackmann, G. M., and J. E. Overland, 1989: Atmospheric structure and momentum balance during a gap-wind event in Shelikof strait, Alaska. Mon. Wea. Rev.,117, 1817–1833.

  • Leise, J. A., 1981: A multidimensional scale-telescoped filter and data extension package. NOAA Tech. Memo. ERL WPL-82, 18 pp. [NTIS PB82-164104.].

  • Li, J., and Y.-L. Chen, 1998: Barrier jets during TAMEX. Mon. Wea. Rev.,126, 959–971.

  • Mass, C. F., and G. K. Ferber, 1990: Surface pressure perturbations produced by an isolated mesoscale topographic barrier. Part I: General characteristics and dynamics. Mon. Wea. Rev.,118, 2579–2596.

  • Matejka, T., and S. A. Lewis, 1997: Improving research aircraft navigation by incorporating INS and GPS information in a variational solution. J. Atmos. Oceanic Technol.,14, 495–511.

  • O’Brien, J. J., 1970: Alternative solutions to the classical vertical velocity problem. J. Appl. Meteor.,9, 197–203.

  • Overland, J. E., 1984: Scale analysis of marine winds in straits and along mountainous coasts. Mon. Wea. Rev.,112, 2530–2534.

  • ——, and N. A. Bond, 1993: The influence of coastal orography: The Yakutat storm. Mon. Wea. Rev.,121, 1388–1397.

  • ——, and ——, 1995: Observations and scale analysis of coastal wind jets. Mon. Wea. Rev.,123, 2934–2941.

  • Parish, T. R., 1982: Barrier winds along the Sierra Nevada mountains. J. Appl. Meteor.,21, 925–930.

  • Parsons, D. B., 1992: An explanation of intense frontal updrafts and narrow cold-frontal rainbands. J. Atmos. Sci.,49, 1810–1825.

  • ——, and P. V. Hobbs, 1983a: The mesoscale and microscale structure and organization of clouds and precipitation in midlatitude cyclones. IX: Some effects of orography on rainbands. J. Atmos. Sci.,40, 1930–1949.

  • ——, and ——, 1983b: The mesoscale and microscale structure and organization of clouds and precipitation in midlatitude cyclones. XI: Comparisons between observational and theoretical aspects of rainbands. J. Atmos. Sci.,40, 2377–2397.

  • Pierrehumbert, R. T., 1984: Linear results on the barrier effects of mesoscale mountains. J. Atmos. Sci.,41, 1356–1367.

  • ——, and B. Wyman, 1985: Upstream effects of mesoscale mountains. J. Atmos. Sci.,42, 977–1003.

  • Rotunno, R., J. B. Klemp, and M. L. Weisman, 1988: A theory for strong, long-lived squall lines. J. Atmos. Sci.,45, 463–485.

  • Roux, F., V. Marécal, and D. Hauser, 1993: The 12/13 January 1988 narrow cold-frontal rainband observed during MFDP/FRONTS 87. Part I: Kinematics and thermodynamics. J. Atmos. Sci.,50, 951–974.

  • Schumacher, P. N., D. J. Knight, and L. F. Bosart, 1996: Frontal interaction with the Appalachian Mountains. Part I: A climatology. Mon. Wea. Rev.,124, 2453–2468.

  • Shapiro, M. A., 1984: Meteorological tower measurements of a surface cold front. Mon. Wea. Rev.,112, 1634–1639.

  • Simpson, J. E., 1969: A comparison between atmospheric and laboratory density currents. Quart. J. Roy. Meteor. Soc.,95, 758–765.

  • ——, and R. E. Britter, 1980: A laboratory model of an atmospheric mesofront. Quart. J. Roy. Meteor. Soc.,106, 485–500.

  • Smith, R. B., 1979: The influence of mountains on the atmosphere. Advances in Geophysics, Vol. 21, Academic Press, 87–230.

  • Trier, S. B., D. B. Parsons, and T. J. Matejka, 1990: Observations of a subtropical cold front in a region of complex terrain. Mon. Wea. Rev.,118, 2449–2470.

  • Yuter, S. E., and R. A. Houze Jr., 1995: Three-dimensional kinematic and microphysical evolution of Florida cumulonimbus. Part II: Frequency distributions of vertical velocity, reflectivity, and differential reflectivity. Mon. Wea. Rev.,123, 1941–1963.

  • Fig. 1.

    Topography along the Pacific Northwest coast of the United States. Terrain height (m MSL) is indicated by shading at 200-m intervals (key at right), and major orographic barriers are noted. Locations of select surface observing stations and marine buoys are indicated by dots along with their three-letter/five-digit identifiers, respectively. Locations of the 915-MHz research wind profiler site at Crescent City (CRC) and operational WSR-88D radar site near Eureka (EKA) are indicated by the square and triangle, respectively.

  • Fig. 2.

    The 850-mb analysis from NCEP valid at 0000 UTC 1 Dec 1995. Full wind barbs correspond to 5 m s−1; half barbs to 2.5 m s−1.

  • Fig. 3.

    Sea level pressure and frontal analysis valid at 0600 UTC on 1 Dec 1995. Isobars are plotted at 2-mb intervals. Full wind barbs correspond to 5 m s−1; half barbs to 2.5 m s−1. The heavy dashed line denotes the pressure trough.

  • Fig. 4.

    Composite radar reflectivity maps from the P-3 lower fuselage radar at (a) 0525–0530, (b) 0640–0645, and (c) 0846–0850 UTC 1 Dec 1995. Shading at 5-dBZ intervals (key at upper left) denotes reflectivities exceeding 25 dBZ. The aircraft track during each composite interval is indicated by small arrows. The aircraft altitude corresponding to these composite intervals is (a) ∼1550 m, (b) ∼1850 m, and (c) ∼1250 m MSL. Inset boxes locate 135 × 135 km2 dual-Doppler analysis domains. In each panel, terrain height contours at 600, 1000, and 1600 m MSL are indicated by the dashed–dotted, thin-solid, and thick-solid contours, respectively. The heavy dashed line in each panel indicates the estimated location of the surface cold front (based on dual-Doppler observations). Prefrontal rainbands A and B (see text) are denoted in (b) and (c), respectively.

  • Fig. 5.

    Objectively analyzed time section of hourly surface winds (wind barbs as in Fig. 3), sea level pressure (solid contours at 1-mb interval), and surface temperature (dashed contours at 1°C interval) along a 1200-km north–south segment of the Pacific Northwest coast for the interval 0100–2000 UTC 1 Dec 1995. Locations of selected surface stations (identifiers along left margin) used to construct this analysis are indicated in Fig. 1. Shading indicates surface winds >10 m s−1. Heavy solid and dashed lines denote subjectively analyzed positions of the surface cold front and pressure trough, respectively. (right) Corresponding meridional profile of maximum terrain height within ∼120 km of the coast.

  • Fig. 6.

    Time–height cross section of hourly averaged winds observed by the 915-MHz profiler at Crescent City from 0000 to 2300 UTC on 1 Dec 1995. Full barbs and flags represent wind speeds of 5 and 25 m s−1, respectively.

  • Fig. 7.

    Vertical profiles of potential temperature and equivalent potential temperature in the prefrontal region (dashed lines) and the postfrontal region (solid lines). The prefrontal profiles were obtained from an aircraft sounding conducted ∼30 km southwest of buoy 46027 (cf. Fig. 1) between 0543 and 0546 UTC, with values below 948 m MSL linearly interpolated from observed values at buoy 46027. The postfrontal profiles were constructed from the dropsonde observation at a point ∼30 km behind the front near 0800 UTC.

  • Fig. 8.

    Ground-relative winds and radar reflectivity (dBZ, key to shading at upper left) at 0.75 km MSL derived from the airborne dual-Doppler analysis at 0515 UTC. (a) Wind vectors (key at upper right) and surface cold-front position (heavy dashed line); (b) corresponding wind speeds with isotach contour interval of 2 m s−1; (c) and (d) corresponding cross-barrier (east–west) and along-barrier (north–south) flow components, respectively; with 2 m s−1 contour interval. In each panel, terrain height thresholds of 600 and 1000 m MSL are indicated by dashed–dotted and thick-solid contours, respectively.

  • Fig. 9.

    As in Fig. 8 except at 0642 UTC. In (a), thick line segments A–B and C–D mark locations of vertical cross sections shown in Figs. 10 and 11, respectively. Flight track as shown in Fig. 4b. In (b), “A” marks location of prefrontal rainband A (see text).

  • Fig. 10.

    Vertical cross section along A–B in Fig. 9a showing radar reflectivity (dBZ, shading key at upper left) in conjunction with (a) cross-barrier and (b) along-barrier component flow (2 m s−1 contour interval) at 0642 UTC. Heavy solid line in lower-right portion of each panel indicates height of coastal topography along the section.

  • Fig. 11.

    Vertical cross section along C–D in Fig. 9a showing time sequence of radar reflectivity (dBZ, shading key at upper left) and front-relative wind vectors (key at upper right) at (a) 0515, (b) 0550, and (c) 0642 UTC. Heavy solid line in lower-right portion of each panel indicates height of coastal topography along the section, while arrow denotes estimated location of surface front.

  • Fig. 12.

    As in Fig. 11 except showing horizontal divergence (contour interval 4 × 10−4 s−1). Solid (dashed) lines denote positive (negative) divergence values, respectively.

  • Fig. 13.

    As in Fig. 8 except for 0843 UTC. In (a), “B” marks location of prefrontal rainband B (see text).

  • Fig. 14.

    Horizontal pattern of cumulative normalized reflectivity (shading according to key at upper left; see text) derived from the Eureka WSR-88D radar during the period of intensive P-3 observations (0500–1100 UTC 1 Dec 1995). Terrain height thresholds of 600, 1000, and 1600 m MSL are indicated by the dashed–dotted, thin-solid, and thick-solid contours, respectively.

  • Fig. 15.

    Low-level composite radar reflectivity maps (dBZ, key to shading at upper left) from the Eureka WSR-88D radar (+ symbol) at (a) 0610, (b) 0622, (c) 0633, and (d) 0645 UTC. The “A” marks location of prefrontal rainband A. Inset box in (d) indicates the airborne dual-Doppler analysis domain.

  • Fig. 16.

    Base scan (0.5° elevation) color display of (a) radar reflectivity (dBZ) and (b) radial velocity (kt) from the Eureka WSR-88D radar at 0929 UTC. Arrow in (b) indicates ill-defined discontinuity marking the front.

  • Fig. 17.

    CFADs (see text) of ground-relative cross-front velocity at (a) 0515, (b) 0550, and (c) 0642 UTC. The bin size is 2.5 m s−1 and the vertical grid increment is 250 m. The contour interval is 10% (m s−1)−1 km−1 and the bold contour represents a threshold value of 10% (m s−1)−1 km−1.

  • Fig. 18.

    As in Fig. 17 except for ground-relative alongfront velocity.

  • Fig. 19.

    Contoured frequency distribution of radar reflectivity for heights ⩽1.5 km as a function of offshore distance (see text) at (a) 0550, (b) 0642, and (c) 0843 UTC. The bin size is 2.5 dBZ and the horizontal increment is 4.5 km. The contour interval is 0.5% (dBZ)−1 km−1 and the bold contour represents a threshold value of 1% (dBZ)−1 km−1.

  • Fig. 20.

    Low-level (below 1.5 km MSL) mean horizontal divergence as a function of offshore distance (i.e., averaged in the alongshore direction) derived from airborne dual-Doppler analyses at 0550, 0642, and 0843 UTC (solid, dotted, and dashed lines, respectively).

  • Fig. 21.

    Horizontal divergence (contour interval 4 × 10−4 s−1) at 0.75 km MSL derived from the dual-Doppler analysis at 0843 UTC. (a) Total divergence, (b) contribution by the variation of zonal (∼cross-barrier) flow in the east–west (∼cross barrier) direction (∂u/∂x), and (c) corresponding contribution by the variation of along-barrier flow in the north–south direction (∂υ/∂y). Solid and dashed lines indicating positive and negative values, respectively; regions of convergence value greater than 4 × 10−4 s−1 are shaded. (a) Heavy dashed line marks estimated surface front position. Terrain height thresholds of 600 and 1000 m MSL are indicated by dashed–dotted and the thick-solid contours, respectively.

  • Fig. 22.

    Vertical profiles of mean cross-front velocity within the prefrontal region (ahead of the surface frontal shear line) derived from dual-Doppler observations at 0515, 0550, and 0642 UTC (solid, dotted, and dashed lines, respectively).

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 382 75 6
PDF Downloads 137 39 1