• Brewer, M., , C. Mass, , and B. Potter, 2012: The West Coast thermal trough: Climatology and synoptic evolution. Mon. Wea. Rev., 140, 38203843.

    • Search Google Scholar
    • Export Citation
  • Chien, F. C., , C. F. Mass, , and Y. H. Kuo, 1997: Interaction of a warm-season frontal system with the coastal mountains of the western United States. Part I: Prefrontal onshore push, coastal ridging, and alongshore southerlies. Mon. Wea. Rev., 125, 17051729.

    • Search Google Scholar
    • Export Citation
  • Colson, D., 1956: Meteorological problems associated with mass fires. Fire Control Notes, 17, 911.

  • Cramer, O. P., 1954: Recognizing weather conditions that affect forest fire behavior. Fire Control Notes, 15, 16.

  • Cramer, O. P., 1957: Frequency of dry east winds over northwest Oregon and southwest Washington. U.S. Forest Service, Pacific Northwest Forest and Range Experiment Station, Research Paper 24, 19 pp.

  • Gisborne, H. T., 1927: Meteorological factors in the Quartz Creek forest fire. Mon. Wea. Rev., 55, 5660.

  • Glickman, T., Ed., 2000: Glossary of Meteorology. 2nd ed. Amer. Meteor. Soc., 855 pp.

  • Jannuzzi, J. A., 1993: The onshore push of marine air into the Pacific Northwest. Wea. Forecasting, 8, 194202.

  • Mass, C. F., , and N. A. Bond, 1996: Coastally trapped wind reversals along the United States West Coast during the warm season. Part II: Synoptic evolution. Mon. Wea. Rev., 124, 446461.

    • Search Google Scholar
    • Export Citation
  • Mass, C. F., , and W. J. Steenburgh, 2000: An observational and numerical study of an orographically trapped wind reversal along the west coast of the United States. Mon. Wea. Rev., 128, 23632397.

    • Search Google Scholar
    • Export Citation
  • Mass, C. F., , M. D. Albright, , and D. J. Brees, 1986: The onshore surge of marine air into the Pacific Northwest: A coastal region of complex terrain. Mon. Wea. Rev., 114, 26022627.

    • Search Google Scholar
    • Export Citation
  • McKendry, I. G., 1994: Synoptic circulation and summertime ground-level ozone concentrations at Vancouver, British Columbia. J. Appl. Meteor., 33, 627641.

    • Search Google Scholar
    • Export Citation
  • Nuss, W. A., 2007: Synoptic-scale structure and the character of coastally trapped wind reversals. Mon. Wea. Rev., 135, 6081.

  • Nuss, W. A., and Coauthors, 2000: Coastally trapped wind reversals: Progress toward understanding. Bull. Amer. Meteor. Soc., 81, 719743.

    • Search Google Scholar
    • Export Citation
  • Werth, P., , and R. Ochoa, 1993: The evaluation of Idaho wildfire growth using the Haines index. Wea. Forecasting, 8, 223234.

  • View in gallery

    Composites of 850-hPa heights (m, contours), temperatures (°C, shading), wind (full barb = 5 m s−1), and 500-hPa heights (m) for the beginning of a WCTT event using NCEP's North American Regional Reanalysis (from BMP12).

  • View in gallery

    Topography and major features of the Pacific Northwest.

  • View in gallery

    (top) 500-hPa heights (m, contours) and wind (full barb = 5 m s−1); (middle) 850-hPa heights (m), temperature (°C, shading), and wind (m s−1); and (bottom) sea level pressure (hPa), 1000-hPa temperature (°C), and 10-m wind (m s−1) from a 36-km WRF Model simulation.

  • View in gallery

    (top) 500-hPa heights (m) and wind (m s−1); (middle) 850-hPa heights (m), temperature (°C), and wind (full barb = 5 m s−1); and (bottom) sea level pressure (hPa), 1000-hPa temperature (°C), and 10-m wind (m s−1) from a 36-km WRF Model simulation.

  • View in gallery

    (top) NARR 500-hPa heights (m, contours) temperature (°C, shading) and wind (full barb = 5 m s−1); (middle) 850-heights (m) temperature (°C) and wind (m s−1); and (bottom) sea level pressure (hPa), 1000-hPa temperature (°C), and 10-m wind (m s−1).

  • View in gallery

    (left) Observed soundings and (right) model soundings from the 4-km WRF Model at Salem, OR. Observed soundings are provided by the University of Wyoming.

  • View in gallery

    Geopotential height (m, contours), temperature (°C, shading), and wind (full barb = 5 m s−1) at (left) 850 and (middle) 925 hPa; and (right) sea level pressure, 1000-hPa temperature (°C), and wind (m s−1) at 1200 UTC 14 May. Temperatures are not shown if the pressure level is below the terrain.

  • View in gallery

    Heights (m, contours), temperatures (°C, shading), and wind (full barb = 5 m s−1) at (top) 850 and (middle) 925 hPa; and (bottom) sea level pressure (hPa), 1000-hPa temperature (°C), and 10-m wind (m s−1) from the 12-km WRF Model.

  • View in gallery

    (top) 6-h changes in height (m, contours) and temperature (°C, shading) at 850 hPa and (bottom) 6-h changes in sea level pressure (hPa) and 1000-hPa temperature (°C).

  • View in gallery

    Backward trajectories for the period of 0000–1200 UTC 14 May 2007. All trajectories end (1200 UTC) at 900 hPa and the width of the trajectories reflect their height.

  • View in gallery

    Cross sections along 43°N from 124.5° to 120.75°W displaying temperature (°C) and wind vectors parallel to the cross section.

  • View in gallery

    Theta tendency for each term in the thermodynamic energy equation on 14 May using the 12-km WRF Model. These cross sections are along 43°N from 124.5° to 121.5°W.

  • View in gallery

    Thermodynamic energy equation analysis at 850 hPa on 14 May.

  • View in gallery

    Heights (m, contours), temperatures (°C, shading), and wind (full barb = 5 m s−1) at (top) 850 and (middle) 925 hPa; and (bottom) sea level pressure (hPa), 1000-hPa temperature (°C), and 10-m wind (m s−1) all using the 12-km WRF Model.

  • View in gallery

    (top) 6-h changes of height (m, contours) and temperature (°C, shading) at 850 hPa and (bottom) 6-h changes in sea level pressure (hPa) and 1000-hPa temperature (°C).

  • View in gallery

    Heights (m, contours), temperatures (°C, shading), and wind (full barb = 5 m s−1) at (top) 850 and (middle) 925 hPa; and (bottom) sea level pressure (hPa), 1000-hPa temperature (°C), and 10-m wind (m s−1) all using the 12-km WRF Model.

  • View in gallery

    (top) 6-h changes of height (m, contours) and temperature (°C, shading) at 850 hPa and (bottom) 6-h changes in sea level pressure (hPa) and 1000-hPa temperature (°C).

  • View in gallery

    Vertical cross sections of terms in the thermodynamic energy equation for the 12-km WRF Model simulation. These cross sections are along 43°N from 122.5° to 118.5°W for the afternoon of 15 May 2007.

  • View in gallery

    The terms of the thermodynamic energy equation at 850 hPa at the same time as in Fig. 17.

  • View in gallery

    Topography (m) for the terrain experiments from the 36-km model.

  • View in gallery

    500-hPa heights (m) and wind (full barb = 5 m s−1) for the control, smooth, no-WC-terrain, and no-terrain runs.

  • View in gallery

    925-hPa heights (m, contours), temperature (°C, color shading), and wind (full barb = 5 m s−1).

  • View in gallery

    925-hPa heights (m, contours), temperature (°C, color shading), and wind (full barb = 5 m s−1) from the 36-km WRF Model.

  • View in gallery

    850-hPa heights (m, contours), temperature (°C, color shading), and wind (full barb = 5 m s−1).

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 49 49 6
PDF Downloads 42 42 5

The West Coast Thermal Trough: Mesoscale Evolution and Sensitivity to Terrain and Surface Fluxes

View More View Less
  • 1 University of Washington, Seattle, Washington
  • | 2 USDA Forest Service, Seattle, Washington
© Get Permissions
Full access

Abstract

Despite the significant impacts of the West Coast thermal trough (WCTT) on West Coast weather and climate, questions remain regarding its mesoscale structure, origin, and dynamics. Of particular interest is the relative importance of terrain forcing, advection, and surface heating on WCTT formation and evolution. To explore such questions, the 13–16 May 2007 WCTT event was examined using observations and simulations from the Weather Research and Forecasting (WRF) Model. An analysis of the thermodynamic energy equation for these simulations was completed, as well as sensitivity experiments in which terrain or surface fluxes were removed or modified. For the May 2007 event, vertical advection of potential temperature is the primary driver of local warming and WCTT formation west of the Cascades. The downslope flow that drives this warming is forced by easterly flow associated with high pressure over British Columbia, Canada. When the terrain is removed from the model, the WCTT does not form and high pressure builds over the northwest United States. When the WCTT forms on the east side of the Cascades, diabatic heating dominates over the other terms in the thermodynamic energy equation, with warm advection playing a small role. If surface heat fluxes are neglected, an area of low pressure remains east of the Cascades, though it is substantially attenuated.

Corresponding author address: Matthew C. Brewer, Department of Atmospheric Sciences, University of Washington, Box 351640, Seattle, WA 98195-164045. E-mail: mcbrewer83@yahoo.com

Abstract

Despite the significant impacts of the West Coast thermal trough (WCTT) on West Coast weather and climate, questions remain regarding its mesoscale structure, origin, and dynamics. Of particular interest is the relative importance of terrain forcing, advection, and surface heating on WCTT formation and evolution. To explore such questions, the 13–16 May 2007 WCTT event was examined using observations and simulations from the Weather Research and Forecasting (WRF) Model. An analysis of the thermodynamic energy equation for these simulations was completed, as well as sensitivity experiments in which terrain or surface fluxes were removed or modified. For the May 2007 event, vertical advection of potential temperature is the primary driver of local warming and WCTT formation west of the Cascades. The downslope flow that drives this warming is forced by easterly flow associated with high pressure over British Columbia, Canada. When the terrain is removed from the model, the WCTT does not form and high pressure builds over the northwest United States. When the WCTT forms on the east side of the Cascades, diabatic heating dominates over the other terms in the thermodynamic energy equation, with warm advection playing a small role. If surface heat fluxes are neglected, an area of low pressure remains east of the Cascades, though it is substantially attenuated.

Corresponding author address: Matthew C. Brewer, Department of Atmospheric Sciences, University of Washington, Box 351640, Seattle, WA 98195-164045. E-mail: mcbrewer83@yahoo.com

1. Introduction

The West Coast thermal trough (WCTT) has a substantial impact on Pacific Northwest weather and climate, particularly during the warm season. As an upper-level ridge moves toward the West Coast and associated low-level high pressure builds over Washington State and southwest Canada, an elongated area of low pressure extends from the southwest United States into the Pacific Northwest (Fig. 1). During the summer, WCTTs are generally associated with clear skies and unusually warm, dry conditions. An extreme example of the conditions accompanying a WCTT event occurred on 29 July 2009 when Seattle–Tacoma Airport reached its all-time high of 103°F (39.4°C). As a WCTT shifts to the east side of the Cascade Mountains, cooler air moves inland from the Pacific Ocean, often quite abruptly, in what is known as an onshore or marine push (Mass et al. 1986). The combined effects of the WCTT and subsequent onshore push have large impacts on wildfires, wind energy, marine interests, and the aviation community (Jannuzzi 1993; Brewer et al. 2012, hereafter BMP12).

Fig. 1.
Fig. 1.

Composites of 850-hPa heights (m, contours), temperatures (°C, shading), wind (full barb = 5 m s−1), and 500-hPa heights (m) for the beginning of a WCTT event using NCEP's North American Regional Reanalysis (from BMP12).

Citation: Monthly Weather Review 141, 8; 10.1175/MWR-D-12-00305.1

The complex topography of the Pacific Northwest is shown in Fig. 2. The interaction of this terrain with the synoptic-scale flow is critical to WCTT initiation and evolution. The Cascade Mountains divide Washington and Oregon into two climatic regions. The west side is greatly influenced by the Pacific Ocean, with moderate temperatures, higher humidity, and greater precipitation than the much drier region to the east of the Cascade crest. The Oregon plateau dominates the east side of Oregon with heights around 1000 m, while the Columbia basin is found in eastern Washington with the Rocky Mountains to the east.

Fig. 2.
Fig. 2.

Topography and major features of the Pacific Northwest.

Citation: Monthly Weather Review 141, 8; 10.1175/MWR-D-12-00305.1

A number of papers have analyzed coastally trapped wind reversals (CTWRs), onshore pushes, wildfires, and northwest U.S. air quality, providing insights into WCTT structure and development. From composites of onshore push events, Mass et al. (1986) described the large-scale evolution of the WCTT and the importance of a synoptic ridge aloft in initiating development. Mass and Bond (1996) described the synoptic evolution of WCTTs using composites of CTWR events. Chien et al. (1997) and Mass and Steenburgh (2000) demonstrated that a CTWR/WCTT can be realistically simulated using a high-resolution mesoscale model [the fifth-generation Pennsylvania State University–National Center for Atmospheric Research (PSU–NCAR) Mesoscale Model (MM5)]. Many studies describe WCTTs as an important meteorological factor favoring wildfire growth in the Pacific Northwest (Gisborne 1927; Cramer 1954; Colson 1956; Cramer 1957; Werth and Ochoa 1993). McKendry (1994) found that days exceeding the Canadian 1-h ambient O3 air quality objective of 82 ppb at Vancouver, British Columbia, Canada, coincide with an upper-level ridge and a WCTT to the west of the city.

BMP12 examined WCTT climatology and synoptic evolution, revealing that WCTTs are most frequent during the autumn, although they are common throughout the year. West of the Cascade crest, WCTTs are most evident during the morning hours since solar heating east of the Cascades results in falling heights and pressure over the interior during the afternoon hours, thus obscuring the WCTT signal on the west side. East of the Cascades, warm-season WCTTs are most frequent during the afternoon/evening. There is no diurnal preference of WCTT occurrence during the cold season.

The composites of WCTT formation in BMP12 showed seasonal differences in WCTT synoptic evolution. Warm season WCTTs are associated with a ridge aloft moving over the northwest United States, resulting in surface high pressure building over southwest Canada and the northward extension of the WCTT along the U.S. West Coast. For cold season WCTT events, there is an upper-level trough over the southwest United States, reflected by low pressure at the surface, and higher pressure over the interior, resulting in offshore flow.

Glickman (2000) describes a thermal low as an area where lower-tropospheric warming leads to a lifting of isobaric surfaces, resulting in divergence aloft and lower pressure near the surface. Much of the literature that discusses WCTTs have suggested that adiabatic warming, forced by downslope flow, plays a significant role in this lower-tropospheric heating (Mass et al. 1986; Mass and Bond 1996; Chien et al. 1997; Nuss et al. 2000; Mass and Steenburgh 2000; Nuss 2007). However, others have suggested that advection also plays an important role in WCTT formation (Mass and Bond 1996; Mass and Steenburgh 2000; Nuss et al. 2000; Nuss 2007) or that diabatic heating is crucial (Chien et al. 1997; Jannuzzi 1993). Chien et al. (1997) simulated a WCTT/CTWR event in which terrain or surface fluxes were removed from the model. They concluded that diabatic heating is the primary driver of WCTT formation, with downslope flow being secondary. However, their model runs were not initialized until the end of the WCTT event, and thus could not provide reliable guidance on WCTT origin. Similarly, Chien et al. (1997) and Mass and Steenburgh (2000) analyzed the thermodynamic energy equation for their respective case studies, but neither examined WCTT initiation.

More recently, BMP12 examined the origin of WCTTs, noting that summer WCTT events are most frequent during the morning hours, and that WCTTs can occur during winter when diabatic heating is minimal. These observations suggest that diabatic heating is not essential for WCTT formation west of the Cascades. In contrast, WCTT events are most frequent during the afternoon/evening east of the Cascades, indicating that diabatic heating plays a larger role in WCTT formation in this region. It was also found that WCTTs are most frequent to the west of the major mountain ranges of the Pacific Northwest, thus suggesting that easterly downslope flow plays an important role.

There is still considerable uncertainty regarding the mesoscale structure of WCTTs west of the Cascades as well as the mechanisms underlying WCTT development and subsequent movement to the east side of the Cascades. There is no study describing the detailed mesoscale evolution of WCTT events, and important questions remain. What is the relative importance of surface fluxes and downslope flow in thermal trough formation and evolution? Does advection play a significant role? Such issues will be addressed in this paper. Specifically, this manuscript will describe the synoptic and mesoscale evolution of a representative WCTT event that occurred on 13–16 May 2007 using observations and simulations from the Weather Research and Forecasting (WRF) Model. This event is similar to the canonical composite event described in BMP12. Particular focus will be given to the hours leading up to the formation of the WCTT on both the west and east sides of the Cascades. An analysis of the thermodynamic energy equation is used to describe the relative contributions of diabatic heating, advection, and terrain forcing in the development of the WCTT. In addition, model runs with modified terrain or without surface fluxes are used to determine the underlying physical mechanisms in the formation of the WCTT.

2. Overview of 13–16 May 2007 event

This analysis makes use of several WRF Model simulations (version 3.4.1) that use an outer nest with 36-km grid spacing covering much of the northeastern Pacific Ocean and western North America, and an interior 12-km nest covering the northwest United States and British Columbia. An additional 4-km nested grid includes Washington, Oregon, and Idaho. The Yonsei University (YSU) boundary layer scheme, the new Thompson microphysics parameterization, the Rapid Radiative Transfer Model (RRTM) longwave radiation scheme, Dudhia shortwave radiation, 38 vertical full-sigma levels, and a model top of 100 hPa were applied. Soil temperature and moisture fields are from the National Centers for Environmental Prediction (NCEP) North American Mesoscale (NAM) model, and sea surface temperatures are from the U.S. Navy Fleet Numerical Meteorology and Oceanography Center's High Resolution Ocean Analysis. The WRF Model was run for 72 h and the model initialization and lateral boundary fields came from the 0000 UTC 13 May 2007 run of NCEP's Global Forecast System (GFS) model. As shown below, this simulation closely matched the observed synoptic/mesoscale evolution.

Figure 3 provides model forecasts at 500 hPa (top) and 850 hPa (middle) and the surface (bottom) during the initiation of the event. At 1200 UTC 13 May, a 500-hPa trough was centered over the Pacific Northwest, with a ridge building over the eastern Pacific. Over the next 24 h, the ridge amplified and moved eastward to the coast. During this same period at sea level and 850 hPa, high pressure built over the northwest United States and British Columbia, a coastal trough extended northward from California along with a tongue of warm air, and low-level winds over Oregon and Washington turned from westerly to northerly to easterly.

Fig. 3.
Fig. 3.

(top) 500-hPa heights (m, contours) and wind (full barb = 5 m s−1); (middle) 850-hPa heights (m), temperature (°C, shading), and wind (m s−1); and (bottom) sea level pressure (hPa), 1000-hPa temperature (°C), and 10-m wind (m s−1) from a 36-km WRF Model simulation.

Citation: Monthly Weather Review 141, 8; 10.1175/MWR-D-12-00305.1

Over the next 36 h, the 500-hPa ridge continued its eastward progression, reaching Idaho at 0000 UTC 16 May (Fig. 4). The low-level high pressure associated with it, once centered over British Columbia, moved southeastward over Montana and then to the east side of the Rockies. At 0000 and 1200 UTC 15 May, the WCTT and the associated tongue of warm air remained on the western side of the Cascades as easterlies continued over the inland portions of Oregon and Washington. By 0000 UTC 16 May, the WCTT had shifted east of the Cascades at both sea level and 850 hPa as low-level high pressure and westerlies pushed inland.

Fig. 4.
Fig. 4.

(top) 500-hPa heights (m) and wind (m s−1); (middle) 850-hPa heights (m), temperature (°C), and wind (full barb = 5 m s−1); and (bottom) sea level pressure (hPa), 1000-hPa temperature (°C), and 10-m wind (m s−1) from a 36-km WRF Model simulation.

Citation: Monthly Weather Review 141, 8; 10.1175/MWR-D-12-00305.1

Since the WRF Model output is used to diagnose the mesoscale evolution and dynamics of the thermal trough event, an evaluation of the fidelity of the model fields is required. Specifically, fields from NCEP's North American Regional Reanalysis (NARR), found in Fig. 5, can be compared to the WRF Model run shown in Figs. 3 and 4. In general, there is good agreement, especially at 500 hPa. There are only minor differences in the simulated and analyzed temperatures, heights, and winds at 850 hPa and sea level. Importantly, both the NARR and the WRF agree on the timing of the initiation of the WCTT, its evolution west of the Cascades, and its eventual formation on the east side of the Cascades.

Fig. 5.
Fig. 5.

(top) NARR 500-hPa heights (m, contours) temperature (°C, shading) and wind (full barb = 5 m s−1); (middle) 850-heights (m) temperature (°C) and wind (m s−1); and (bottom) sea level pressure (hPa), 1000-hPa temperature (°C), and 10-m wind (m s−1).

Citation: Monthly Weather Review 141, 8; 10.1175/MWR-D-12-00305.1

Comparing simulated and observed soundings is another way of evaluating model realism. To that end, observed and 4-km WRF Model soundings at Salem, Oregon, for this case were constructed (Fig. 6). In general, model temperatures are in good agreement with the observations, though there are substantial differences in dewpoint at middle and upper levels. Wind directions are also in good agreement, with model and observations generally within 10° of each other. Model wind speeds appear to be around 5 kt (~2.6 m s−1) higher than observed for middle- and upper-tropospheric levels. These soundings will be used in the next section to describe the mesoscale evolution of this case.

Fig. 6.
Fig. 6.

(left) Observed soundings and (right) model soundings from the 4-km WRF Model at Salem, OR. Observed soundings are provided by the University of Wyoming.

Citation: Monthly Weather Review 141, 8; 10.1175/MWR-D-12-00305.1

3. Mesoscale overview

Before describing the mesoscale evolution of this event, it is important to understand how heights and temperatures are extrapolated below terrain by the display software, and the implications of that extrapolation. The graphics shown below are produced by the Read/Interpolate/Plot (RIP) postprocessing package, which computes “underground” temperature, geopotential heights, and sea level pressure by taking temperatures at the lowest model layer and extrapolating them below ground using a constant lapse rate (6.5°C, hereafter Γm). Vertically averaging the underground temperature and plugging it into the hypsometric equation produces geopotential heights and sea level pressure. This method of vertical extrapolation is common in models and postprocessing software used by the operational community. Pressure and height artifacts arise in complex/high terrain when the observed lapse rate (Γo) differs from Γm, a frequent occurrence. The pressure reduction problem can be significant in the early morning since stability increases at night, resulting in a sounding not well represented by Γm. During summer days, areas such as the Central Valley of California and the Columbia basin warm up, resulting in a nearly dry adiabatic lapse rate. Because Γo > Γm during the daytime, an area of low pressure forms in these valleys and basins relative to the higher terrain, where the assumed vertical gradient in temperature is less and thus low-level temperatures are cooler.

To illustrate some of these issues, Fig. 7 shows model temperatures, wind, and geopotential heights at 850, 925, and 1000 hPa at 1200 UTC 14 May, with temperature displayed only when the model level is above the terrain. A significant amount of the northwest U.S. terrain extends above 925 hPa, while 850 hPa is generally above the surface except for the higher mountains and plateaus. Since the Sierra and Rocky Mountains rise much higher than 850 hPa, caution should be exercised when viewing maps at 850 hPa and below in these regions.

Fig. 7.
Fig. 7.

Geopotential height (m, contours), temperature (°C, shading), and wind (full barb = 5 m s−1) at (left) 850 and (middle) 925 hPa; and (right) sea level pressure, 1000-hPa temperature (°C), and wind (m s−1) at 1200 UTC 14 May. Temperatures are not shown if the pressure level is below the terrain.

Citation: Monthly Weather Review 141, 8; 10.1175/MWR-D-12-00305.1

At 850 hPa, the axis of the WCTT was along the coast and broad in scale, while at sea level and 925 hPa, the axis was over the Cascades and far narrower. At this time, there was a strong surface-based inversion at Salem that extended up to 850 hPa (Fig. 6). Because Γo was much less than Γm, the reduction approach put the warmest air and thus the lowest pressure under the Cascades, producing an unphysical narrow pressure trough under the highest terrain.

a. Initiation of the WCTT

Using the 12-km WRF simulation, a more detailed view of the initiation of the May 2007 WCTT event is shown in Fig. 8. On 0000 UTC 14 May, cool lower-tropospheric temperatures were found offshore and over the lowlands of western Oregon and Washington. Winds along the coast and offshore were generally northerly at all three levels, while over inland Washington they were westerly. Coastal winds were more northeasterly along the Oregon–California border, likely resulting in downslope warming and low-level troughing over the western slopes of the coastal and Siskiyou–Klamath Mountains. A tongue of warm air at 850 hPa was found along the southern Oregon coast and was associated with low-level troughing. As suggested above, some of the troughing at lower levels, especially the narrow tongue of troughing under the highest terrain, was probably amplified by the pressure reduction issue. Over the next 12 hours, an area of low pressure in the northeast corner of the domain moved eastward at all three levels, high pressure built into the northwest United States in its wake, and a region of easterly flow, warm air, and coastal troughing extended northward. The axis of the coastal trough appears to shift progressively westward and northward between sea level and 850 hPa.

Fig. 8.
Fig. 8.

Heights (m, contours), temperatures (°C, shading), and wind (full barb = 5 m s−1) at (top) 850 and (middle) 925 hPa; and (bottom) sea level pressure (hPa), 1000-hPa temperature (°C), and 10-m wind (m s−1) from the 12-km WRF Model.

Citation: Monthly Weather Review 141, 8; 10.1175/MWR-D-12-00305.1

The observed and modeled soundings over Salem show the transition from westerly to easterly flow and rapid low-level warming during the development of the WCTT (Fig. 6). The sounding at 1200 UTC 13 May was moist at low levels with northwesterlies increasing with height. Winds turned northeasterly below 800 hPa over the next 24 h, with substantial warming between 700 and 900 hPa leading to the development of a strong surface-based inversion and a large drop in relative humidity.

Figure 9 shows 6-h changes of 850-hPa heights, 850-hPa temperatures, and sea level pressure and 1000-hPa temperature during the development phase of the thermal trough. From 1800 UTC 13 May to 0000 UTC 14 May, 850-hPa heights rose over the northern two-thirds of the domain, with the largest increases over eastern Washington. Similar pressure changes occurred at sea level, though pressure falls were more extensive, particularly over western Oregon. The increases in height/pressure over the northern portion of the domain were associated with the ridging aloft and the eastward movement of low pressure out of the domain. Warming was found only over land at 1000 hPa, but extended offshore at 850 hPa. During 0000–0600 UTC 14 May, heights/pressure rose on the east side of the domain, with pressure falls along the southern Oregon and Northern California coasts at both levels. Temperatures warmed west of the Cascades at 850 hPa, while cooling predominated at 1000 hPa with the sole exception being the coastal zone in the lee of the coastal mountains near the Oregon–California border. The 850-hPa warming to the west of the Cascade crest continued during the next 6-h period (0600–1200 UTC 14 May), and was key to the hydrostatic reduction in pressure leading to the formation of the WCTT. A strip of warming caused by downslope flow on the western side of the Cascades is evident at 1000 hPa. Height and pressure fell over the ocean and western Oregon and Washington during this later period.

Fig. 9.
Fig. 9.

(top) 6-h changes in height (m, contours) and temperature (°C, shading) at 850 hPa and (bottom) 6-h changes in sea level pressure (hPa) and 1000-hPa temperature (°C).

Citation: Monthly Weather Review 141, 8; 10.1175/MWR-D-12-00305.1

Figure 10 shows backward trajectories for the 12 h leading to WCTT development (0000–1200 UTC 14 May). Trajectories over the ocean indicate subsidence due to synoptic descent associated with the large-scale ridge. Over land, subsidence is further enhanced by easterly downslope flow, evident by the rapid narrowing of the trajectories along the western slopes of the Cascades and Siskiyou/Klamath Mountains.

Fig. 10.
Fig. 10.

Backward trajectories for the period of 0000–1200 UTC 14 May 2007. All trajectories end (1200 UTC) at 900 hPa and the width of the trajectories reflect their height.

Citation: Monthly Weather Review 141, 8; 10.1175/MWR-D-12-00305.1

Figure 11 shows cross sections along 43°N and from 124.5° to 120.75°W (roughly central Oregon to the coast), with temperatures and winds in the plane of the cross section during the formation of the WCTT. At 1200 UTC 13 May, there was offshore flow at low levels near the coast, with cool nighttime temperatures and an inversion near 800 hPa. Twelve hours later, strong boundary layer heating led to upslope flow and westerlies over and west of the Cascade crest, as well as a weakening of the inversion. At 0400 UTC 14 May residual warm air from daytime heating was still evident on both sides of the Cascades, and offshore flow had begun over the lowest 100 hPa above the terrain. Four hours later (0800 UTC), downslope flow strengthened on the upper western slopes of the Cascades, resulting in an area of adiabatically warmed air that was advected westward by offshore flow. Significantly, between 0400 and 0800 UTC, the warmest air switched from the east to the west side of the Cascades. As easterly downslope flow continued through the night and into the morning (1200–1600 UTC), downslope flow and westward advection of adiabatically warmed air continued. Large-scale subsidence is evident by warming near the 700-mb level between 1200 UTC 13 May and 1200 UTC 14 May. This level is well above the crest, and thus the warming would not be a result of adiabatic compression with downslope flow.

Fig. 11.
Fig. 11.

Cross sections along 43°N from 124.5° to 120.75°W displaying temperature (°C) and wind vectors parallel to the cross section.

Citation: Monthly Weather Review 141, 8; 10.1175/MWR-D-12-00305.1

To evaluate the origin of the warm air west of the Cascade crest that led to the formation of the WCTT, an analysis of model output was completed using the thermodynamic energy equation:
eq1
The term on the left-hand side of the equation is the local potential temperature (θ, theta) change. On the right-hand side of the equation, the left term is diabatic heating, the middle term is horizontal advection of theta, and the term on the right is vertical advection of theta. Diabatic heating is calculated by adding together the contributions to theta tendency from the PBL parameterization, the cumulus scheme, microphysical latent heating, and radiation. After summing these terms and adding them to the advection terms, the sum was compared to the model total theta tendency, and it was found that there were only very small residual, thus ensuring that the budget was closed.

Using output from the 12-km WRF simulation, the terms from the thermodynamic energy equation were plotted in a cross section along 43°N for the period of WCTT development along the West Coast (Fig. 12). At 0200 UTC 14 May, there was lower-troposphere warming west of the Cascades starting roughly 100 hPa above the slopes. This warming moved closer to the surface, particularly over the upper slopes, during the subsequent hours, even with nighttime diabatic cooling near the surface. The origin of the warming appears to be twofold. Vertical advection of higher values of θ aloft due to downslope flow was dominant near the surface. This higher theta air was then advected westward away from the Cascade slopes. In addition, vertical advection warming from large-scale synoptic subsidence contributed to a general warming of the whole region, particularly aloft. There appears to be two mountain-related subsidence areas: one associated with the Cascades and the other with the Coast Range. As the WCTT extended northward during the night, the diabatic term did not contribute to the formation of the WCTT, and vertical theta advection remained dominant.

Fig. 12.
Fig. 12.

Theta tendency for each term in the thermodynamic energy equation on 14 May using the 12-km WRF Model. These cross sections are along 43°N from 124.5° to 121.5°W.

Citation: Monthly Weather Review 141, 8; 10.1175/MWR-D-12-00305.1

The magnitudes of the various thermodynamic terms at 850 hPa during thermal trough development are shown in plan view in Fig. 13. The theta tendency term shows warming along the west slopes of the Cascades and offshore throughout the night. The diabatic term generally shows cooling, except at 0200 UTC. Downward vertical advection warming dominates, with particularly large values along the western slopes of the Cascades. Horizontal advection is weak in general, with a gradual strengthening of warm advection over the western Oregon lowlands and offshore during this period.

Fig. 13.
Fig. 13.

Thermodynamic energy equation analysis at 850 hPa on 14 May.

Citation: Monthly Weather Review 141, 8; 10.1175/MWR-D-12-00305.1

b. Evolution of the WCTT west of the Cascade Mountains

At 1800 UTC 14 May, the WCTT axis shifted substantially westward at 850 hPa compared to the morning hours, while moving only slightly at lower levels (Fig. 14). The tongue of warm air at 850 hPa had broadened since 1200 UTC 14 May as a result of the westward advection of warm air and the surface heating of the interior. The warmest air in the 1000-hPa analysis was under the southern Oregon Cascades, reflecting that Γo < Γm. Westward advection of adiabatically warmed air likely contributed to a westward shift in the WCTT axis with height, since the cool sea surface temperatures prevented the thermal ridge from extending over the ocean near the surface.

Fig. 14.
Fig. 14.

Heights (m, contours), temperatures (°C, shading), and wind (full barb = 5 m s−1) at (top) 850 and (middle) 925 hPa; and (bottom) sea level pressure (hPa), 1000-hPa temperature (°C), and 10-m wind (m s−1) all using the 12-km WRF Model.

Citation: Monthly Weather Review 141, 8; 10.1175/MWR-D-12-00305.1

Over the next 12 h (through 0600 UTC 15 May), the WCTT weakened at 850 hPa, even though offshore flow continued and a tongue of warm air was still evident west of the Cascades at 0600 UTC. The sea level trough extended farther north into Canada. The soundings over Salem revealed very dry conditions throughout the troposphere at 0000 UTC 15 May, the warmest period for this location (Fig. 6).

Figure 15 illustrates the low-level pressure, height, and temperature change during this period. Between 1200 and 1800 UTC 14 May, heights rose substantially at 850 hPa over the eastern side of the domain in connection with the southeastward movement of high pressure from southwest Canada. This evolution strengthened the downslope and offshore flow, particularly over the eastern side of the domain. Warming was observed offshore at 850 hPa, but only over land at 1000 hPa.

Fig. 15.
Fig. 15.

(top) 6-h changes of height (m, contours) and temperature (°C, shading) at 850 hPa and (bottom) 6-h changes in sea level pressure (hPa) and 1000-hPa temperature (°C).

Citation: Monthly Weather Review 141, 8; 10.1175/MWR-D-12-00305.1

Between 1800 and 0000 UTC, warming extended over the entire domain at 850 hPa and continued over land near the surface, with height/pressure falls over most of the region at both levels. The 850-hPa heights fell over land due to strong surface heating, particularly over eastern Washington and over the ocean due an incoming synoptic low, the latter weakening WCTT amplitude. Finally, during the evening period of 0000–0600 UTC 15 May, cooling spread over most of the domain at 850 hPa, with the exception of western Oregon and Washington and the offshore waters. Heights/pressure fell over the northwestern portion of the domain due to the low-level warming as well as a result of the incoming synoptic low pressure system.

c. WCTT movement to the east side of the Cascades

At 1200 UTC 15 May, the trough at 850 hPa was very weak, but apparent at lower levels in a weakened form (Fig. 16). Winds turned more southerly at all levels, especially over Oregon, resulting in weakened downslope flow over the western Cascade slopes. A tongue of warm air extended northward over western Oregon and Washington. The WCTT weakened further at all levels at 1800 UTC. Westerly onshore flow had commenced along the coast, bringing an influx of cool, marine air over western Oregon. By 0000 UTC 16 May, the WCTT had redeveloped east of the Cascades at all three levels, with westerly flow moving cool air to the crest of the Cascade Mountains. A coastal pressure ridge had developed along the coast and the warmest air then resided east of the Cascade crest at all three levels.

Fig. 16.
Fig. 16.

Heights (m, contours), temperatures (°C, shading), and wind (full barb = 5 m s−1) at (top) 850 and (middle) 925 hPa; and (bottom) sea level pressure (hPa), 1000-hPa temperature (°C), and 10-m wind (m s−1) all using the 12-km WRF Model.

Citation: Monthly Weather Review 141, 8; 10.1175/MWR-D-12-00305.1

Figure 17 shows that between 0600–1200 UTC 15 May temperatures cooled at both sea level and 850 hPa, with the exception of western Washington, over the ocean, and at 850 hPa above the Columbia basin, where residual offshore flow continued. This warming was likely a combination of downslope flow and warm air advection. The approach of an area of low pressure offshore was evident as heights/pressure dropped over the northwest area of the domain. Over the next 6 h (1200–1800 UTC), the developing coastal pressure ridge resulted in rising heights/pressure and falling temperatures along the Oregon coast and areas offshore, while inland temperatures warmed because of surface heating. By the afternoon hours (1800–0000 UTC), sea level pressure and 850 hPa heights rose over the coastal region while temperatures rose and pressure/heights fell east of the Cascade crest, thus strengthening the inland WCTT.

Fig. 17.
Fig. 17.

(top) 6-h changes of height (m, contours) and temperature (°C, shading) at 850 hPa and (bottom) 6-h changes in sea level pressure (hPa) and 1000-hPa temperature (°C).

Citation: Monthly Weather Review 141, 8; 10.1175/MWR-D-12-00305.1

An important question is the origin of the WCTT on the east side of the Cascades. An analysis of the thermodynamic energy equation along 43°N for the hours leading up to WCTT formation east of the Cascade crest is shown in Fig. 18. This analysis was done in the same way as the one described earlier. At 1800 UTC 15 May, the theta tendency term shows substantial warming over the Oregon Plateau that is dominated by the diabatic term, while over the western slopes diabatic heating was opposed by strong horizontal advection of cool ocean air. During the early afternoon (2100 UTC), the east side continued to warm, though at a lesser rate, predominantly because of diabatic heating supported by horizontal advection of warmer air from the south. Over the western portion of the domain strong onshore advection of cool air resulted in to cooling from the Cascade crest westward. Finally, by late afternoon (0000 UTC) strong cold air advection resulted in cooling over the western two-thirds of the cross section.

Fig. 18.
Fig. 18.

Vertical cross sections of terms in the thermodynamic energy equation for the 12-km WRF Model simulation. These cross sections are along 43°N from 122.5° to 118.5°W for the afternoon of 15 May 2007.

Citation: Monthly Weather Review 141, 8; 10.1175/MWR-D-12-00305.1

Figure 19 shows terms of the thermodynamic energy equation at 850 hPa in plan view for the same time period. Areas west of the Cascades generally cooled through this period, with both horizontal and vertical advection contributing. There is diabatic warming over eastern Oregon and Washington, with a small contribution from horizontal advection. Thus, formation of the WCTT on the east side is predominantly the result of strong diabatic heating east of the Cascades, with cool air advection causing increasing pressure to its west.

Fig. 19.
Fig. 19.

The terms of the thermodynamic energy equation at 850 hPa at the same time as in Fig. 17.

Citation: Monthly Weather Review 141, 8; 10.1175/MWR-D-12-00305.1

4. Sensitivity experiments

Several experiments were conducted to determine the impact of terrain and surface fluxes on WCTT structure and evolution. First, the control run (described in the last section) was compared to three additional simulations, each with terrain modifications shown in Fig. 20. For the first experiment (smooth), all terrain encompassed by line L1 was reduced by 25%. To avoid producing a cliff along L1, all the terrain inside of line L2 was smoothed 10 times using a five-point smoother. This smoother averaged each grid point with the four surrounding points, giving equal weight to each. In the next experiment (no-WC-terrain), all the terrain inside of L1 was reduced to 1 m. Then, to smooth the cliff created along L1, all the terrain inside of L2 was smoothed 10 times using the same smoother. In the third terrain experiment (no-terrain), all terrain in the domain was reduced to 1 m.

Fig. 20.
Fig. 20.

Topography (m) for the terrain experiments from the 36-km model.

Citation: Monthly Weather Review 141, 8; 10.1175/MWR-D-12-00305.1

In a separate experiment, surface heat and moisture fluxes in the WRF Model were reduced to zero (using the original terrain). Two runs were performed for this experiment, one initialized at 0000 UTC 13 May and the other at 0000 UTC 15 May, to capture the development of the WCTT on both the west side and east side of the Cascades. These runs were initialized similarly to the control, using the GFS (for three-dimensional fields) and NAM (for surface temperature) fields. This is obviously an experimental limitation since the GFS has terrain and surface fluxes, but one mitigated by the distance of the model boundaries.

For the smooth and no-WC-terrain experiments, there was almost no change in the 500-hPa height field compared to the control throughout the case (Fig. 21). However, for the no-terrain experiment, the 500-hPa trough on the eastern edge of the domain had a tighter gradient and was more amplified.

Fig. 21.
Fig. 21.

500-hPa heights (m) and wind (full barb = 5 m s−1) for the control, smooth, no-WC-terrain, and no-terrain runs.

Citation: Monthly Weather Review 141, 8; 10.1175/MWR-D-12-00305.1

Figure 22 compares the terrain experiments to the control at 925 hPa. Consider first the control and smooth runs. At 1200 UTC 14 May, the thermal low over the southwest United States and the area of high pressure over Canada were centered over the same areas in both the control and smooth runs, and their 925-hPa height values were very similar. There was a tongue of low-level warm air in the smooth run that extended into Oregon that was not as strong or extended as far north as in the control. The WCTT in the smooth run was broader than in the control, likely because the slope in the smooth run was more gradual.

Fig. 22.
Fig. 22.

925-hPa heights (m, contours), temperature (°C, color shading), and wind (full barb = 5 m s−1).

Citation: Monthly Weather Review 141, 8; 10.1175/MWR-D-12-00305.1

At 2100 UTC 14 May, the WCTT was weaker in the smooth run than in the control. Finally, at 0600 UTC 15 May, as heights rose over the east side of the domain, the WCTT strengthened in the smooth run, though it still did not extend as far north as the control. By reducing the terrain height and smoothing it, there was less downslope warming, especially over the Washington Cascades, thus a weaker WCTT was produced in this region.

The no-WC-terrain experiment, in which the coastal terrain was removed, had a substantially different mesoscale evolution than the control. At 1200 UTC 14 May, when there was a strong WCTT in the control, there was no WCTT in the no-WC-terrain experiment, nor was there a tongue of warm air extending into Oregon. Rather, high pressure built over the northwest United States. A WCTT-like feature began to form on the slope along L1 as high pressure built east of the Rockies late on 14 May, thus driving downslope flow over the modified slopes.

For the no-terrain experiment, there was no WCTT in any location. A deeper trough at 500 hPa and a stronger area of low pressure near the surface was found over the northeast corner of the domain, while high pressure built over the remaining region for the duration of the run. Without terrain, mesoscale troughing did not develop anywhere in the domain, and warm air did not extend northward. There was still offshore flow in Oregon and Northern California due to the synoptic circulation around the high. This suggests that for a typical thermal trough case, the origin of the easterly flow that gives rise to downslope flow is from the synoptic circulation associated with low-level high pressure to the north and the thermal low over the southwest United States.

Figure 23 shows the first portion of an experiment in which heat and moisture fluxes were set to zero in the model. This run was initialized at 0000 UTC 13 May to capture the development of the WCTT west of the Cascades. There was little change in the synoptics at upper (not shown) or lower levels. Without solar heating, the thermal low over the Great Basin did not amplify during the day and low-level temperatures cooled over the Desert Southwest. The coastal thermal trough appears to be almost unchanged compared to the control, suggesting that solar heating is not important for WCTT initiation and development.

Fig. 23.
Fig. 23.

925-hPa heights (m, contours), temperature (°C, color shading), and wind (full barb = 5 m s−1) from the 36-km WRF Model.

Citation: Monthly Weather Review 141, 8; 10.1175/MWR-D-12-00305.1

Heat and moisture fluxes appear to be much more important for WCTT development on the east side of the Cascades (Fig. 24). In a model run without surface fluxes and initialized at 0000 UTC 15 May, the WCTT was substantially attenuated when it moved to the east side of the Cascades on the afternoon of 15 May. The weakened low pressure trough appeared to represent the interface between the coastal ridge to the west and synoptic high pressure to the east.

Fig. 24.
Fig. 24.

850-hPa heights (m, contours), temperature (°C, color shading), and wind (full barb = 5 m s−1).

Citation: Monthly Weather Review 141, 8; 10.1175/MWR-D-12-00305.1

5. Discussion and conclusions

Using the WRF Model, a West Coast thermal trough (WCTT) case on 13–16 May 2007 is analyzed in order to describe its evolution and origin. Because this event is similar to the composite event described in BMP12, the results from this study are valuable in understanding WCTTs in general.

The event began as high pressure associated with a ridge aloft extended over the northwest United States and British Columbia. As high pressure built inland, winds turned from northerly to easterly, resulting in downslope flow over the western slopes of the Cascade Mountains. This downslope flow produced adiabatic warming, which led to horizontal divergence aloft and subsequent hydrostatic reduction in pressure, thus leading to the development of the WCTT along the U.S. West Coast on the morning of the 14 May. The WCTT developed a westward tilt with height as warm air was advected westward in the lower troposphere, while near the surface, cool ocean temperatures kept the WCTT axis inland.

The ridge aloft and the surface high pressure associated with it continued eastward, resulting in a loss of downslope flow on the western slopes of the Cascade Mountains. As the interior heated on the afternoon of 15 May, cool air rushed inland, resulting in a marine push west and over the Cascades crest. While low-level pressure/heights rose on the west side of the Cascades from this cooling, pressure/heights fell on the east side due to the strong surface heating in that region. Under these conditions, the WCTT transitioned to the east side of the Cascades that afternoon.

An analysis of the thermodynamic energy equation on the morning of 14 May, just before the WCTT extended along the U.S. West Coast, found that vertical advection of potential temperature due to downslope flow was dominant in producing the warming that led to WCTT formation. Horizontal advection spread the adiabatically warmed air westward, thus broadening the trough to the west. Diabatic effects were minimal during the initiation stage.

An analysis of the thermodynamic energy equation was also performed for the 6 h of WCTT intensification on the east side of the Cascades on the afternoon of 15 May. In contrast to the west-side results, the diabatic term was the major contributor to the warming that led to the intensification of the WCTT on the east side. Vertical advection did not contribute to warming, and horizontal advection produced cooling as marine air pushed inland. There was some horizontal warm advection by southerly flow east of the Cascades that contributed to height falls.

When heat and moisture fluxes were removed from the WRF Model, the WCTT developed on the west side as intensely as in the control run. In fact, the pressure gradient across the Cascade Mountains was stronger in the run without surface fluxes since diabatic heating in the control run led to height falls during the day on the east side.

In another run, heat and moisture fluxes were removed during the period in which the observed trough shifted eastward. The WCTT did move to the east side of the Cascades in this simulation, but the area of low pressure was substantially weaker. Therefore, the WCTT on the east side was mainly a diabatic forced low, with areas of synoptic high pressure on either side.

Other sensitivity experiments were performed to explore the role of terrain in the formation of the WCTT. In an experiment where the U.S. West Coast terrain was reduced by 25% and smoothed, the WCTT did form, but in a weakened state. In another experiment, the West Coast terrain was completely removed, though the Rockies were retained. Because this created a large slope, the altered terrain resulted in the formation of an inverted trough along the western slopes of the Rockies. In the last experiment, all terrain was removed from the model. As a result, no inverted pressure trough formed in the domain. High pressure built over the northwest United States, with flow around the area of high pressure resulting in easterlies from Oregon southward.

An important aspect of the WCTT is its impact on wildfires when it moves across the Cascades and reforms on the east side of the Cascades. When WCTTs pass over the Cascades, existing wildfires have often been known to experience rapid growth (J. Saltenberger, Northwest Interagency Coordination Center, 2012, personal communication). It has been found that when WCTTs move past the Cascade crest, a strong area of convergence and rising motion often forms, coinciding with the trough axis of the WCTT. This phenomenon will be addressed in a future paper.

Acknowledgments

This research was supported by the USDA Forest Service through Agreement PNW-10-JV-11261987-033. We wish to thank Mr. John Saltenberger and Ms. Julia Ruthford for helpful conversations regarding the fire impacts of thermal troughs, and Jeff Baars and Rick Steed for support in running the WRF Model and postprocessing.

REFERENCES

  • Brewer, M., , C. Mass, , and B. Potter, 2012: The West Coast thermal trough: Climatology and synoptic evolution. Mon. Wea. Rev., 140, 38203843.

    • Search Google Scholar
    • Export Citation
  • Chien, F. C., , C. F. Mass, , and Y. H. Kuo, 1997: Interaction of a warm-season frontal system with the coastal mountains of the western United States. Part I: Prefrontal onshore push, coastal ridging, and alongshore southerlies. Mon. Wea. Rev., 125, 17051729.

    • Search Google Scholar
    • Export Citation
  • Colson, D., 1956: Meteorological problems associated with mass fires. Fire Control Notes, 17, 911.

  • Cramer, O. P., 1954: Recognizing weather conditions that affect forest fire behavior. Fire Control Notes, 15, 16.

  • Cramer, O. P., 1957: Frequency of dry east winds over northwest Oregon and southwest Washington. U.S. Forest Service, Pacific Northwest Forest and Range Experiment Station, Research Paper 24, 19 pp.

  • Gisborne, H. T., 1927: Meteorological factors in the Quartz Creek forest fire. Mon. Wea. Rev., 55, 5660.

  • Glickman, T., Ed., 2000: Glossary of Meteorology. 2nd ed. Amer. Meteor. Soc., 855 pp.

  • Jannuzzi, J. A., 1993: The onshore push of marine air into the Pacific Northwest. Wea. Forecasting, 8, 194202.

  • Mass, C. F., , and N. A. Bond, 1996: Coastally trapped wind reversals along the United States West Coast during the warm season. Part II: Synoptic evolution. Mon. Wea. Rev., 124, 446461.

    • Search Google Scholar
    • Export Citation
  • Mass, C. F., , and W. J. Steenburgh, 2000: An observational and numerical study of an orographically trapped wind reversal along the west coast of the United States. Mon. Wea. Rev., 128, 23632397.

    • Search Google Scholar
    • Export Citation
  • Mass, C. F., , M. D. Albright, , and D. J. Brees, 1986: The onshore surge of marine air into the Pacific Northwest: A coastal region of complex terrain. Mon. Wea. Rev., 114, 26022627.

    • Search Google Scholar
    • Export Citation
  • McKendry, I. G., 1994: Synoptic circulation and summertime ground-level ozone concentrations at Vancouver, British Columbia. J. Appl. Meteor., 33, 627641.

    • Search Google Scholar
    • Export Citation
  • Nuss, W. A., 2007: Synoptic-scale structure and the character of coastally trapped wind reversals. Mon. Wea. Rev., 135, 6081.

  • Nuss, W. A., and Coauthors, 2000: Coastally trapped wind reversals: Progress toward understanding. Bull. Amer. Meteor. Soc., 81, 719743.

    • Search Google Scholar
    • Export Citation
  • Werth, P., , and R. Ochoa, 1993: The evaluation of Idaho wildfire growth using the Haines index. Wea. Forecasting, 8, 223234.

Save