1. Introduction
Strong nocturnal low-level-wind jets have been documented at several locations worldwide at canyon exits where a valley or canyon ends abruptly at a plain (Whiteman 1990; Zardi and Whiteman 2012). Valley-exit jets are known to be of two types (Banta et al. 1995): those produced by the channeling of moderate to strong background winds from aloft down a valley axis (dynamically driven exit jets) and those produced by the cascade of cold air out the valley exit fed by nocturnal down-valley flows inside the canyon (thermally driven exit jets). Thermally driven exit jets, the focus of this paper, occur most readily under the relatively quiescent large-scale weather conditions that produce diurnal mountain circulations. Despite anecdotal descriptions of the regular occurrence of these flows in many mountain regions, relatively little scientific attention has been focused on thermally driven canyon exit jets (hereinafter referred to simply as canyon or valley-exit jets), even though the scientific literature on thermally driven winds inside mountain valleys is quite extensive [see recent reviews by Zardi and Whiteman (2012) and Poulos and Zhong (2008)].
This study describes the exit jet at Weber Canyon, Utah. Canyon exit jets have gained attention recently because of their potential for harnessing wind energy from the jets through the use of wind turbines. Utah’s first wind farm, the 19-MW Spanish Fork Canyon wind project, which is located 120 km south of Weber Canyon, uses canyon exit-jet winds to generate electricity that is sold to a local electrical utility.
The goals of the Weber Canyon experiments were to gain basic observational evidence on the temporal and spatial characteristics of the exit jet. Of particular interest were seasonal and diurnal variations in the strength and structure of the exit jet and the spatial evolution of vertical wind and temperature profiles in the canyon exit region.
The paper is organized as follows. In section 2 we provide a review of the scientific literature on valley-exit jets and related gap flows. Section 3 presents the experimental setup for the Weber Canyon experiments, including a description of the experimental goals, the terrain, and the long- and short-term meteorological instrumentation assets. Section 4 presents the results. This is followed by a discussion in section 5 and conclusions and recommendations for future research in section 6.
2. Literature review of valley-exit jets and gap flows
a. Valley-exit jets
Thermally driven low-level jets at the exits of valleys and canyons have been documented at only a few locations around the world. The most detailed observational study appears to be that of Stilke (1984) and Pamperin and Stilke (1985), who observed the exit jet of the Inn Valley as it enters the Rosenheim Basin near Thalreit, Germany, during the Mesoscale Experiment in the Region Kufstein-Rosenheim (MERKUR) in March and April of 1982. The exit jet formed where the weak but deep down-valley flow in the Inn Valley made a transition to a shallower layer at and beyond the exit, where the flow accelerated. The nocturnal low-level jet strengthened during the night and attained a maximum speed of 15 m s−1 at a height of around 200 m. Following sunrise the jet rapidly degraded and by 0815 local time the jet profile had almost completely disappeared. Zängl (2004) used a numerical model to gain further understanding of the Inn Valley jet, finding that just beyond the valley constriction at the exit, the flow accelerated and formed a pronounced low-level jet that maintained its structure several tens of kilometers into the foreland. In the United States, Banta et al. (1995) observed a low-level jet at the exit of Eldorado Canyon in Colorado during the wintertime but noted that the jet was present for only a few nighttime hours. The jet had a complex vertical structure with peak speeds greater than 6 m s−1 at a height of around 600–700 m. Darby and Banta (2006) used a scanning Doppler lidar to observe nocturnal low-level jets at the exits of several canyons that drain from the Wasatch Mountains into the Salt Lake Valley during the Vertical Transport and Mixing Experiment (VTMX) of October 2000. They found that the canyon outflows were sensitive to the strength of the down-valley flows in the Salt Lake Valley and the strength of the synoptic-scale flows above the valleys. When these flows strengthened, the canyon outflows decreased.
b. Gap flows
Several analogies can be drawn between small-scale thermally driven, diurnally varying valley-exit jets and larger-scale synoptically driven gap winds. Both are a result of pressure gradients that form between two air masses that are separated by gap openings such as channels, valleys, or mountain passes. Studies of gap winds have generally been focused on two areas: 1) gap winds through channels such as in the Strait of Juan de Fuca south of Vancouver Island (Overland and Walter 1981, Mass et al. 1995), in the Shelikof Strait in Alaska (Lackmann and Overland 1989), or in the Columbia River Gorge (Sharp and Mass 2002, 2004) and 2) gap winds through elevated mountain passes such as those described by Mayr et al. (2007). Overland and Walter (1981) show that the airflow magnitudes in the Strait of Juan de Fuca can be explained by the along-valley pressure gradient that exists between the high pressure region over British Columbia, Canada, and a low pressure system on the Washington coast and that the strongest winds are found beyond the exit of the gap and not within the narrowest region within the gap. Sharp and Mass (2004) found similar characteristics at the exit of the Columbia River gorge and argue that this finding indicates that Venturi effects cannot explain the strong gap winds. Rather, it is the along-valley pressure gradient that maximizes at the exit of the gap that produces the strongest winds downstream of the gap. As we shall see, observations at Weber Canyon show similar characteristics where the strongest winds were observed beyond the canyon exit and not within the narrow part of the canyon upstream of the exit. In this sense, exit jets at Weber Canyon are like gap winds embedded within a valley flow. Lackmann and Overland (1989) completed a momentum balance for a gap flow event in the Shelikof Strait in Alaska and found that the flow can be described as a balance among the pressure gradient force, inertia, entrainment, and friction.
Differences between gap winds and thermally driven valley-exit winds have to do with the spatial scales of the wind systems, the driving mechanisms, and the climatological characteristics. Thermally driven valley-exit jets occur on a much smaller scale and are caused by diurnally varying temperature contrasts between a mountain valley and a plain due to differences in radiative heating and cooling. They are generally nocturnal and occur regularly from day to day when synoptic weather conditions are rather quiescent. In contrast, traditional gap winds are larger in scale, and may occur at any time of day when strong synoptic-scale pressure gradients develop across the gap. The low-level structure of the jet and its spatial extent are not well known for either gap winds or valley-exit winds.
3. Topography, observations, and synoptic conditions
The Wasatch Mountains of Utah, which run from the northern Utah border south into central Utah, form the border between the Rocky Mountains to the east and the Great Basin farther west. The Great Basin contains isolated mountain ranges that are separated by broad basins (i.e., basin-and-range topography). Weber Canyon flows westward out of the Wasatch Mountains into the adjacent Great Salt Lake basin (GSLB) as shown in Fig. 1a. Weber Canyon is typical of other Wasatch canyons that narrow significantly as they cut through a final ridge before flowing westward out of the Wasatch Mountains. Most canyons are fed by upstream drainage basins that originate in higher mountainous terrain. The canyons thus provide a path where localized mountain wind circulations flow between the canyon drainage basin and the GSLB. In the case of Weber Canyon, a subbasin (the Morgan Basin) is present immediately east of the narrow 10-km lower section of the canyon.
Shaded relief map of (a) Weber Canyon and (b) Lower Weber Canyon. Measurement locations are shown as diamonds.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
Shaded relief map of (a) Weber Canyon and (b) Lower Weber Canyon. Measurement locations are shown as diamonds.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
Shaded relief map of (a) Weber Canyon and (b) Lower Weber Canyon. Measurement locations are shown as diamonds.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
A measurement campaign was designed to gather data to elucidate the seasonal, diurnal, and spatial characteristics of winds at the exit of Weber Canyon as well as the mechanisms responsible for producing the strong exit jet. The scope and budget of the experiment did not allow for a full suite of meteorological instruments. Thus, results are somewhat limited by available resources. The locations of measurement equipment are provided in Fig. 1b. Long-term data came from two sources: 1) 10 years of data from a surface meteorological station (INPWR) at a hydropower plant 1.7 km upstream from the Weber Canyon exit and operated by the Utah Department of Transportation and 2) 12 months of wind data at the exit region from a 50-m-tall meteorological tower (EXTTWR) operated by the State of Utah anemometer loan program. These long-term data were supplemented by data from a 3-month field study, which provided additional high-resolution measurements. A pulsed Doppler sonic detection and ranging (sodar) instrument (EXTSDR) was operated near the exit of Weber Canyon adjacent to the EXTTWR site during the 3-month general observation period (GOP) of July–September 2010 to continuously monitor wind profiles at the canyon exit (Fig. 1b). A pulsed Doppler WINDCUBE lidar measured wind speed profiles within the canyon 5 km upstream from the canyon exit (INLDR) for 3 weeks during the GOP. The WINDCUBE was chosen for its narrow beam geometry and its ability to capture wind profiles continuously within the canyon without interference from canyon walls and other flow obstructions. Additionally, four overnight intensive observational periods (IOPs) were conducted during the GOP in late September and early October 2010 using a tethered balloon sounding system located inside the canyon (INSONDE) adjacent to the INLDR. Rawinsondes were launched from the canyon exit region (EXTSONDE) at a site adjacent to EXTTWR. The purpose of these additional measurements was to observe how wind and temperature profiles evolved from within the canyon to the exit during typical events. A novel new wind profiling system (ValidWind) developed at Utah State University (Wilkerson et al. 2012) that automatically tracks pilot balloons using a laser range finder was tested during several IOPs inside the canyon (INVW) in the vicinity of the INSONDE and INLDR. During IOP4 this instrument collected wind profiles continuously during the entire night. The Ogden Peak (OGP) station provided measurements of ridgetop wind conditions and Hill Air Force Base (HIF) provided an indicator of exit-jet flow penetration into the GLSB (Fig. 1a). Figure 2 provides a timeline of the measurements, and Table 1 shows additional information about the IOPs.
Measurements timeline. INPWR, the 10-m surface reference station, is not shown but provided a long-term 10-yr data record from 2000 to 2010.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
Measurements timeline. INPWR, the 10-m surface reference station, is not shown but provided a long-term 10-yr data record from 2000 to 2010.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
Measurements timeline. INPWR, the 10-m surface reference station, is not shown but provided a long-term 10-yr data record from 2000 to 2010.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
The IOPs.
IOP dates were chosen on the basis of weather forecasts and knowledge obtained about flow development from previous nights during the GOP. The weather conditions sought were high pressure, clear skies, weak synoptic pressure gradients, and light background winds, as these conditions are conducive to the development of localized thermally driven flows. Synoptic conditions for IOPs 1–4 are visualized in Fig. 3 using 700-hPa upper-air analysis charts from the National Climatic Data Center. IOPs 2–4 exhibited similar synoptic characteristics, with weak regional pressure gradients over Utah and light winds aloft. IOP1, however, exhibited a stronger pressure gradient that was oriented across Utah, producing moderate upper-level winds. Well-developed outflow jets were observed at the exit of Weber Canyon on all four IOP nights, and clear skies also prevailed during the four IOPs.
The 700-hPa upper-air analysis charts for IOPs 1–4. The 1200 UTC (0500 MST) charts correspond to early morning of the second day for each IOP during 2010: (a) IOP1, 19 Sep; (b) IOP2, 25 Sep; (c) IOP3, 29 Sep; and (d) IOP4, 1 Oct.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
The 700-hPa upper-air analysis charts for IOPs 1–4. The 1200 UTC (0500 MST) charts correspond to early morning of the second day for each IOP during 2010: (a) IOP1, 19 Sep; (b) IOP2, 25 Sep; (c) IOP3, 29 Sep; and (d) IOP4, 1 Oct.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
The 700-hPa upper-air analysis charts for IOPs 1–4. The 1200 UTC (0500 MST) charts correspond to early morning of the second day for each IOP during 2010: (a) IOP1, 19 Sep; (b) IOP2, 25 Sep; (c) IOP3, 29 Sep; and (d) IOP4, 1 Oct.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
4. Results
This section is focused on gaining an understanding of the temporal and spatial characteristics of the valley-exit jet at Weber Canyon. Long-term and seasonal aspects will be discussed first, followed by a discussion of shorter-term and spatial aspects of the phenomenon.
a. Diurnal and seasonal flow patterns
Hourly and daily variations of the wind speed (positive for down-valley flow and negative for up-valley flow) at the exit of Weber Canyon (EXTTWR) are displayed for the October 2009–October 2010 period in Fig. 4, illustrating the diurnal and seasonal variations in exit region wind strength. Down-valley winds are strongest and most consistent at night during the summer and fall. Up-valley winds occur predominantly during daytime, but are weaker, especially in midafternoon. Down-canyon flow duration varies seasonally and peaks in winter and fall. The initiation and cessation of the down-valley winds, and thus their duration, are closely tied to the times of astronomical sunrise and sunset, which vary seasonally. At Weber Canyon, down-valley flows begin 1–3 h after sunset and continue 5–6 h after sunrise regardless of season. Up-canyon flow is strongest and most consistent during the daytime in spring. Because of the atypical nature of katabatic and anabatic flow duration at the exit of Weber Canyon, for many of the subsequent analyses wind speeds have been separated into periods when either katabatic or anabatic flow was expected based on the seasonal pattern observed in Fig. 4 and discussed previously.
Hourly and daily courses of the maximum 10-min wind speeds (m s−1) for each hour for a 1-yr period at the 50-m level of the meteorological tower at the exit of Weber Canyon (EXTTWR). Positive (negative) wind speeds denote down- (up-) canyon flow. Dashed lines indicate astronomical sunrise (bottom) and sunset (top) times at this location.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
Hourly and daily courses of the maximum 10-min wind speeds (m s−1) for each hour for a 1-yr period at the 50-m level of the meteorological tower at the exit of Weber Canyon (EXTTWR). Positive (negative) wind speeds denote down- (up-) canyon flow. Dashed lines indicate astronomical sunrise (bottom) and sunset (top) times at this location.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
Hourly and daily courses of the maximum 10-min wind speeds (m s−1) for each hour for a 1-yr period at the 50-m level of the meteorological tower at the exit of Weber Canyon (EXTTWR). Positive (negative) wind speeds denote down- (up-) canyon flow. Dashed lines indicate astronomical sunrise (bottom) and sunset (top) times at this location.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
Frequency distributions of wind speed and direction at the EXTTWR and INPWR stations have been built (Fig. 5) using 10-min-average data over the 1-yr period of October 2009–October 2010. At the exit of Weber Canyon (EXTTWR) wind speed frequencies exhibit a bimodal peak (Fig. 5a) during katabatic flow periods. The lower-speed peak at 2 m s−1 is indicative of daytime wind speeds that occur around morning and evening transition, and the high-speed peak at 11 m s−1 is indicative of stronger easterly nighttime and morning flows (exit jets) coming out of the Weber Canyon exit. Wind speeds at the exit during anabatic flow periods exhibit a peak around 3 m s−1. Inside the canyon (INPWR) wind speed magnitudes are strongest during anabatic flow and are weaker during katabatic flow periods (Fig. 5b). The prevailing winds at the exit of Weber Canyon, as determined from the 50-m level at EXTTWR with a 1-yr period of record, are easterly (Fig. 5c). The strong easterly, down-canyon flows occur almost 60% of the time and predominantly during nighttime. Westerly up-canyon flow occurs mostly during daytime but is much weaker and variable in direction. Winds inside the canyon, over the concurrent 1-yr period, as measured at the 10-m INPWR station are also predominantly oriented in the down-canyon direction (Fig. 5d) but exhibit a more pronounced up-canyon flow component during the day.
(a) Wind speed frequency distribution at 50 m on EXTTWR; (b) wind speed frequency distribution at 10 m on INPWR; (c) wind rose showing the joint frequency distribution of wind directions and speeds at the 50-m level of EXTTWR, combined katabatic and anabatic periods; and (d) wind rose showing the joint frequency distribution of wind directions and speeds at the 10-m level of INPWR, combined katabatic and anabatic periods for the same 1-yr period shown in Fig. 4. In (a) and (b), wind speeds are divided into periods when either katabatic or anabatic flow conditions were expected based on seasonal exit-jet patterns.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
(a) Wind speed frequency distribution at 50 m on EXTTWR; (b) wind speed frequency distribution at 10 m on INPWR; (c) wind rose showing the joint frequency distribution of wind directions and speeds at the 50-m level of EXTTWR, combined katabatic and anabatic periods; and (d) wind rose showing the joint frequency distribution of wind directions and speeds at the 10-m level of INPWR, combined katabatic and anabatic periods for the same 1-yr period shown in Fig. 4. In (a) and (b), wind speeds are divided into periods when either katabatic or anabatic flow conditions were expected based on seasonal exit-jet patterns.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
(a) Wind speed frequency distribution at 50 m on EXTTWR; (b) wind speed frequency distribution at 10 m on INPWR; (c) wind rose showing the joint frequency distribution of wind directions and speeds at the 50-m level of EXTTWR, combined katabatic and anabatic periods; and (d) wind rose showing the joint frequency distribution of wind directions and speeds at the 10-m level of INPWR, combined katabatic and anabatic periods for the same 1-yr period shown in Fig. 4. In (a) and (b), wind speeds are divided into periods when either katabatic or anabatic flow conditions were expected based on seasonal exit-jet patterns.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
Monthly average wind speeds measured by the anemometers at different levels (10, 30, and 50 m) at EXTTWR during the 1-yr period from October 2009 through September 2010 are shown in Fig. 6. Wind speeds are separated into katabatic and anabatic periods and averaged for each month. For the 1-yr period shown the average wind speeds range from 3 to 12 m s−1, with the strongest katabatic flow in summer and early fall, peaking in September, and the strongest anabatic flow in June. As we will see later in this section, the peak in summer and fall katabatic flow at the exit is consistent with the peak in frequency of thermally driven katabatic valley winds within the canyon as determined for a 10-yr period of record. An interesting feature in Fig. 6 is the tendency for positive wind shear between the 30- and 50-m tower levels during summer months and negative wind shear during winter months of January and February, suggesting that the height of the wind speed maximum varies seasonally such that during the winter the jet maximum is contained within the tower layer and during the summer the jet maximum is elevated above the tower.
Monthly average wind speed measured at the 50-m meteorological tower (EXTTWR) during October 2009–October 2010. Average wind speeds are divided into periods when either katabatic or anabatic flow conditions were expected based on seasonal exit-jet patterns.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
Monthly average wind speed measured at the 50-m meteorological tower (EXTTWR) during October 2009–October 2010. Average wind speeds are divided into periods when either katabatic or anabatic flow conditions were expected based on seasonal exit-jet patterns.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
Monthly average wind speed measured at the 50-m meteorological tower (EXTTWR) during October 2009–October 2010. Average wind speeds are divided into periods when either katabatic or anabatic flow conditions were expected based on seasonal exit-jet patterns.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
The INPWR site provided 10 years of anemometer data from a 10-m tower from which average monthly and interannual variations in flow strength within the lower Weber Canyon could be analyzed. Mean monthly wind speeds are shown for each of the 10 years in Fig. 7. Wind speeds are generally consistent from year to year and display similar seasonal trends, with the strongest flows during the summer and early fall. This agrees well with measurements obtained on the meteorological tower at the valley exit during 2010 when flow strength also peaked in September. Overall, the spread of the data is small, especially during the summer months. Thus, there is a high consistency in summertime thermally driven flows from year to year. Data quality issues were found in the 10-yr dataset, reducing the number of available observations in some of the months and leading to some of the scatter in the monthly average wind speeds in a few of the years in Fig. 7 such as 2003.
Monthly average wind speeds at the INPWR surface weather station within Weber Canyon over a 10-yr period of record from 2000 through 2010.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
Monthly average wind speeds at the INPWR surface weather station within Weber Canyon over a 10-yr period of record from 2000 through 2010.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
Monthly average wind speeds at the INPWR surface weather station within Weber Canyon over a 10-yr period of record from 2000 through 2010.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
b. Case studies
Overall, the conditions during the 3-month GOP were conducive to down-canyon flow development, with 75 of 90 nights (83%) displaying exit-jet outflow as measured by EXTSDR. The only times when canyon flows were completely absent were during the passage of low pressure systems or frontal zones associated with increased cloud cover and precipitation. Because IOPs 2–4 were conducted under synoptically undisturbed conditions and the temperature and wind structures evolved similarly during all three IOPs, we illustrate the weak synoptic flow evolution case with IOP4, which had the most ValidWind data (see Table 1). IOP1 is then used to illustrate the evolution of the vertical temperature and wind structure under moderate synoptic background flows.
Wind profiles within the canyon from the INVW system and at the canyon exit from the EXTSONDE balloon launches are shown for IOP4 in Fig. 8. To obtain the wind profiles from the 1.5-s-resolution GPS track of the rawinsondes released from the EXTSONDE site, a 13-point smoothing function was applied to the subpoint trajectories, with a subsequent 5-point smoothing function applied to the wind speed profiles. Within the canyon there was a relatively uniform flow layer that occupied the depth of the canyon, which is about 1000 m but decreases to around 400 m near the exit. At the exit region the flow was much shallower with a pronounced low-level-jet wind maximum. Potential temperature θ profiles measured using INSONDE and EXTSONDE during IOP4 are plotted together in Fig. 9 at several times throughout the night. Within the canyon at night a 20–40-m-deep mixed layer was present near the ground, with a strong stable layer above this extending to heights of ~100 m. Above this height the profile maintained weaker stability to around 400 m, the upper limit of the tethersonde soundings. Strong nighttime cooling also resulted in a shallow potential temperature deficit near the surface within the canyon. Potential temperatures were generally higher at the canyon exit than inside the canyon, with a strong stable layer present to around 200 m and near-neutral conditions above. The atmosphere within the canyon cooled continuously throughout the night, as shown in Fig. 9. In contrast, at the exit, cooling was limited to the positive shear zone below the jet maximum (i.e., below about 100 m) with negligible cooling above. The continuous cooling aloft within the canyon at night coupled with the negligible cooling aloft at the exit resulted in an increasing temperature difference between the canyon and the exit above 200 m. Thus, the horizontal temperature contrast between the elevated air in the canyon and the air at the exit, which leads to a down-canyon pressure gradient that produces the exit jet, increased with time during the night. Turbulence at the exit region is believed to be responsible for the mixing of warmer air from aloft, creating warmer conditions near the surface at the exit. Because flow tends to follow along isentropes or contours of potential temperature in the absence of diabatic effects (Cramer 1972), a diagnosis of flow motion can be made by comparing the height at which air parcels having equal potential temperature are located. With this assumption, Fig. 9 shows that air parcels that originate at higher elevations within the canyon are brought down closer to the surface at the exit.
Wind speed profiles (a) in the canyon measured with ValidWind (INVW) and (b) at the canyon exit measured with rawinsondes (EXTSONDE) during IOP4 on 30 Sep–1 Oct 2010.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
Wind speed profiles (a) in the canyon measured with ValidWind (INVW) and (b) at the canyon exit measured with rawinsondes (EXTSONDE) during IOP4 on 30 Sep–1 Oct 2010.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
Wind speed profiles (a) in the canyon measured with ValidWind (INVW) and (b) at the canyon exit measured with rawinsondes (EXTSONDE) during IOP4 on 30 Sep–1 Oct 2010.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
Potential temperature profiles in the canyon (dashed line, INSONDE) and at the exit (solid line, EXTSONDE) during IOP4 at (a) 1700, (b) 2000, (c) 2300, (d) 0200, (e) 0500, and (f) 0800 MST.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
Potential temperature profiles in the canyon (dashed line, INSONDE) and at the exit (solid line, EXTSONDE) during IOP4 at (a) 1700, (b) 2000, (c) 2300, (d) 0200, (e) 0500, and (f) 0800 MST.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
Potential temperature profiles in the canyon (dashed line, INSONDE) and at the exit (solid line, EXTSONDE) during IOP4 at (a) 1700, (b) 2000, (c) 2300, (d) 0200, (e) 0500, and (f) 0800 MST.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
Figure 10 shows time series of wind speed and direction at the 80-m level within and at the exit of the canyon during IOP4 and IOP1 from INLDR and EXTSDR observations, respectively. We chose 80 m because the jet maximum at the exit commonly occurred near this height. First, we will consider the evolution of the atmospheric structure during the synoptically undisturbed period of IOP4 (Figs. 10a and 10b). In the afternoon before sunset, the up-valley winds inside the canyon were stronger and less variable than those over the exit region at the same heights. At both locations, the IOP4 transition to down-canyon flow occurred shortly after sunset. Wind speeds increased to reach a nearly steady state during the night by 0000 Mountain Standard Time (MST), but with the strongest winds occurring just before sunrise at both locations. Following sunrise, the winds began to slow as the ground was heated by the sun, with winds becoming calm and then reversing to up valley around noon. Winds at the exit of the canyon were the first to undergo a morning wind direction reversal while winds within the canyon reversed less than 1 h later. During the day, anabatic winds are about half as strong at the canyon exit as inside the canyon. During the night, wind speeds at the exit are about 2.5 times as strong as inside the canyon.
Time series of wind speed and direction during (a),(b) IOP4 and (c),(d) IOP1. Shown are data from the 80-m level in Weber Canyon from lidar observations (dark lines) and at the canyon exit from sodar observations (lighter lines). Sunset and sunrise times (MST) are noted.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
Time series of wind speed and direction during (a),(b) IOP4 and (c),(d) IOP1. Shown are data from the 80-m level in Weber Canyon from lidar observations (dark lines) and at the canyon exit from sodar observations (lighter lines). Sunset and sunrise times (MST) are noted.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
Time series of wind speed and direction during (a),(b) IOP4 and (c),(d) IOP1. Shown are data from the 80-m level in Weber Canyon from lidar observations (dark lines) and at the canyon exit from sodar observations (lighter lines). Sunset and sunrise times (MST) are noted.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
In contrast to the undisturbed IOP4, IOP1 occurred during a period when southwest winds aloft were moderate and a synoptic-scale pressure gradient was present over the region (see Fig. 3). This background wind affected some aspects of the evolution of the valley-exit jet (Figs. 10c and 10d). The up-valley winds during the afternoon were stronger in the canyon than at the exit, in agreement with the undisturbed case. The reversal to down-valley winds occurred, as with the undisturbed case, nearly simultaneously at the canyon exit and in the canyon, but the reversal was somewhat earlier than for the undisturbed case. The exit-jet and down-valley flows approached a steady state earlier in the evening, but with lower speeds than for the undisturbed case. The valley exit-jet speed began to decrease well before sunrise and the morning transition to up-canyon flow at the exit occurred about 3 h sooner, possibly due to the strong aiding westerly wind component in the GSLB, whereas down-canyon winds remained quite strong within the canyon for another 1–2 h. During both IOPs, winds at the exit of the canyon were the first to undergo a morning wind direction reversal and winds within the canyon followed shortly thereafter.
Time–height cross sections of horizontal wind speed (from INLDR and EXTSDR observations) and potential temperature (from INSONDE and EXTSONDE observations) are used to understand how the vertical structure of the canyon atmosphere evolved during the two IOPs. In agreement with INVW profiles during IOP4, there was a deep uniform flow within the canyon (Fig. 11a). At the exit the flow descended and thinned, leading to strong vertical and horizontal accelerations. This descent brought potentially warmer air at the upper levels of the canyon down closer to the surface at the canyon exit (Fig. 11b). A very strong stable layer was present across the core of the jet, as the flow appeared to compress the isentropes.
Horizontal wind speed (shades of gray) and potential temperature (labeled contour lines) for IOP4 and IOP1 from (a),(c) sodar observations at EXTSDR and (b),(d) lidar observations at INLDR. No potential temperature observations were available for (c).
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
Horizontal wind speed (shades of gray) and potential temperature (labeled contour lines) for IOP4 and IOP1 from (a),(c) sodar observations at EXTSDR and (b),(d) lidar observations at INLDR. No potential temperature observations were available for (c).
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
Horizontal wind speed (shades of gray) and potential temperature (labeled contour lines) for IOP4 and IOP1 from (a),(c) sodar observations at EXTSDR and (b),(d) lidar observations at INLDR. No potential temperature observations were available for (c).
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
The flow at the canyon exit during IOP1 was confined to a shallower layer (Fig. 11c), the wind speeds were weakened, and the jet began to decay well before sunrise compared to the undisturbed case. Unfortunately, potential temperature profiles were unavailable at the canyon exit during IOP1. The potential temperatures in the canyon, however, were similar to those observed during IOP4 (Fig. 11d).
Potential temperature data support the conclusion that slower-moving air from aloft within the canyon descends along constant potential temperature surfaces and accelerates into a jet at the exit region of Weber Canyon. Vertical winds observed at the EXTSDR location show strong downward motions at the exit (Fig. 12), where downward winds are indicated by negative values and upward winds by positive values. Vertical wind speeds during IOP4 were −2.5 m s−1 (Fig. 12a), and vertical wind speeds were −1.5 m s−1 during IOP1 (Fig. 12b). The strong vertical winds form as a deep layer of air drains out of the canyon, thins, and accelerates downward into a shallower layer forming the jet. This converts potential energy from the deep in-canyon flow layer to kinetic energy at the exit as the flow speed increases.
Vertical wind speeds (m s−1) measured by EXTSDR at the exit of Weber Canyon during (a) IOP4 and (b) IOP1. Downward winds are indicated by negative values, and upward winds are indicated by positive values.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
Vertical wind speeds (m s−1) measured by EXTSDR at the exit of Weber Canyon during (a) IOP4 and (b) IOP1. Downward winds are indicated by negative values, and upward winds are indicated by positive values.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
Vertical wind speeds (m s−1) measured by EXTSDR at the exit of Weber Canyon during (a) IOP4 and (b) IOP1. Downward winds are indicated by negative values, and upward winds are indicated by positive values.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
The net vertical motion is caused both by flow descent and the downward slope of the canyon floor. An estimate of the vertical motion can be obtained by assuming the flow descends continuously along the terrain in the canyon and also descends rapidly along adiabats near the exit. The average slope of the terrain in the down-canyon direction in Weber Canyon is approximately 1.5%. With wind speeds on the order of 20 m s−1, this would produce a vertical wind component of 0.3 m s−1. As shown in Fig. 9 at 2000 MST, air from a level of ~400 m within the canyon is brought down to a level of ~100 m at the exit. Considering that the two locations are separated by a distance of ~5 km and that the mean flow motion is on the order of 20 m s−1, this would produce average vertical winds on the order of 1 m s−1. This approach, however, assumes that adiabatic descent is continuous over a horizontal distance of ~5 km and therefore is most likely a conservative estimate of the magnitude of vertical motion. Overall, the net effects of terrain slope and flow descent can partially explain the magnitude of the vertical winds measured at the exit region.
Concurrent measurements of wind speed within the canyon and at the canyon exit are plotted in the scatter diagram in Fig. 13 for the 3-week period that INLDR was available. Wind speeds are separated into anabatic (up canyon) flow and katabatic (down canyon) flow. Katabatic winds inside the canyon are nearly always greater than 2 m s−1, while winds at the canyon exit can become calm. Canyon exit speeds increase as in-canyon wind speeds increase, but the winds at the exit are much stronger than inside the canyon. When the katabatic winds reach a nighttime steady state, they are typically around 17 m s−1 at the exit while only 7 m s−1 inside the canyon at the same 80-m level, despite a horizontal separation of only a few kilometers. Anabatic winds at the 80-m height are weak at the valley exit (typically 2–3 m s−1) while taking on a broader range of speeds (2–10 m s−1) inside the canyon.
Concurrent measurements of wind speed at 80 m AGL in the canyon and at the canyon exit during (a) nighttime katabatic flow and (b) daytime anabatic flow. In-canyon measurements were made with lidar, while canyon-exit measurements were made with sodar. Data points are 10-min averages over the 14 Sep–6 Oct 2010 period for which there were concurrent measurements.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
Concurrent measurements of wind speed at 80 m AGL in the canyon and at the canyon exit during (a) nighttime katabatic flow and (b) daytime anabatic flow. In-canyon measurements were made with lidar, while canyon-exit measurements were made with sodar. Data points are 10-min averages over the 14 Sep–6 Oct 2010 period for which there were concurrent measurements.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
Concurrent measurements of wind speed at 80 m AGL in the canyon and at the canyon exit during (a) nighttime katabatic flow and (b) daytime anabatic flow. In-canyon measurements were made with lidar, while canyon-exit measurements were made with sodar. Data points are 10-min averages over the 14 Sep–6 Oct 2010 period for which there were concurrent measurements.
Citation: Journal of Applied Meteorology and Climatology 52, 5; 10.1175/JAMC-D-12-0221.1
5. Discussion
The coevolution of the down-canyon flow inside the canyon and the canyon-exit jet supports the conclusion that the exit jet is a local modification of the cold down-canyon flow as it exits from the canyon. This is supported by the covariation of the exit jet with down-valley winds inside the canyon, its regular occurrence on undisturbed nights when synoptic-scale influences are weak (as these are the conditions that favor the development of thermally driven mountain flows), and by its attainment of maximum strength in the summer and fall when the horizontal temperature contrasts along the canyon between the warm GSL basin and the cool air formed in the Morgan Basin at the canyon’s east end could be expected to be a maximum. The temperature contrast would produce a horizontal pressure gradient along the canyon axis that drives the canyon winds. The local modification of the down-canyon flow at the canyon exit would then represent the sinking, thinning, and acceleration of the cold, deep, and slow-moving air in the canyon as it reaches the canyon mouth, converting potential energy to kinetic energy.
Several interesting research questions arise that can be tested in further experiments. First, why do the down-valley flow and the exit jet in Weber Canyon continue so late into the morning following sunrise? Most down-valley flows that have been studied to date reverse within a few hours following sunrise. The exceptions are large valleys, such as Austria’s Inn Valley (Whiteman 2000), where additional time is required for the large air mass in the valley to be heated to reverse the nighttime along-valley pressure gradient. We can hypothesize that the Weber Canyon situation is caused by the nighttime buildup and persistence into the late morning of a cold-air pool in the Morgan Basin on the east end of the Weber Canyon, and the accompanying along-valley pressure gradient from the east to the west end of the canyon. This hypothesis can be tested in the future with temperature soundings and/or pressure measurements at the canyon’s east end. Second, if the temperature contrast between the persistent elevated warm air in the Salt Lake Basin and the falling temperatures inside the canyon increases during the night, why do the canyon winds and exit winds approach a steady state? This is surely the result of the approach to a steady-state balance of forces in the along-valley momentum equation as the drag of the canyon sidewalls increases.
With respect to the wind power implications of the Weber Canyon exit jet, the wind resource can be characterized as quite strong during the katabatic phase of the wind system but weak during the anabatic phase. Further, the down-valley-phase winds are from a narrow range of directions, which is advantageous for some types of wind turbines. The height of the wind maximum above ground also is within the range of normal turbine hub heights. The repeatability of the exit-jet winds from day to day, even when synoptic-scale weather systems and pressure gradients were absent and most other locations would not have usable winds, is another positive feature of this wind resource. On the other hand, this wind resource is not available during the afternoon and early evening. Energy storage techniques might be used to distribute the wind resource over all hours of the day. For example, the down-valley-phase winds could be used to generate electricity to pump water to an elevated reservoir, and the water could be released in the afternoon to generate needed power through hydroturbines. The valley exit-jet wind resource has possible drawbacks, as turbines might be unacceptable at heavily populated canyon mouths and transmission lines might not be available at canyon mouths or might be unacceptable to local populations. Finally, the wind resource is in the form of a wind jet, with significant wind shear both above and below the jet maximum. Turbine blades and gearboxes would need to be engineered to withstand the flexing that would derive from the shear profile.
6. Conclusions and future work
A valley-exit jet at Weber Canyon, Utah, has been investigated observationally using both long- and short-term meteorological data. The jet is a local, diurnal phenomenon occurring at the canyon exit where the down-valley flows inside the canyon are modified as relatively cold air flowing down valley from the basin east of the canyon exits the canyon and comes into the warmer Great Salt Lake basin. Upon exiting the canyon, the wind descends, thins, and accelerates to produce a shallow high-speed jetlike flow at (and beyond) the canyon exit. The jetlike shape of the wind speed profile is a result of the flow compression and friction that forces the wind speed to vanish at the surface. As a diurnal thermally driven flow, it forms most readily and is best developed under fair weather conditions when synoptic-scale forcing is weak. The initiation, maintenance, and dissipation of the jet, and thus its duration, are closely tied to the times of sunrise and sunset, varying with day of year. The jet is typically initiated within 1–3 h following sunset. The strength increases rapidly after initiation, and tends to approach a steady state during the night with jet maximum wind speeds of 15–20 m s−1 at levels of about 80–120 m above the ground. The maximum development occurs near sunrise, and the speeds decrease slowly from then until about 5–6 h later, when the wind direction reverses and daytime up-valley flows form in the canyon. The strongest, deepest, and most consistent down-canyon and exit-jet flows occur during summer and early fall nights. Highly consistent valley-exit jets were observed in Weber Canyon during the summer and early fall of 2010 with ~83% of the nights from July through September exhibiting exit jets. The katabatic flow was observed to rapidly accelerate from inside the canyon to the exit region where winds were typically 2.5 times as strong as those inside the canyon. Strong downward vertical wind speeds were observed at the canyon exit, supporting the theory that a deep but weak down-valley flow inside the canyon thins, descends, and accelerates at the exit of the canyon, converting potential energy into kinetic energy. Through this descent mechanism, potentially warmer air from aloft within the canyon is brought down closer to the surface at the exit. The exit jet speeds at Weber Canyon are greater than for other previously reported valley-exit jets, such as that at the exit of Austria’s Inn Valley. The strongest winds were contained within a strong surface-based temperature inversion both inside and at the exit of the canyon. There is little interannual variation of wind speeds in Weber Canyon, as determined from a 10-yr period of record.
Weber Canyon exit-jet characteristics can be significantly modified by passing synoptic-scale weather systems. An IOP was run on 18–19 September 2010 in which a moderate synoptic-scale pressure gradient existed over the region. During this IOP, the jet began normally but the nighttime jet was weaker (12–15 m s−1) and the jet dissipated by midmorning.
This research project had a limited scope and budget, leaving much room for further research. A further expansion of the research into other drainage basins using modeling and observations could produce a more general understanding of the valley-exit jet phenomenon and help answer fundamental questions about the characteristics of canyon flows and their evolution. Additional research could help determine if parameters such as drainage basin size, cold-air production potential, and canyon exit geometry play important roles in valley-exit jet dynamics. With respect to Weber Canyon, we do not yet know how far the exit jet extends into the Great Salt Lake basin and whether the Morgan Basin cold-air pool on the east side of the canyon plays a major role in extending the down-valley flow and jet into late morning.
Acknowledgments
This material is based upon work supported by the National Science Foundation under Grant ATM-0938397. The authors thank the reviewers for their comments and suggestions that helped to improve the manuscript. The first author thanks Sebastian W. Hoch, Joseph S. Young, and Christopher Ander for their assistance during the measurement campaign. We also thank the State of Utah anemometer loan program for providing data from the meteorological tower and to NRG Systems, Inc., for loaning the WINDCUBE lidar for use in this research. Thanks are given to Gregory S. Poulos and Julie K. Lundquist for their assistance in loaning the lidar and to Alan Marchant and Tom Apedaile from Utah State University for operating and providing data from the ValidWind system during the measurement campaign. Several organizations provided land on which measurement equipment was temporarily installed, and we are grateful for their cooperation.
REFERENCES
Banta, R. M., L. D. Olivier, W. D. Neff, D. H. Levinson, and D. Ruffieux, 1995: Influence of canyon-induced flows on flow and dispersion over adjacent plains. Theor. Appl. Climatol., 52, 27–41.
Cramer, P., 1972: Potential temperature analysis for mountainous terrain. J. Appl. Meteor., 11, 44–50.
Darby, L. S., and R. M. Banta, 2006: The modulation of canyon flows by larger-scale influences. Preprints, 12th Conf Mountain Meteorology, Santa Fe, NM. Amer. Meteor. Soc., 14.4. [Available online at https://ams.confex.com/ams/pdfpapers/114383.pdf.]
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.
Mass, C. F., S. Businger, M. D. Albright, and Z. A. Tucker, 1995: A windstorm in the lee of a gap in a coastal mountain barrier. Mon. Wea. Rev., 123, 315–331.
Mayr, G. J., and Coauthors, 2007: Gap flows: Results from the Mesoscale Alpine Programme. Quart. J. Roy. Meteor. Soc., 133, 881–896.
Overland, J. E., and B. A. Walter Jr., 1981: Gap winds in the Strait of Juan de Fuca. Mon. Wea. Rev., 109, 2221–2227.
Pamperin, H., and G. Stilke, 1985: Nächtliche Grenzschicht und LLJ im Alpenvorland nahe dem Inntalausgang (Nocturnal boundary layer and low-level jet near the Inn Valley exit). Meteor. Rundsch., 38, 145–156.
Poulos, G. S., and S. Zhong, 2008: The observational history of small-scale katabatic winds in mid-latitudes. Geogr. Compass, 2, 1798–1821.
Sharp, J., and C. F. Mass, 2002: Columbia Gorge gap flow: Insights from observational analysis and ultra-high-resolution simulation. Bull. Amer. Meteor. Soc., 83, 1757–1762.
Sharp, J., and C. F. Mass, 2004: Columbia Gorge gap winds: Their climatological influence and synoptic evolution. Wea. Forecasting, 19, 970–992.
Stilke, G., 1984: Nocturnal boundary layer and low-level jet in the pre-Alpine region near the outlet of the Inn Valley. Proc. 18th Int. Conf. on Alpine Meteorology, Opatija, Germany, ICAM, 68–71.
Whiteman, C. D., 1990: Observations of thermally developed wind systems in mountainous terrain. Atmospheric Processes over Complex Terrain, Meteor. Monogr., No. 45. Amer. Meteor. Soc., 5–42.
Whiteman, C. D., 2000: Mountain Meteorology: Fundamentals and Applications. Oxford University Press, 355 pp.
Wilkerson, T. D., A. B. Marchant, and T. J. Apedaile, 2012: Wind field characterization from the trajectories of small balloons. J. Atmos. Oceanic Technol., 29, 1236–1249.
Zängl, G., 2004: A reexamination of the valley wind system in the Alpine Inn Valley with numerical simulations. Meteor. Atmos. Phys., 87, 241–256.
Zardi, D., and C. D. Whiteman, 2012: Diurnal mountain wind systems. Mountain Weather Research and Forecasting, F. K. Chow, S. F. J. DeWekker, and B. Snyder Eds., Springer, 35–119.
