1. Introduction
Mountainous terrain is often characterized by large spatial variations in orography and surface cover. The various terrain features give rise to thermally driven wind circulations of different spatial and temporal scales, from large mountain–plain circulations (e.g., Lugauer and Winkler 2005; Bica et al. 2007) to shallow slope flows or drainage flows in small gullies (e.g., Mahrt et al. 2001, 2014). The thermally driven circulation systems can interact with one another (e.g., Doran et al. 1990; Whiteman and Zhong 2008), or with other flows such as synoptic winds (e.g., Barr and Orgill 1989) and other complex-terrain phenomena such as valley cold pools (e.g., Catalano and Cenedese 2010; Mahrt et al. 2010).
In this paper, we describe the nocturnal boundary layer development during a quiescent spring night on a low-angle slope at the foot of a steep desert mountain that borders a large valley (Fig. 1a). The data were taken on the lower east sidewall of Granite Mountain at the U.S. Army’s Dugway Proving Ground in Utah during the Mountain Terrain Atmospheric Modeling and Observations (MATERHORN) field campaign. The field campaign consisted of two 1-month field experiments, which were conducted in October of 2012 and May of 2013, with the overarching objectives of studying interactions of motions of different scales in complex terrain and of improving the predictability of weather in complex terrain (Fernando et al. 2015). In this paper, we focus exclusively on the night of spring intensive observational period (IOP) 4, which ran for 24 h starting at 1300 mountain standard time (MST) 11 May 2013. The observations from IOP4 highlight the complexity of the nocturnal boundary layer development over the slope and some of the interactions that are the focus of the MATERHORN project. During this IOP, tethersonde ascents on the slope supplemented other MATERHORN field measurements, which thus provided additional information on the vertical structure of the atmosphere above the slope that is not available for other IOP nights. The nocturnal boundary layer development started with the evening-transition period, which is characterized by strong surface cooling and the development of a surface-based inversion, as well as by the transition from daytime upslope/up-valley winds to nighttime downslope/down-valley winds. The evening transition was followed by a period of almost undisturbed boundary layer conditions on the sidewall until approximately 2230 MST. After 2300 MST, the inversion that had formed over the adjacent valley sloshed up the sidewall repeatedly, engulfing the lower part of the mountain slope in colder air and disturbing the local downslope flows.
(a) Map of Granite Mountain and Dugway Valley. The rectangular outline over Granite Mountain shows the domain of (b). (b) Detailed map of the instrumented east sidewall of Granite Mountain and the locations of measurement sites. The symbols in (a) and (b) labeled Sagebrush and TS mark the locations of tethersonde measurements, and the gray dots in (a) show the network of mini-SAMS sites. In (b) the filled squares labeled ES show the locations of the five towers, and the open circles labeled PW show the locations of PWIDS surface stations. The gray symbol labeled LID and line indicate the location of the lidar and the direction of the lidar scans shown in Fig. 14. Contour intervals in (a) and (b) are 100 and 10 m, respectively. The floor of the Dugway Valley slopes slightly down to the northwest but is almost flat.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
(a) Map of Granite Mountain and Dugway Valley. The rectangular outline over Granite Mountain shows the domain of (b). (b) Detailed map of the instrumented east sidewall of Granite Mountain and the locations of measurement sites. The symbols in (a) and (b) labeled Sagebrush and TS mark the locations of tethersonde measurements, and the gray dots in (a) show the network of mini-SAMS sites. In (b) the filled squares labeled ES show the locations of the five towers, and the open circles labeled PW show the locations of PWIDS surface stations. The gray symbol labeled LID and line indicate the location of the lidar and the direction of the lidar scans shown in Fig. 14. Contour intervals in (a) and (b) are 100 and 10 m, respectively. The floor of the Dugway Valley slopes slightly down to the northwest but is almost flat.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
(a) Map of Granite Mountain and Dugway Valley. The rectangular outline over Granite Mountain shows the domain of (b). (b) Detailed map of the instrumented east sidewall of Granite Mountain and the locations of measurement sites. The symbols in (a) and (b) labeled Sagebrush and TS mark the locations of tethersonde measurements, and the gray dots in (a) show the network of mini-SAMS sites. In (b) the filled squares labeled ES show the locations of the five towers, and the open circles labeled PW show the locations of PWIDS surface stations. The gray symbol labeled LID and line indicate the location of the lidar and the direction of the lidar scans shown in Fig. 14. Contour intervals in (a) and (b) are 100 and 10 m, respectively. The floor of the Dugway Valley slopes slightly down to the northwest but is almost flat.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
The current state of knowledge on downslope winds is summarized in Zardi and Whiteman (2012). Several studies have documented the thermal and dynamic characteristics of steady-state downslope winds and the dominant forces on the basis of both observations (e.g., Doran et al. 1990; Haiden and Whiteman 2005) and numerical simulations (e.g., Skyllingstad 2003). Downslope winds are also sensitive to disturbances that are due to other circulations, however. Zardi and Whiteman (2012) list, among others, valley winds, synoptic-scale flows, sea breezes, seiches, and gravity waves. Downslope winds on mountain sidewalls bordering a valley are oftentimes influenced by strengthening down-valley winds (e.g., Whiteman and Zhong 2008), which can produce large directional wind shear with height and can eventually overpower the downslope winds as the night progresses. The nocturnal boundary layer over the east sidewall of Granite Mountain is frequently impacted by a sloshing of the valley inversion (Fernando et al. 2015), resulting in large temperature oscillations and intermittent disturbances to the slope winds. Seiches in basin cold pools have been observed, for example, in model simulations for Arizona’s Meteor Crater by Fritts et al. (2010). Rapid temperature changes caused by the displacement of a cold-air pool were observed by Lareau et al. (2013) in the Salt Lake basin of Utah, and Whiteman et al. (2001) described strong temperature changes along the slope of Rattlesnake Mountain in Washington, which were produced by seiches or gravity waves on top of a strong inversion layer.
Different mechanisms leading to the evening transition from upslope to downslope flows have recently been reported in the literature. Fernando et al. (2013) described two different evening-transition mechanisms for sloping terrain: 1) the “sliding slab,” that is, a simultaneous stagnation and onset of the downslope flow along the entire slope, and 2) a “transition front,” that is, the formation of a stagnation front that travels down the slope, lifting the weakening upslope-flow layer from the slope surface. The latter mechanism of a transition front is based on a slope-flow evening-transition model developed by Hunt et al. (2003). Downward-traveling transition fronts were observed in the Phoenix basin in Arizona by Brazel et al. (2005) and Pardyjak et al. (2009), on a gentle northwest-facing slope (Mahrt et al. 2010), and on Mt. Hymettos in Greece, where the shadow propagated down the slope (Papadopoulos and Helmis 1999). In contrast, Nadeau et al. (2013) observed an upward-propagating flow transition on a west-facing slope in an Alpine valley, where the shadow moved up the slope. Martínez Villagrasa et al. (2013) found an almost simultaneous transition at two sites on the short slope of Meteor Crater on several days and a relation of the transition time to the decay of daytime convection and turbulent kinetic energy in the crater. In this study, we compare the evening flow transition on the east slope of Granite Mountain with these previously described mechanisms.
The tethersonde and other measurements used in this study are described in section 2, followed by a brief overview of the synoptic situation during IOP4 in section 3. Section 4 presents the results, and a summary and conclusions are provided in section 5.
2. Data
Measurements were made on the lower east-facing sidewall of Granite Mountain [2159 m above mean sea level (MSL)]. Granite Mountain is located to the west of Dugway Valley, which is a basinlike valley that is enclosed by higher terrain to the east, south, and southwest (Fig. 1a). The instrumented slope at the foot of Granite Mountain is shown in more detail in Figs. 1b and 2. The upper part of the instrumented slope is bounded by higher terrain on both the north and south sides, similar to a small canyon. The canyonlike part is approximately 1.2 km long and 200 m deep, with an approximately 300-m-wide valley floor in the upper part, which widens to approximately 700 m in the lower part. Below the canyon, the terrain opens onto an alluvial fan and is completely open to the north, south, and east. The slope angle decreases from ~3°–5° in the canyonlike upper part to ~1.5°–3° on the alluvial fan. Because of the east-facing aspect of the terrain, the shadow propagates down the slope in the evening. The local topography at the measurement site can lead to complex interactions between slope winds in the west–east direction (west is downslope, and east is upslope) and along-valley winds in the north–south direction (north is up-valley, and south is down-valley). The data used in this study come from two tethered-balloon sounding systems, five meteorological towers, 12 Dugway Proving Ground Portable Weather Information Display Systems (PWIDS), a network of the miniature configuration of the Dugway Proving Ground Surface Atmospheric Measurement Systems (“mini-SAMS”), and a scanning Doppler lidar. The instrumentation is briefly described below. A more detailed description of all of the MATERHORN instrumentation and the accuracy of the instruments can be found in Fernando et al. (2015).
The instrumented east slope of Granite Mountain, looking to the west-northwest. The locations of towers ES2–ES5 are marked for guidance. (Copyright Google.)
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
The instrumented east slope of Granite Mountain, looking to the west-northwest. The locations of towers ES2–ES5 are marked for guidance. (Copyright Google.)
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
The instrumented east slope of Granite Mountain, looking to the west-northwest. The locations of towers ES2–ES5 are marked for guidance. (Copyright Google.)
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
During the night of IOP4, a Vaisala, Inc., “DigiCORA” tethersonde system was operated on the lower part of the slope at an elevation of 1343 m MSL (Fig. 1b). The sonde, which measured pressure, temperature, relative humidity, and wind speed and direction, was attached to a line below a balloon that was tethered to an electric winch at the ground. Regular ascents were made to a height of approximately 200 m above ground level (AGL). After midnight, some soundings were extended to a height of 400 m AGL to probe the atmosphere farther away from the surface. Soundings were made between 1730 and 0830 MST, with sounding intervals of approximately 15 min during the evening (1730–2200 MST) and morning (0530–0830 MST) periods and approximately 30 min during the rest of the night. Only data from the ascents were used for the analysis; data from the rapid descents were discarded. An identical tethersonde system was operated in the center of the valley at the Sagebrush site (Fig. 1a) at an elevation of 1316 m MSL, that is, approximately 30 m lower than the tethersonde site on the east slope of Granite Mountain. Soundings at Sagebrush were made between 1715 and 0800 MST, with soundings extending to a height of 400 m AGL, except for the evening (1830–2010 MST) and morning (0550–0800 MST) periods, when ascents were made more frequently at the expense of height, with soundings extending up to 200 m AGL.
Tethersonde data were supplemented with data from a network of 12 PWIDS surface weather stations (Fig. 1b). Local topographical characteristics of the sites are listed in Table 1. PWIDS stations measured wind, temperature, relative humidity, and pressure at 2 m AGL. Five-minute-averaged temperatures that deviated by more than 15°C from the mean over all other PWIDS temperatures at a given time or that were outside 3 standard deviations were excluded from the analysis. Further data came from a network of 51 mini-SAMS that was deployed in the Dugway Valley east of Granite Mountain (Fig. 1a). The mini-SAMS consisted of 10-m towers, which were instrumented with wind sensors at 2 and 10 m, as well as with temperature, relative humidity, and pressure sensors at 2 m.
Elevation (m MSL), slope azimuth angle (°), and slope angle (°) at the PWIDS sites. The elevation was determined from the 1/3-arc-s National Elevation Dataset (NED, provided through the courtesy of the U.S. Geological Survey; Gesch et al. 2002). Local azimuth and slope angle were determined using a geographic information system. Since the slope angle at PW80 is 0.0, no azimuth angle could be determined, and the value of the upstream site PW89 was used.
Five 20–32-m-high meteorological towers (ES1–ES5) were deployed along the sidewall (Fig. 1b). Local topographical characteristics at the tower sites are listed in Table 2 together with information on the instrumentation levels. Sonic anemometers were installed at different levels between approximately 0.5 m AGL and the tops of the towers. Thermocouples were installed at ES2, ES3, and ES5 to measure temperature profiles. The thermocouples were collocated with the sonic anemometers at ES5, but their heights differed from sonic-anemometer levels at ES2 and ES3.
Elevation (m MSL), slope azimuth angle (°), slope angle (°), and instrumentation heights (m AGL) at the five instrumented towers ES1–ES5. Elevations were determined from the 1/3-arc-s NED. Local azimuth and slope angle were determined using a geographic information system. Since the slope angle at ES1 is 0.0, the azimuth angle at the upstream site PW89 was used. The first row of instrumentation levels for all sites is for sonic anemometers, and the second row for ES2, ES3, and ES5 is for thermocouple heights.
In the following discussion, u and υ refer to the west–east and south–north wind components, respectively. While the general slope is oriented almost perfectly in the west–east direction, horizontal wind components in the local downslope direction us were calculated using the local slope azimuth angles listed in Tables 1 and 2 and were used for part of the analysis. All of the data from the PWIDS, the five meteorological towers, and the mini-SAMS reported below are 5-min averages unless specified otherwise, with the indicated times referring to the end of the averaging period.
3. Synoptic conditions
During the night of 11–12 May, a ridge of high pressure was located over the western part of the United States (Fig. 3). Northern Utah was under weak geopotential-height gradients with predominantly southeasterly to southwesterly flow at 850 hPa. Radiosoundings at Salt Lake City International Airport showed a weak southeasterly flow below 850 hPa that was topped by northwesterly flow below 500 hPa (Fig. 4). Temperature profiles showed a deep close-to-adiabatic layer above the nocturnal surface-based inversion, with a shallow stable layer near 500 hPa. High cirrostratus clouds were present in the late afternoon, but the clouds cleared in the early evening, and clear skies prevailed during the night. A decrease in humidity above the 650-hPa pressure level occurred between the 0500 and 1700 MST 11 May soundings. As the jet region north of Utah moved farther to the south during the second half of the night, northern Utah came under the influence of a southwesterly flow (Fig. 3).
ERA-Interim reanalysis (Dee et al. 2011) of 850-hPa geopotential height (black solid lines; 50-m contour interval), temperature (white dashed lines; 2-K contour interval), wind (arrows), and relative humidity (color contours). The red symbol indicates the location of Granite Mountain, and the blue L and H label the locations of surface pressure lows and highs, respectively.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
ERA-Interim reanalysis (Dee et al. 2011) of 850-hPa geopotential height (black solid lines; 50-m contour interval), temperature (white dashed lines; 2-K contour interval), wind (arrows), and relative humidity (color contours). The red symbol indicates the location of Granite Mountain, and the blue L and H label the locations of surface pressure lows and highs, respectively.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
ERA-Interim reanalysis (Dee et al. 2011) of 850-hPa geopotential height (black solid lines; 50-m contour interval), temperature (white dashed lines; 2-K contour interval), wind (arrows), and relative humidity (color contours). The red symbol indicates the location of Granite Mountain, and the blue L and H label the locations of surface pressure lows and highs, respectively.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
Radiosoundings from the Salt Lake City International Airport (40.77°N, 111.95°W; ~130 km northeast of Granite Mountain) at 0500 MST 11 May (black), 1700 MST 11 May (red), and 0500 MST 12 May (blue). Profiles of temperature (rightmost curves), dewpoint temperature (leftmost curves), and wind (wind barbs) are shown. Feathers on the wind barbs are for 25 (filled), 5 (long), and 2.5 (short) m s−1.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
Radiosoundings from the Salt Lake City International Airport (40.77°N, 111.95°W; ~130 km northeast of Granite Mountain) at 0500 MST 11 May (black), 1700 MST 11 May (red), and 0500 MST 12 May (blue). Profiles of temperature (rightmost curves), dewpoint temperature (leftmost curves), and wind (wind barbs) are shown. Feathers on the wind barbs are for 25 (filled), 5 (long), and 2.5 (short) m s−1.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
Radiosoundings from the Salt Lake City International Airport (40.77°N, 111.95°W; ~130 km northeast of Granite Mountain) at 0500 MST 11 May (black), 1700 MST 11 May (red), and 0500 MST 12 May (blue). Profiles of temperature (rightmost curves), dewpoint temperature (leftmost curves), and wind (wind barbs) are shown. Feathers on the wind barbs are for 25 (filled), 5 (long), and 2.5 (short) m s−1.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
4. Results
Data from the tethered-balloon soundings (Fig. 5) and the five meteorological towers (Fig. 6) on the slope document the nocturnal boundary layer development at and above the surface, starting with the evening flow transition. For this study, we define the period of the evening flow transition as the period between the passage of the shadow at the first site in the upper part of the slope and the final onset of downslope flows at the last site in the lower part of the slope. The evening-flow-transition period thus defined started around 1800 MST and lasted for about 1 h. Following the short transition period, a second and somewhat longer period of boundary layer development occurred between approximately 1900 and 2230 MST. This period was characterized by an almost undisturbed surface inversion, a katabatic flow that formed locally over the slope, and near-surface temperatures that stayed almost constant (Fig. 6). Small disturbances initiated by an increase in wind speed above the shallow surface boundary layer affected the slope temperature and wind fields intermittently. After approximately 2300 MST, the atmosphere over the east slope was strongly affected by the inversion that had formed over the Dugway Valley. The valley inversion pushed up the slope repeatedly and retreated again, resulting in an oscillatory behavior of the temperature and wind direction at ES1–ES5 (Fig. 6). These three different phases or their close variants were also observed during other IOPs (Hocut et al. 2014; Leo et al. 2014).
Time–height cross sections of (a) temperature and (b) wind speed (colors) and horizontal wind arrows from tethered-balloon soundings on the east slope between 1700 and 0800 MST. Contour lines in (a) are drawn every 2°C. Wind profiles were smoothed using a nine-point moving-average filter. Vertical black lines indicate the boundaries between the three different phases.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
Time–height cross sections of (a) temperature and (b) wind speed (colors) and horizontal wind arrows from tethered-balloon soundings on the east slope between 1700 and 0800 MST. Contour lines in (a) are drawn every 2°C. Wind profiles were smoothed using a nine-point moving-average filter. Vertical black lines indicate the boundaries between the three different phases.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
Time–height cross sections of (a) temperature and (b) wind speed (colors) and horizontal wind arrows from tethered-balloon soundings on the east slope between 1700 and 0800 MST. Contour lines in (a) are drawn every 2°C. Wind profiles were smoothed using a nine-point moving-average filter. Vertical black lines indicate the boundaries between the three different phases.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
Time series of temperature, wind direction, and wind speed at ES1–ES5 (ES1 and ES4 are wind only) between 1700 and 0800 MST. For ES2 and ES3, the leftmost of the legends on the right side of the panel lists temperature levels and the rightmost lists wind levels.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
Time series of temperature, wind direction, and wind speed at ES1–ES5 (ES1 and ES4 are wind only) between 1700 and 0800 MST. For ES2 and ES3, the leftmost of the legends on the right side of the panel lists temperature levels and the rightmost lists wind levels.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
Time series of temperature, wind direction, and wind speed at ES1–ES5 (ES1 and ES4 are wind only) between 1700 and 0800 MST. For ES2 and ES3, the leftmost of the legends on the right side of the panel lists temperature levels and the rightmost lists wind levels.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
Tethersonde operators visually observed the arrival of the shadow at the site at about 1835 MST; astronomical sunset was at 1932 MST. Surface cooling at the tethersonde site started to produce an inversion before local sunset, between 1812 and 1830 MST. Whereas the 1812 MST sounding showed a neutral near-surface layer, the subsequent 1830 MST sounding already had a weak and shallow surface-based inversion. The surface inversion maintained a depth of 20–30 m until approximately 2230 MST. The strongest temperature gradient occurred within the lowest 10 m AGL, with temperature increases of 6°C or more in some of the soundings. Above the inversion layer, the atmosphere remained close to neutral while cooling continuously.
The wind, which was from a northerly up-valley direction in the late afternoon, changed to a westerly downslope direction in the lowest 30 m AGL around local sunset (Fig. 5b). Above the downslope-flow layer, the wind changed initially from northerly to easterly and eventually to a southerly down-valley wind. After 2230 MST, a relatively strong southwesterly wind was present throughout the entire depth of the approximately 200-m-deep sounding.
Tethered-balloon soundings at Sagebrush (Fig. 7) in the center of the valley (see Fig. 1 for location) show a different evolution of the vertical temperature structure than do observations on the slope. Surface cooling at Sagebrush started around 1930 MST, that is, somewhat later than on the sidewall, in agreement with later shading at the relatively flat and open valley site. In contrast to the slope sites, the surface inversion at Sagebrush continued to grow in depth after the initiation of surface cooling until the morning. The continuous growth was only interrupted by short, intermediate mixing events at the top of the inversion, for example, around 0500 MST. A direct comparison of the soundings at Sagebrush and the slope shows that temperatures in the lowest 50 m AGL were generally colder at Sagebrush, an indication of cold-air pooling on the valley floor.
(a),(c) Vertical profiles and (b),(d) time–height cross sections of (top) temperature and (bottom) horizontal wind speed from tethered-balloon soundings at Sagebrush. Profiles from the east-slope tethersonde are also included in (a) and (c) for comparison (gray lines). Horizontal wind arrows are included in (d). Contour lines in (b) are drawn every 2 K. Wind profiles were smoothed using a nine-point moving-average filter.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
(a),(c) Vertical profiles and (b),(d) time–height cross sections of (top) temperature and (bottom) horizontal wind speed from tethered-balloon soundings at Sagebrush. Profiles from the east-slope tethersonde are also included in (a) and (c) for comparison (gray lines). Horizontal wind arrows are included in (d). Contour lines in (b) are drawn every 2 K. Wind profiles were smoothed using a nine-point moving-average filter.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
(a),(c) Vertical profiles and (b),(d) time–height cross sections of (top) temperature and (bottom) horizontal wind speed from tethered-balloon soundings at Sagebrush. Profiles from the east-slope tethersonde are also included in (a) and (c) for comparison (gray lines). Horizontal wind arrows are included in (d). Contour lines in (b) are drawn every 2 K. Wind profiles were smoothed using a nine-point moving-average filter.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
a. Phase 1—Evening flow transition
Time series of temperature and wind direction at ES1–ES5 are shown in Fig. 6. Temperature dropped first at ES5 shortly after 1800 MST. Farther down the slope at ES3 and ES2 the temperature drop occurred slightly later at approximately 1840 and 1835 MST, respectively. The respective times of cooling onset agree well with the time of local sunset, that is, the time of shadow passage (Fig. 8a) and the time of heat-flux reversal at the lowest tower levels, from an upward heat flux to a downward heat flux at 1810 MST at ES5 and at 1835 MST at ES2 and ES3 (not shown). Here, cooling onset refers to the beginning of the rapid near-surface temperature decrease after local sunset. Weaker cooling started already at an earlier time. The spatial distribution of cooling-onset times at the PWIDS stations and at the towers is shown in Fig. 8b. It shows the gradual propagation of surface cooling down the slope from northwest to southeast, in agreement with the shadow propagation on the slope (Fig. 8a).
Map of time (MST) of (a) local sunset, (b) cooling onset, and (c) downslope-flow onset. Local sunset in (a) is determined from a shadow algorithm on a 10-m-resolution digital elevation model with a temporal resolution of 5 min, the cooling onset in (b) is defined as the time for which the 2-m temperature (1 m for ES2 and ES3) dropped by at least 1.1°C over the preceding 10 min, and the downslope-flow onset time in (c) is defined as the time when the 2-m wind direction changed to 270° ± 45°.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
Map of time (MST) of (a) local sunset, (b) cooling onset, and (c) downslope-flow onset. Local sunset in (a) is determined from a shadow algorithm on a 10-m-resolution digital elevation model with a temporal resolution of 5 min, the cooling onset in (b) is defined as the time for which the 2-m temperature (1 m for ES2 and ES3) dropped by at least 1.1°C over the preceding 10 min, and the downslope-flow onset time in (c) is defined as the time when the 2-m wind direction changed to 270° ± 45°.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
Map of time (MST) of (a) local sunset, (b) cooling onset, and (c) downslope-flow onset. Local sunset in (a) is determined from a shadow algorithm on a 10-m-resolution digital elevation model with a temporal resolution of 5 min, the cooling onset in (b) is defined as the time for which the 2-m temperature (1 m for ES2 and ES3) dropped by at least 1.1°C over the preceding 10 min, and the downslope-flow onset time in (c) is defined as the time when the 2-m wind direction changed to 270° ± 45°.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
Wind profiles from the tethered-balloon soundings showed a northerly up-valley flow in the late afternoon throughout most of the sounding depth (Fig. 9). The u component of the wind was relatively weak at this time but was still directed up the slope during the first sounding (Fig. 9c). The slope azimuth angle at the tethersonde site is 90°, that is, the west–east u component of the wind is identical to the local downslope-wind component. In the second sounding, u had already turned to a downslope direction in the lowest 100 m. The westerly downslope component increased during the following hour, and a 1–2 m s−1 downslope flow became clearly distinguishable in the lowest 30 m after 1830 MST (Fig. 9b). The northerly up-valley component weakened at the same time and turned eventually to a southerly down-valley component, starting near the surface at 1900 MST (Fig. 9d). The flow near the surface, however, remained dominated by the westerly downslope flow. The transition from an up-valley flow to a down-valley flow therefore started above 100 m, where the northerly up-valley flow was replaced by easterly winds at 1900 MST while the up-valley flow continued between the top of the approximately 30-m-deep downslope flow and 100 m (Fig. 9b). The up-valley flow was subsequently completely removed, and easterly winds prevailed above the downslope-flow layer during the entire transition period between up-valley flow and down-valley flow. The easterly flow was eventually replaced by a southerly down-valley flow above the downslope-flow layer, starting after 2000 MST.
Time–height cross sections of (a) temperature, (b) horizontal wind, (c) u component of the wind, and (d) υ component of the wind from tethered-balloon soundings. Note that the height range in (a) differs from the height range in (b)–(d). Contour lines are drawn every 2°C in (a) and every 0.5 m s−1 in (c) and (d). Positive velocities in (c) and (d) are indicated by solid lines, and negative velocities are shown by dotted lines. The thick contour line marks 0 m s−1. Annotations in (b) indicate up-valley flow (UV), downslope flow (DS), a transition from up-valley flow to down-valley flow (UV→DV), and down-valley flow (DV). Wind profiles were smoothed using a nine-point moving-average filter.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
Time–height cross sections of (a) temperature, (b) horizontal wind, (c) u component of the wind, and (d) υ component of the wind from tethered-balloon soundings. Note that the height range in (a) differs from the height range in (b)–(d). Contour lines are drawn every 2°C in (a) and every 0.5 m s−1 in (c) and (d). Positive velocities in (c) and (d) are indicated by solid lines, and negative velocities are shown by dotted lines. The thick contour line marks 0 m s−1. Annotations in (b) indicate up-valley flow (UV), downslope flow (DS), a transition from up-valley flow to down-valley flow (UV→DV), and down-valley flow (DV). Wind profiles were smoothed using a nine-point moving-average filter.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
Time–height cross sections of (a) temperature, (b) horizontal wind, (c) u component of the wind, and (d) υ component of the wind from tethered-balloon soundings. Note that the height range in (a) differs from the height range in (b)–(d). Contour lines are drawn every 2°C in (a) and every 0.5 m s−1 in (c) and (d). Positive velocities in (c) and (d) are indicated by solid lines, and negative velocities are shown by dotted lines. The thick contour line marks 0 m s−1. Annotations in (b) indicate up-valley flow (UV), downslope flow (DS), a transition from up-valley flow to down-valley flow (UV→DV), and down-valley flow (DV). Wind profiles were smoothed using a nine-point moving-average filter.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
The depth of the westerly downslope-flow layer near the surface varied between approximately 20 and 50 m during the transition from up-valley winds to down-valley winds above. The downslope-flow layer rapidly increased in depth immediately after the decay of the up-valley flow. As the easterly winds above the downslope-flow layer increased in speed, they reached close to the surface, and the downslope-flow layer depth was reduced to about 15 m at approximately 2000 MST. Afterward, as the easterly flow was gradually replaced by initially weak down-valley winds, the downslope flow dominated again over an approximately 60-m deep layer. Once the southerly down-valley flow had increased in strength, the downslope-flow layer depth again decreased. Then, as the down-valley winds increased further, the exact transition between downslope and down-valley flow became difficult to distinguish (Fig. 9b).
The tethered-balloon soundings captured the complete transition from up-valley winds to down-valley winds above the surface layer. We will now focus on the near-surface layer and the transition to downslope flows. In contrast to the lower part of the slope, where a northerly to northeasterly up-valley wind prevailed in the late afternoon, the wind was generally from an easterly upslope direction in the upper part of the slope, including ES4 and ES5 (Figs. 6, 10). The time of transition to a westerly downslope flow agreed with the time of local sunset and the onset of surface cooling at most sites along the slope (Fig. 8). The flow transition thus propagated down the slope from the northwest to the southeast together with the shadow front. Exceptions were PW79 at the top of the slope and ES3 and ES4. At PW79, the upslope–downslope flow transition and the onset of surface cooling occurred 35 min after local sunset, close to the transition in the lower part of the slope. At ES3 and ES4, the flow transition at 2 m AGL also occurred approximately 0.5 h after local sunset. The transition at the lower 0.5-m levels, however, had already occurred at 1840 and 1834 MST, respectively. Papadopoulos and Helmis (1999) used more-complex criteria for defining the onset time of katabatic flows. Their criteria required a positive temperature gradient at the surface, a near-surface wind speed maximum, and a flow lasting for at least 1 h. Since no information on the vertical profiles was available for the PWIDS stations, we used a much simpler definition that is only based on temporal evolution.
Map of the horizontal wind components at 2 m AGL at the PWIDS and tower sites at six different times in the evening.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
Map of the horizontal wind components at 2 m AGL at the PWIDS and tower sites at six different times in the evening.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
Map of the horizontal wind components at 2 m AGL at the PWIDS and tower sites at six different times in the evening.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
The transition on the upper slope above ES4, which is sheltered by surrounding topography, differed from the transition on the lower open sidewall. In the upper part, the afternoon upslope flow weakened until the surface winds became almost stagnant and then a downslope flow developed and increased in strength. In the lower part, wind speeds also decreased during the transition but did not reach a near-zero value; rather, the wind gradually turned counterclockwise from a northerly up-valley direction to a westerly downslope direction (Fig. 10). A counterclockwise turning of diurnal winds is typical for sidewalls on the left side of a valley, looking down-valley (Whiteman 1990).
The different flow-transition behaviors that have been found in different locations (e.g., Fernando et al. 2013; Nadeau et al. 2013; Martínez Villagrasa et al. 2013) may indicate that the local terrain plays a large role in the transition process. The evening flow transition on the east sidewall of Granite Mountain during IOP4 was governed by local surface cooling at the individual sites. The flow transition propagated down the slope, similar to the “transition front” described by Fernando et al. (2013), but following the movement of the shadow down the sidewall from northwest to southeast. This is in agreement with the findings by Nadeau et al. (2013), who observed an upslope propagation of the shadow and the flow transition on a west-facing sidewall in a Swiss Alpine valley. Analysis of different nights from the first MATERHORN experiment in the autumn of 2012, however, revealed the occurrence of a transition front during several IOPs and a simultaneous downslope-flow onset along the slope during other nights (Pardyjak et al. 2014; Fernando et al. 2015). The movement of the shadow during the evening transition in the autumn, however, differed from the spring, which may explain the different behavior of the flow transition.
At the topmost site on the upper slope, the transition occurred much later than local sunset, at the same time as the transition farther down the slope, that is, after a large part of the slope was shaded and had started to cool. This behavior is reminiscent of the findings in Meteor Crater by Martínez Villagrasa et al. (2013), who observed an almost simultaneous transition at two sites on the crater sidewall after the decay of daytime convection. The terrain of the small (~1.2-km diameter) and completely enclosed Meteor Crater is generally very different from the approximately 3-km-long and comparatively open east sidewall of Granite Mountain. The canyonlike upper part of the slope, where the comparatively late transition occurred, is, however, similar in scale to the Meteor Crater.
b. Phase 2—Undisturbed nocturnal boundary layer
1) Near-surface temperatures and heat budget
After the initial rapid cooling, the near-surface temperature decrease at the lowest 1–2 m AGL was small until midnight, with a stronger temperature decrease at or above 3 m AGL (Fig. 6). For example, at ES2 and ES3, temperatures decreased by approximately 0.23° and 0.47°C h−1 at or below 1 m AGL between 2000 and 2300 MST, respectively. At ES5, temperatures below 2 m AGL even increased by approximately 0.3°C h−1 during this 3-h period. In a similar way, temperature changes at the PWIDS stations ranged from a 0.5°C h−1 cooling to a 0.27°C h−1 warming. The temperature decreases at, for example, 3 m AGL at ES2 and ES3, on the other hand, were 2.5° and 1.6°C, respectively.
Potential temperature storage, horizontal advection, and heat-flux divergence were calculated for the PWIDS stations, the towers, and the tethersonde, when possible (Fig. 11). The slope-parallel wind component usp was approximated by the east–west wind component parallel to the slope up, and the cross-slope wind component was approximated by the north–south component. When calculating horizontal temperature advection, it needs to be taken into account that small inaccuracies in the measurement heights could result in noticeable errors because of the strong vertical temperature gradient in the near-surface layer. On the basis of postexperiment measurements of the exact PWIDS temperature sensor heights, the measurement heights varied by about 30 cm. With a vertical temperature gradient of 1 K m−1, a typical downslope-wind component of 2 m s−1, and a distance of 0.5–1 km between two stations, the resulting maximum errors would be 0.6 × 10−3–1.2 × 10−3 K s−1. Temperature tendency was small at all sites, particularly before approximately 2130 MST, with values on the order of 10−3 K s−1. Afterward, the tendency increased slightly and became more variable. Temperature advection in the along-slope direction calculated between ES2 and ES3 and between pairs of PWIDS was generally negative; that is, downslope winds on the sidewall advected warmer air. Cross-slope temperature advection was very weak during this downslope-flow regime. Heat-flux divergence at the towers between 0.5 and 5 m AGL produced a cooling near the surface, with values on the order of 0.5 × 10−2–1 × 10−2 K s−1. The residual could only be calculated for ES2 and ES3 on the basis of the temperature storage, advection between the two towers, and the local heat-flux divergence. The residual tendency includes both radiative-flux divergence and slope-normal advection. Weak residual cooling on the order of 10−3 K s−1 was present before 2000 MST, which was followed by an approximately 1.5–2-h period of close-to-zero residual cooling. If it is assumed that the downslope flow is parallel to the underlying terrain, vertical temperature advection should be negligible. Previous studies reported radiative-flux divergences on the order of ±10−4 K s−1. For example, Mahrt et al. (2001) calculated a heating of 0.5 W m−2 (~3 × 10−4 K s−1) due to radiative-flux divergence within the lowest 1.7 m of a shallow gully for a case study of the 1999 Cooperative Atmosphere–Surface Exchange Study (CASES-99). Again during CASES-99, Sun et al. (2003) observed radiation differences of approximately −5 W m−2 (~1.1 × 10−4 K s−1) between 2 and 48 m AGL. Simulations by Hoch et al. (2011) produced height-independent radiative cooling rates of approximately 1.15 × 10−4 K s−1 in Meteor Crater. The weak residual cooling observed on the east slope of Granite Mountain thus agrees with these reports from previous studies.
(a) Potential temperature tendency at all PWIDS stations, at ES2 (1 m AGL), ES3 (1 m AGL), ES5 (2 m AGL), and from the tethersonde profile (2 m AGL). (b) Potential-temperature tendency due to horizontal advection at 1 m AGL between ES3 and ES2 and between pairs of PWIDS in the along-slope direction [(89 and 80), (90 and 89), (96 and 33), (79 and 72), (93 and 90), (93 and 01), (78 and 01), and (75 and 92)] and pairs of PWIDS in the cross-slope direction [(96 and 72), (96 and 93), (93 and 75), (01 and 92), (33 and 01), and (33 and 90)]. (c) Potential-temperature tendency due to heat-flux divergence at ES2 and ES3, calculated between 0.5 and 5 m AGL. (d) Residual at ES2 and ES3.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
(a) Potential temperature tendency at all PWIDS stations, at ES2 (1 m AGL), ES3 (1 m AGL), ES5 (2 m AGL), and from the tethersonde profile (2 m AGL). (b) Potential-temperature tendency due to horizontal advection at 1 m AGL between ES3 and ES2 and between pairs of PWIDS in the along-slope direction [(89 and 80), (90 and 89), (96 and 33), (79 and 72), (93 and 90), (93 and 01), (78 and 01), and (75 and 92)] and pairs of PWIDS in the cross-slope direction [(96 and 72), (96 and 93), (93 and 75), (01 and 92), (33 and 01), and (33 and 90)]. (c) Potential-temperature tendency due to heat-flux divergence at ES2 and ES3, calculated between 0.5 and 5 m AGL. (d) Residual at ES2 and ES3.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
(a) Potential temperature tendency at all PWIDS stations, at ES2 (1 m AGL), ES3 (1 m AGL), ES5 (2 m AGL), and from the tethersonde profile (2 m AGL). (b) Potential-temperature tendency due to horizontal advection at 1 m AGL between ES3 and ES2 and between pairs of PWIDS in the along-slope direction [(89 and 80), (90 and 89), (96 and 33), (79 and 72), (93 and 90), (93 and 01), (78 and 01), and (75 and 92)] and pairs of PWIDS in the cross-slope direction [(96 and 72), (96 and 93), (93 and 75), (01 and 92), (33 and 01), and (33 and 90)]. (c) Potential-temperature tendency due to heat-flux divergence at ES2 and ES3, calculated between 0.5 and 5 m AGL. (d) Residual at ES2 and ES3.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
2) Temperature and flow profiles
As already seen in the tethered-balloon profiles, the developing slope inversion layer remained shallow and slope-parallel during the first part of the night. The top of the inversion clearly lay below the top of towers ES2, ES3, and ES5, with the atmosphere away from the slope surface being mixed or close to isothermal (Fig. 12). The strongest part of the inversion layer (defined as ∂T/∂z > 0.3°C m−1) was extremely shallow, with depths between 5 and 10 m (Fig. 13b). The depth of this layer was thus relatively constant along the slope and did not change with time. Note that fewer temperature sensors were placed on ES5 than on ES2 and ES3 and that no measurements were made between 5 and 10 m. The tower and tethersonde temperatures converged at the top of the towers, indicating that the isotherms above the slope were horizontally homogeneous. An along-slope temperature profile can be produced by connecting the tower and PWIDS surface temperatures in Fig. 12. In contrast to the atmosphere above the slope surface, the along-slope temperature gradient was stable after approximately 1930 MST, with lower temperatures in the lower part of the slope. This means that the near-surface inversion must have been stronger on the lower slope. This is confirmed by the inversion strength shown in Fig. 13a, which is defined as the temperature difference between the top level of the inversion and 1 m AGL (interpolated at ES5).
Vertical profiles of (a) temperature, (b) downslope component of the wind us, and (c) cross-slope component of the wind υs for ES1–ES5, PWIDS (asterisks), and the slope tethersonde. Tower and PWIDS data have been averaged over a 15-min period, with the time in the figure indicating the ending time of the averaging period. Tethersonde data are from nonaveraged soundings during the 15-min period. Line color indicates the distance east from 307 km easting (see Fig. 1b).
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
Vertical profiles of (a) temperature, (b) downslope component of the wind us, and (c) cross-slope component of the wind υs for ES1–ES5, PWIDS (asterisks), and the slope tethersonde. Tower and PWIDS data have been averaged over a 15-min period, with the time in the figure indicating the ending time of the averaging period. Tethersonde data are from nonaveraged soundings during the 15-min period. Line color indicates the distance east from 307 km easting (see Fig. 1b).
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
Vertical profiles of (a) temperature, (b) downslope component of the wind us, and (c) cross-slope component of the wind υs for ES1–ES5, PWIDS (asterisks), and the slope tethersonde. Tower and PWIDS data have been averaged over a 15-min period, with the time in the figure indicating the ending time of the averaging period. Tethersonde data are from nonaveraged soundings during the 15-min period. Line color indicates the distance east from 307 km easting (see Fig. 1b).
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
Evening time series of (a) inversion strength, (b) inversion depth, (c) jet maximum downslope wind speed, (d) height of the downslope jet maximum, (e) vertical wind speed shear, (f) vertical directional wind shear, and (g) gradient Richardson number at ES1–ES5. The inversion top for (a) and (b) is defined as the tower level where dT/dz drops below 0.3°C m−1. Inversion strength is calculated as the temperature difference between the top of the inversion and 1 m (interpolated at ES5). Wind speed shear and directional wind shear in (e) and (f) are calculated between the 20-m tower level and the lowest tower level. Wind speed shear is calculated using total horizontal wind speed. Positive and negative directional shear indicate clockwise and counterclockwise turning with height, respectively. Rig is calculated using vertical temperature and wind gradients between the 5- and 10-m tower levels.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
Evening time series of (a) inversion strength, (b) inversion depth, (c) jet maximum downslope wind speed, (d) height of the downslope jet maximum, (e) vertical wind speed shear, (f) vertical directional wind shear, and (g) gradient Richardson number at ES1–ES5. The inversion top for (a) and (b) is defined as the tower level where dT/dz drops below 0.3°C m−1. Inversion strength is calculated as the temperature difference between the top of the inversion and 1 m (interpolated at ES5). Wind speed shear and directional wind shear in (e) and (f) are calculated between the 20-m tower level and the lowest tower level. Wind speed shear is calculated using total horizontal wind speed. Positive and negative directional shear indicate clockwise and counterclockwise turning with height, respectively. Rig is calculated using vertical temperature and wind gradients between the 5- and 10-m tower levels.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
Evening time series of (a) inversion strength, (b) inversion depth, (c) jet maximum downslope wind speed, (d) height of the downslope jet maximum, (e) vertical wind speed shear, (f) vertical directional wind shear, and (g) gradient Richardson number at ES1–ES5. The inversion top for (a) and (b) is defined as the tower level where dT/dz drops below 0.3°C m−1. Inversion strength is calculated as the temperature difference between the top of the inversion and 1 m (interpolated at ES5). Wind speed shear and directional wind shear in (e) and (f) are calculated between the 20-m tower level and the lowest tower level. Wind speed shear is calculated using total horizontal wind speed. Positive and negative directional shear indicate clockwise and counterclockwise turning with height, respectively. Rig is calculated using vertical temperature and wind gradients between the 5- and 10-m tower levels.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
Radial wind velocities from lidar scans up the slope at (a) 2124 and (b) 2224 MST. See Fig. 1b for the lidar location and scan path.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
Radial wind velocities from lidar scans up the slope at (a) 2124 and (b) 2224 MST. See Fig. 1b for the lidar location and scan path.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
Radial wind velocities from lidar scans up the slope at (a) 2124 and (b) 2224 MST. See Fig. 1b for the lidar location and scan path.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
3) Intermittent disturbances
Time series of wind direction in Fig. 6 show that the locally driven slope flows during this period were not completely undisturbed. On the basis of the wind direction, three disturbance events, which did not affect all five towers equally, can be identified before 2300 MST. The first event occurred shortly before 2000 MST, indicated by a brief change to a more northerly direction at ES2–ES4 while ES5 remained completely unaffected. Also, the wind direction close to the surface remained downslope at ES2–ES4, thus producing strong directional wind shear (Fig. 13f). During the same period, the wind speed dropped perceptibly, even at ES5, where the wind direction remained unaffected (Fig. 6). The northerly flow was also present in the 1957 MST tethersonde profile (Fig. 9). The tethersonde soundings generally recorded a thin layer of northerly flow during the transition from up-valley to down-valley flow. This layer of northerly flow was located between the layer of westerly downslope flow above the surface and the easterly flow aloft and was thus most of the time above the tops of the towers. At 1957 MST, however, the easterly flow aloft increased, reducing the downslope-flow-layer depth and bringing the thin layer of northerly flow closer to the surface. The origin of this layer of northerly flow, however, is not clear.
The second event occurred at approximately 2130 MST with a shift of the wind direction to a more southerly direction at all levels of ES2 and ES3. At the same time, the previously constant temperature started to oscillate slightly. Similar to the first event, the wind speed dropped during this event, although only at ES2–ES4. Wind speeds above the slope showed a perceptible increase between the 2108 sounding and the subsequent 2129 MST sounding (Fig. 9), suggesting that the disturbance at the surface was linked to a coupling of the near-surface atmosphere to the atmosphere aloft as a result of increased wind speed. This event lasted only for a very short time before the surface layer decoupled again from the atmosphere aloft and winds at ES2 and ES3 returned to their initial westerly direction. Another event followed immediately afterward, shortly after 2200 MST. This event produced another shift in the wind direction to a more southerly direction at the upper tower levels of ES2 and ES3, increasing the vertical directional wind shear. In contrast to the previous disturbance, it also affected the upper part of the slope at ES4 and ES5. At ES5, the wind direction even remained west-southwesterly for most of the night after this wind shift. This event also marked the end of the relatively undisturbed period. Another lidar scan at 2224 MST (Fig. 14b) also captured the change in flow pattern. In this scan, the shallow katabatic flow was replaced with a pocket of near-stagnant air over the slope that was topped by a deeper layer of westerly winds.
c. Phase 3—Sloshing valley inversion
Fernando et al. (2015) suggest a formation mechanism for the sloshing of the valley inversion over the Dugway Valley, which was observed during all quiescent IOPs of the MATERHORN autumn experiment. As southwesterly flows through the gap south of Granite Mountain converge with the southeasterly down-valley winds, the resulting easterly flow pushes the valley inversion up the slope, where it converges and interacts with the local slope winds (Hocut et al. 2014). After the subsequent retreat of the valley inversion, the process can repeat. The resulting oscillatory behavior of temperatures and wind over the slope, as the valley inversion immerses the slope and retreats again, dominated the night of IOP4 after approximately 2300 MST. Similar changes in temperature caused by the displacement of a valley cold-air pool have also been observed in the Salt Lake basin (Lareau et al. 2013).
After the onset of cooling over the valley floor, a southerly flow developed, which then turned to a more southeasterly direction between 2100 and 2200 MST. The downslope flow over the east sidewall ran out onto the valley floor, where it converged with the valley winds (Fig. 15a). Surface temperatures on the valley floor were much colder than over the slope at that time. As the flow over the valley turned to a more easterly direction, the effect of the valley inversion first reached ES1, where a sudden change in wind direction occurred from southwesterly to easterly, except for the lowest 1.95-m level, while the flow over the rest of the east-facing slope was still downslope (Fig. 6). Westerly downslope flows and easterly valley winds converged near the foot of the sidewall, and the easterly flow pushed the valley inversion up the slope. As the valley inversion pushed up the slope, it started to affect other slope sites, which cooled and developed a southwesterly flow (Figs. 6, 15b). Temperatures dropped around midnight at ES2 and ES3, as well as in the tethersonde profiles. ES5 higher up on the slope was affected less by the valley inversion, with only a small temperature oscillation occurring around 0200 MST. At the same time, tethersonde profiles also showed a change in the wind field near the surface, with a decrease of the 4–5 m s−1 southwesterly flow to less than 2 m s−1 (Fig. 5).
The 2-m temperature (colors) and wind fields (vectors) at (a) 2300, (b) 0000, (c) 0100, (d) 0200, (e), 0300, and (f) 0330 MST.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
The 2-m temperature (colors) and wind fields (vectors) at (a) 2300, (b) 0000, (c) 0100, (d) 0200, (e), 0300, and (f) 0330 MST.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
The 2-m temperature (colors) and wind fields (vectors) at (a) 2300, (b) 0000, (c) 0100, (d) 0200, (e), 0300, and (f) 0330 MST.
Citation: Journal of Applied Meteorology and Climatology 54, 4; 10.1175/JAMC-D-14-0223.1
At 0100 MST, the valley inversion was clearly extending up the sidewall (Fig. 15c). Surface temperatures along the slope were similar to temperatures on the valley floor while the wind direction on the slope had changed to northerly. The sounding at 0056 MST showed a 23-m deep inversion, which grew to a depth of 50 m over the next half hour (Fig. 5). Wind speeds were weak within this 50-m deep inversion layer. The following retreat of the valley inversion resulted in the redevelopment of a weak southwesterly downslope flow and rising temperatures. At 0200 MST, temperatures over the slope were again higher than over the valley floor (Fig. 15d). At the same time, the easterly flow over the valley turned more southeasterly. Increasing wind speeds over the slope during and after the retreat of the valley inversion may have contributed somewhat to the temperature increase through turbulent mixing, but the stability near the surface remained relatively constant during this warming. At 0300 MST winds over the valley floor started to increase again (Fig. 15e), and at 0330 MST the valley inversion pushed up the slope a second time (Fig. 15f), to retreat again 30 min later (not shown).
While temperatures above 1 m AGL at ES2 and ES3 showed successive temperature increases and decreases, temperatures at or below 1 m AGL remained low after the inversion retreated the first time. At ES2, the temperature away from the slope showed three distinct drops and subsequent increases before 0700 MST (Fig. 6). The decreases were generally more abrupt than the increases, and the duration of the tower being immersed in the valley cold pool decreased throughout the night. The first of the three periods lasted for almost 3 h, whereas the second one lasted for about 2 h and the last one only lasted for approximately 1.5 h. While the same cooling and warming periods can be distinguished at ES3, additional intermittent warming periods occurred around 0130 MST and around 0430 MST, suggesting that ES3 was at the edge of the cold pool that was pushed up the slope and that there were slight variations in the location of this border.
Together with the arrival of the valley cold pool, the surface winds along the slope generally turned to a northerly or northwesterly direction within the deepened slope inversion (Fig. 6). The cold pool also affected the wind profiles (Fig. 12). As the inversion layer started to encompass ES1–ES3, the jet profiles in the u component disappeared at ES2 and ES3. At ES1, an easterly jet developed in agreement with the easterly valley flow pushing up the slope.
5. Summary and conclusions
Observations were presented of the nocturnal boundary layer development at the foot of an east-facing sidewall of Granite Mountain at Dugway Proving Ground during IOP4 (11–12 May 2013) of the spring MATERHORN field experiment. During this IOP, the boundary layer development over the slope could be divided into three distinct phases, as described below.
Phase 1 is marked by an evening flow transition. Because of the local terrain characteristics, local sunset on the instrumented slope occurred first in the upper part of the slope around 1800 MST. In the lower part of the slope, the shadow propagated down the slope from northwest to southeast, arriving last at the foot of the slope around 1900 MST. The transition from daytime upslope (in the upper part of the slope) and up-valley (in the lower part of the slope) flows to nighttime downslope flows closely followed the propagation of the shadow, thus propagating down the slope from northwest to southeast. Various mechanisms for the evening flow transition on mountain slopes have been reported in the literature, including a simultaneous onset of the downslope flow along the entire slope (Fernando et al. 2013), a frontal propagation down the slope (Hunt et al. 2003), a transition that follows the direction of shadow propagation (Nadeau et al. 2013), and a transition that follows the decrease of daytime turbulent kinetic energy (Martínez Villagrasa et al. 2013). The observed flow transition on the east slope of Granite Mountain during IOP4 was consistent with the observations of Nadeau et al. (2013), who observed an upward shadow and flow-transition propagation on a west-facing slope. During several nights of the first MATERHORN experiment, however, the transition on the east sidewall of Granite Mountain occurred almost simultaneously along the entire slope (Pardyjak et al. 2014).
Phase 2 represents the undisturbed nocturnal slope boundary layer. The evening flow transition was followed by a period of almost undisturbed boundary layer development between approximately 1900 and 2230 MST. This period was characterized by little cooling of the near-surface temperatures over the slope and a shallow but strong slope-parallel inversion. An analysis of the heat budget close to the surface indicated that warm-air advection by downslope winds was mostly balanced by cooling resulting from heat-flux divergence. Downslope wind speed maxima during this period were between 2 and 4 m s−1 with a jet maximum at approximately 5 m AGL. The height and strength of the jet maximum were relatively constant along the slope, with somewhat lower wind speeds at the foot of the slope. The downslope flows ran down to the bottom of the slope and into the adjacent valley, where they converged with the southeasterly down-valley flows. The slope boundary layer was subject to three small disturbances caused by the coupling of the boundary layer to the atmosphere aloft, which affected wind and temperature fields along the slope to various degrees.
Phase 3 is characterized by a sloshing valley inversion. The slope faces a large valley to the east, which is surrounded by terrain to the east, south, southwest, and west. The basinlike terrain thus favors the formation of nocturnal cold-air pools that can affect the boundary layer over the east sidewall of Granite Mountain. After approximately 2300 MST, the inversion that had formed over the adjacent valley was repeatedly pushed up the slope and retreated again, producing an oscillatory behavior along the slope. When the inversion pushed up the slope, slope-surface temperatures dropped quickly, only to increase again as the valley inversion retreated from the slope. Westerly downslope flows were generally replaced by northerly winds on the slope during these events and then redeveloped between events. Similar sloshing of the valley inversion and interactions with the boundary layer over the slope were observed during quiescent IOPs of the autumn experiment (Fernando et al. 2015).
Acknowledgments
This paper is dedicated to the memory of Dr. Elford Astling, a long-time civilian employee at Dugway Proving Ground, who began many local meteorological studies at Dugway that culminated in the MATERHORN program. The research was funded by the Office of Naval Research Award N00014-11-1-0709, Mountain Terrain Atmospheric Modeling and Observations (MATERHORN) Program. University of Utah students A. Charland and B. Blaylock assisted with tethersonde ascents, and M. Jeglum helped with data access. Further, we thank three anonymous reviewers for their helpful comments on the manuscript. ERA-Interim data were obtained through the European Centre for Medium-Range Weather Forecasts data archive.
REFERENCES
Barr, S., and M. M. Orgill, 1989: Influence of external meteorology on nocturnal valley drainage winds. J. Appl. Meteor., 28, 497–517, doi:10.1175/1520-0450(1989)028<0497:IOEMON>2.0.CO;2.
Bica, B., and Coauthors, 2007: Thermally and dynamically induced pressure features over complex terrain from high-resolution analysis. J. Appl. Meteor. Climatol., 46, 50–65, doi:10.1175/JAM2418.1.
Brazel, A. J., H. J. S. Fernando, J. C. R. Hunt, N. Selover, B. C. Hedquist, and E. Pardyjak, 2005: Evening transition observations in Phoenix, Arizona. J. Appl. Meteor., 44, 99–112, doi:10.1175/JAM-2180.1.
Catalano, F., and A. Cenedese, 2010: High-resolution numerical modeling of thermally driven slope winds in a valley with strong capping. J. Appl. Meteor. Climatol., 49, 1859–1880, doi:10.1175/2010JAMC2385.1.
Dee, D. P., and Coauthors, 2011: The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553–597, doi:10.1002/qj.828.
Doran, J. C., T. W. Horst, and C. D. Whiteman, 1990: The development and structure of nocturnal slope winds in a simple valley. Bound.-Layer Meteor., 52, 41–68, doi:10.1007/BF00123177.
Fernando, H. J. S., B. Verhoef, S. Di Sabatino, L. S. Leo, and S. Park, 2013: The Phoenix Evening Transition Flow Experiment (TRANSFLEX). Bound.-Layer Meteor., 147, 443–468, doi:10.1007/s10546-012-9795-5.
Fernando, H. J. S., and Coauthors, 2015: The MATERHORN—Unraveling the intricacies of mountain weather. Bull. Amer. Meteor. Soc., doi:10.1175/BAMS-D-13-00131.1, in press.
Fritts, D. C., D. Goldstein, and T. Lund, 2010: High-resolution numerical studies of stable boundary layer flows in a closed basin: Evolution of steady and oscillatory flows in an axisymmetric Arizona Meteor Crater. J. Geophys. Res., 115, D18109, doi:10.1029/2009JD013359.
Gesch, D., M. Oimoen, S. Greenlee, C. Nelson, M. Steuck, and D. Tyler, 2002: The National Elevation Dataset. Photogramm. Eng. Remote Sensing, 68, 5–11.
Haiden, T., and C. D. Whiteman, 2005: Katabatic flow mechanisms on a low-angle slope. J. Appl. Meteor., 44, 113–126, doi:10.1175/JAM-2182.1.
Hoch, S. W., C. D. Whiteman, and B. Mayer, 2011: A systematic study of longwave radiative heating and cooling within valleys and basins using a three-dimensional radiative transfer model. J. Appl. Meteor. Climatol., 50, 2473–2489, doi:10.1175/JAMC-D-11-083.1.
Hocut, C. M., and Coauthors, 2014: Slope and valley flow interactions in MATERHORN-1. 18th Joint Conf. on the Applications of Air Pollution Meteorology with the A&WMA, Atlanta, GA, Amer. Meteor. Soc., 15.5. [Available online at https://ams.confex.com/ams/94Annual/webprogram/Paper237292.html.]
Hunt, J. C. R., H. J. S. Fernando, and M. Princevac, 2003: Unsteady thermally driven flows on gentle slopes. J. Atmos. Sci., 60, 2169–2182, doi:10.1175/1520-0469(2003)060<2169:UTDFOG>2.0.CO;2.
Lareau, N. P., E. Crosman, C. D. Whiteman, J. D. Horel, S. W. Hoch, W. O. J. Brown, and T. W. Horst, 2013: The Persistent Cold-Air Pool Study. Bull. Amer. Meteor. Soc., 94, 51–63, doi:10.1175/BAMS-D-11-00255.1.
Leo, L. S., S. Di Sabatino, A. Grachev, H. J. S. Fernando, C. M. Hocut, E. R. Pardyjak, and D. Jensen, 2014: Structure and dynamics of katabatic flows: results from MATERHORN X-1. 18th Joint Conf. on the Applications of Air Pollution Meteorology with the A&WMA, Atlanta, GA, Amer. Meteor. Soc., 15.4. [Available online at https://ams.confex.com/ams/94Annual/webprogram/Paper236027.html.]
Lugauer, M., and P. Winkler, 2005: Thermal circulation in South Bavaria—Climatology and synoptic aspects. Meteor. Z., 14, 15–30, doi:10.1127/0941-2948/2005/0014-0015.
Mahrt, L., D. Vickers, R. Nakamura, M. R. Soler, J. Sun, S. Burns, and D. H. Lenschow, 2001: Shallow drainage flows. Bound.-Layer Meteor., 101, 243–260, doi:10.1023/A:1019273314378.
Mahrt, L., S. Richardson, N. Seaman, and D. Stauffer, 2010: Non-stationary drainage flows and motions in the cold pool. Tellus, 62A, 698–705, doi:10.1111/j.1600-0870.2010.00473.x.
Mahrt, L., J. Sun, S. P. Oncley, and T. W. Horst, 2014: Transient cold air drainage down a shallow valley. J. Atmos. Sci., 71, 2534–2544, doi:10.1175/JAS-D-14-0010.1.
Martínez Villagrasa, D., M. Lehner, C. D. Whiteman, S. W. Hoch, and J. Cuxart, 2013: The upslope–downslope flow transition on a basin sidewall. J. Appl. Meteor. Climatol., 52, 2715–2734, doi:10.1175/JAMC-D-13-049.1.
Nadeau, D. F., E. R. Pardyjak, C. W. Higgins, H. Huwald, and M. B. Parlange, 2013: Flow during the evening transition over steep Alpine slopes. Quart. J. Roy. Meteor. Soc., 139, 607–624, doi:10.1002/qj.1985.
Oerlemans, J., and B. Grisogono, 2002: Glacier winds and parameterisation of the related surface heat fluxes. Tellus, 54A, 440–452, doi:10.1034/j.1600-0870.2002.201398.x.
Oldroyd, H. J., G. Katul, E. R. Pardyjak, and M. B. Parlange, 2014: Momentum balance of katabatic flow on steep slopes covered with short vegetation. Geophys. Res. Lett., 41, 4761–4768, doi:10.1002/2014GL060313.
Papadopoulos, K. H., and C. G. Helmis, 1999: Evening and morning transition of katabatic flows. Bound.-Layer Meteor., 92, 195–227, doi:10.1023/A:1002070526425.
Pardyjak, E. R., H. J. S. Fernando, J. C. R. Hunt, A. A. Grachev, and J. Anderson, 2009: A case study of the development of nocturnal slope flows in a wide open valley and associated air quality implications. Meteor. Z., 18, 85–100, doi:10.1127/0941-2948/2009/362.
Pardyjak, E. R., and Coauthors, 2014: Evening transition characteristics on a slope in an arid environment. 21st Symp. on Boundary Layers and Turbulence, Leeds, United Kingdom, Amer. Meteor. Soc., 13B.8. [Available online at https://ams.confex.com/ams/21BLT/webprogram/Paper248697.html.]
Skyllingstad, E. D., 2003: Large-eddy simulation of katabatic flows. Bound.-Layer Meteor., 106, 217–243, doi:10.1023/A:1021142828676.
Sun, J., S. P. Burns, A. C. Delany, S. P. Oncley, T. W. Horst, and D. H. Lenschow, 2003: Heat balance in the nocturnal boundary layer during CASES-99. J. Appl. Meteor., 42, 1649–1666, doi:10.1175/1520-0450(2003)042<1649:HBITNB>2.0.CO;2.
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., and S. Zhong, 2008: Downslope flows on a low-angle slope and their interactions with valley inversions. Part I: Observations. J. Appl. Meteor. Climatol., 47, 2023–2038, doi:10.1175/2007JAMC1669.1.
Whiteman, C. D., S. Zhong, W. J. Shaw, J. M. Hubbe, X. Bian, and J. Mittelstadt, 2001: Cold pools in the Columbia basin. Wea. Forecasting, 16, 432–447, doi:10.1175/1520-0434(2001)016<0432:CPITCB>2.0.CO;2.
Zardi, D., and C. D. Whiteman, 2012: Diurnal mountain wind systems. Mountain Weather Research and Forecasting, F. K. Chow, S. F. J. De Wekker, and B. Snyder, Eds., Springer, 35–119, doi:10.1007/978-94-007-4098-3_2.
