Cite this Article
WANG Lin-Lin, GUO Xiao-Feng, WAN Bing-Cheng. Characteristics of Thermally-Induced Near-Surface Flows over An Enclosed Crater: Observations of the Meteor Crater Experiment (METCRAX). Atmospheric and Oceanic Science Letters, 2014, 7(2): 162-167  
Permissions
Copyright?2014, Editorial office of Atmospheric and Oceanic Science Letters
This is an Open Access article under the terms of CCAL.
Characteristics of Thermally-Induced Near-Surface Flows over An Enclosed Crater: Observations of the Meteor Crater Experiment (METCRAX)
WANG Lin-Lin, GUO Xiao-Feng, WAN Bing-Cheng
State Key Laboratory of Atmospheric Boundary Layer Physics and Atmospheric Chemistry, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing 100083, China
Corresponding author: WANG Lin-Lin,linlinwang@mail.iap.ac.cn
Abstract

Meteorological measurements of the Meteor Crater Experiment (METCRAX) were used to analyze the characteristics of the thermally-induced flows over an enclosed crater (Arizona, USA). Despite its relatively small size, this crater acts as an effective wind shelter. Daytime upslope winds and nocturnal downslope flows are regularly observed on its west/east sidewalls. The times of diurnal wind-direction shift (i.e., around sunrise and sunset) were slightly mismatched between the sites based on the east and west sidewalls, owing to the crater’s shadowing effects. Under conditions of relatively weak synoptic forcing, the nocturnal downslope flows prevail throughout the night, despite quite a low wind velocity near the surface.

Keyword: atmospheric physics; mountain meteorology; local thermal circulation; diurnal behavior
1 Introduction

Nocturnal downslope flows are a characteristic phenomenon over mountainous areas, and these types of airflow play an important role in the atmospheric transport and dispersion processes over complex terrains, such as mountains, valleys, and basins. Field observational data available from various areas of complex terrain have hitherto led to a fundamental understanding of the peculiar characteristics of downslope flows—formation mechanisms, interactions with large-scale circulation, and typical response to specific terrain configurations ( Papadopoulos and Helmis, 1999; Haiden and Whiteman, 2005; Witheman and Zhong, 2008; Oltmanns et al., 2013). For instance, Mahrt (1982) calculated different terms of the momentum equation, and by examining the dominant forcing mechanisms of the downslope flows, proposed new formulations to enhance their physical description, which incorporated both advective and frictional effects. Manins (1992) found that the vertical divergence of the radiative flux was an important factor in generating static and dynamic instability in downslope flows, which led to the argument that the radiative cooling in their momentum budget should not be neglected. Zhong et al. (2008) found that the daytime downslope wind system in the lee of the Sierra Nevada, USA was driven primarily by a regional- scale pressure gradient that developed because of the asymmetric heating of the atmosphere. They also found that regional westerly winds over the Sierra Nevada were not a prerequisite for the development of the daytime downslope flows. In the literature for mountain meteorology, the majority of observational analyses are devoted to investigations over mountain slopes and large-scale slopes. There have been, nonetheless, very few studies that addressed relatively gentle or small-scale topographic slopes, but these less-studied environments represent an important area of research because turbulent fluxes within shallow downslope flows are critical to the land surface- atmosphere exchange. For instance, turbulent heat exchange could have a modulating effect on the minimum nocturnal air temperature, as well as the dispersion of air pollutants in the surface boundary layer ( Soler et al., 2002).

In October 2006, the Meteor Crater Experiment (METCRAX) was conducted to investigate the multi-scale structure and temporal evolution of near-surface temperature inversions and cold-air pools that commonly form over an enclosed crater. Previous analyses of the dataset have relied mainly on observations made outside the meteor crater; for example, Savage et al. (2008) analyzed the data of an Integrated Sounding System (ISS) installed about 5 km away from the crater. In our work, we focused on meteorological measurements made inside the crater to gain a further understanding of the characteristics of thermally-induced near-surface flows over such a small, enclosed basin.

2 Experimental sites and measurements

The meteor crater (35.07°N, 111.03°W) is located about 40 km east of Flagstaff in northern Arizona, and was created about 50000 years ago during the Pleistocene epoch. The crater is approximately 160 m in depth and 1200 m in diameter at the rim (about 50 m above the mean level of the surrounding plain over the Colorado Plateau) (see Fig. 1 for a field photograph). The crater is largely an axisymmetric circular basin, though enhanced erosion along two faults running southwest-to-northeast across the crater floor have created a slightly squared-off geometry ( Whiteman et al., 2008). The crater floors and sidewalls are composed primarily of rocks that are sparsely covered with short grass and small bushes.

Figure 1 Photograph of the crater used in the Meteor Crater Experiment (METCRAX).

During the METCRAX experiment, meteorological observations were based both inside and outside the cra- ter, in order to capture a full picture of the wind system and concurrent synoptic conditions. One measurement site was located about 2.5 km to the southwest of the crater (site ID: SW), and equipped with a 10-m meteorological tower and a mini-sodar with a Radio Acoustic Sounding System (RASS). Another site was located 5 km north- northeast of the crater and was equipped with the ISS composed of a 915-MHz Radar wind profiler with RASS and an enhanced surface meteorology tower (ISS site). Inside the crater, an array of five micrometeorological flux towers, or Integrated Surface Flux Facility (ISFF), were installed to make continuous measurements of mean meteorological parameters, turbulent fluxes, and the radiative energy budget. These towers were located in an east-west cross section with three 9-m towers positioned either near to the center of the crater floor (site ID: FLR) or adjacent to the lower east and west sidewalls (site IDs: EL and WL), and two 6-m towers adjacent to the upper slopes of the east and west sidewalls (site IDs: EU and WU). Related information about the tower sites and meteorological observations is listed in Table 1. In brief, the five towers inside the crater were all equipped with hygrothermometers (10-Hz data) and 3-D sonic anemometers (20-Hz data) at multiple heights (see Table 1). Additionally, a 10-m tripod was installed near the north rim of the crater (site ID: RIM) to collect data from higher above the crater.

Table 1 Geographic information of the measurement sites and meteorological observations.

Prior to the related analyses, turbulence data from the 3D sonic anemometers were corrected for tilt by applying the planar fit technique, with wind-vector rotation angles determined using the 5-min-averaged wind velocities; subsequently, raw data of turbulence were rotated into a surface-parallel, geographic coordinate system. Ideally,the transformed coordinate system renders the overall averaged vertical velocity to be negligible at the individual sites. Parenthetically, the U wind component is defined herein to be positive from west to east and the V component positive from south to north.

3 Results
3.1 Characteristics of near-surface winds inside the crater

Characteristics of the daytime and nighttime near- surface winds inside the crater are first illustrated by wind frequency distributions; namely, the percentage of 5-min runs per 10° (wind roses) and 1 m s-1 (histograms) intervals. Figure 2 shows the results for the six tower sites, i.e., at a height of 3 m for those adjacent to the crater’s sidewalls (EL, WL, EU, and WU) and at a height of 2 m for those on the crater floor (FLR) and near the crater’s rim (RIM).

During the daytime, the near-surface winds measured at the four sites (Fig. 2a) saw a wide range of wind directions, while those measured adjacent to the sites at the crater’s west sidewalls were predominantly from the northeast (i.e., northeasterlies). Moreover, the wind speeds measured inside the crater (falling into the interval of 0-2 m s-1 for over 70% of the runs) were markedly less than those measured outside, namely near the crater’s rim where wind speeds exceeded 2 m s-1 for over 70% of the time. Therefore, the crater, even though it is relatively small and shallow, acts as an effective wind shelter.

Figure 2 Wind frequency distributions by means of (a, b) wind roses and (c, d) histograms separately for the daytime, 0600-1800 MST (Mountain Standard Time), and the nighttime, 1800-0600 MST. See Table 1 for information about the different measurement sites.

Nocturnal wind directions measured at most of the sites covered a narrower range compared to those observed during the daytime, except for one site on the FLR site. Specifically, nocturnal winds were most frequently from the west (i.e., westerlies) at the sites near the crater’s RIM site and the sites on the west sidewalls, while being from the east (i.e., easterlies) at those near the sites on the east sidewalls. At the FLR site, the dominant wind directions were less discernible, as they covered an extended range of distribution. Overall, the nocturnal wind speeds appeared to be weaker than those during the day; for instance, more than 70% of the measurements were less than 1 m s-1 inside the crater, while those observed on the RIM site were similarly weak, i.e., less than 3 m s-1 for over 80% of the runs. Meanwhile, moderately strong southwesterly low-level winds (4-6 m s-1) were often observed outside the crater; namely, by the tower instruments and rawinsonde soundings at the ISS site and the mini-sodar at the SW site (figures omitted). Such relatively weak synoptic flows could facilitate the development of nocturnal downslope flows, as more frequently observed near the east sidewalls (see Fig. 2).

Diurnal variations of the near-surface wind pattern are further examined in Fig. 3, which shows daytime cross sections of the 15-min averaged along-slope wind component ( U in m s-1) at the 3-m height on the five meteorological towers. The white-colored areas indicate data gaps due to instrument malfunction. Although inter-daily variations were significant, the along-slope wind speeds were generally weak (mostly within the range of 1-3 m s-1). During the METCRAX experiment, the sign of the measured U components was typically opposite between the sites adjacent to the east and west sidewalls of the crater, which indicated diurnally shifting, thermally- induced wind features, namely upslope winds during the day, and downslope winds during the night. Interestingly, such an evident diurnal behavior observed at the four sidewall sites was largely absent from the site on the FLR.

In Fig. 3, several transient wind events are also noticeable during the experiment. For instance, the wind directions were consistently from the west at these five sites from approximately 1700 to 2400 MST (Mountain Standard Time) 17 October 2006. During this period, the measured wind speeds had a spatial variation of an apparent increase from the WU to the WL, reaching higher values on the FLR, before decreasing towards the east sidewalls. Such short-term wind events were similarly found on both 4 and 25 October. For these atypical transient events, downslope flows were largely not found to develop on the east side of the crater, possibly due to the enhanced effects of fairly strong synoptic airflows from the west.

Figure 3 Daytime cross sections of the along-slope wind velocity component (m s-1) as measured at the 3 m-level at the four sidewalls towers: (a) WU, (b) WL, (d) EL, (e) EU, and (c) 2 m-level on the FLR tower.

An additional point of interest arises from Fig. 3 regarding the somewhat mismatched times of diurnal wind- direction shift between the sites adjacent to the east and west sidewalls of the crater. Specifically, the wind directions as observed to the west shifted from westerly to easterly typically at about 0500 MST, nearly one hour earlier than the time when the winds on the opposing side (EL and EU) shifted from easterly to westerly in the early morning. Furthermore, the reverse commonly occurred at the time of sunset: the wind directions to the west changed from easterly to westerly at about 1700 MST, nearly one hour earlier than the time when the winds to the east shifted from westerly to easterly at times of sunset. From an intuitive perspective, the above findings should result from the spatially non-uniform heating associated with the topographic shadowing effects incurred near times of sunrise and sunset.

3.2 Case study

Nocturnal downslope flows typically developed when the buoyancy-induced acceleration due to the near-surface air cooling exceeded the opposing effects due to the along-slope pressure gradient. Meanwhile, the downslope flows could be attenuated by ambient synoptic conditions.

Relatively weak synoptic forcing and a clear night sky were conducive to the formation of nocturnal surface- based temperature inversions that produced the negative buoyancy needed to drive the downslope flows. As shown in Fig. 3, the downslope flows were well developed at both sides inside the crater (WL and WU to the west; EL and EU to the east). For a brief case study below, our focus is on measurements made on the east side, given that airflows observed on the west sidewalls were influenced more often by the regional-scale southwest drainage flows during the experiment.

The approach of Yao and Zhong (2009) is adopted herein for the brief case study. The integrated thermal-inversion strength was calculated to characterize the temperature inversions inside the crater; in this way, different atmospheric conditions could be described. The integrated inversion strength ( I) was formulated as

, (1)

where t1 and t2 are the approximate times of sunset and sunrise, respectively, as observed inside the meteor crater; and and TFLR are the average temperatures measured on the upper sidewalls and the temperature measured on the FLR, respectively; all these variables were measured at a height of 0.5 m. The time step for the numerical integration was 5 min. In the following analysis, we use the calculated integrated inversion strength as an alternative measure of thermal forcings of the downslope flow. Owing to the relatively weak synoptic weather systems experienced and the fairly strong inversion strength ( I > 6000°C min), atmospheric conditions on the night of 22-23 October were particularly conducive to the development of nocturnal downslope flows. Therefore, this time period is addressed below in our brief case study.

We first consulted the occurrence frequency of the near- surface wind maximum for the different measurement heights at the EL and EU sites (22-23 October), and found that, for either site, the wind maximum was recognized at the height of 0.5 m for over 50% of the time. Moreover, the wind maximum at EL was found at the height of 8.5 m for about 25% of the time, with that at EU occurring at the height of 1.5 m for about 32% of the time. Therefore, in spite of favorable synoptic conditions with optimal thermal forcings (namely, negative buoyancy), the nocturnal downslope flows had a peak velocity extremely close to the crater’s surface, which was a phenomenon probably indicative of a fairly shallow layer of air constituted by the downslope flows.

We now proceed to a succinct description of the evolving meteorological parameters to conclude this case study. The 5-min averaged measurements for the 0.5-m height at the EL site are presented in Fig. 4, including components of the longwave radiation, air temperature, wind direction, and the along-slope velocity. It is noticeable that both the outgoing longwave radiation and the air temperature decreased progressively after sunset (22 October), and the former reached its minimum (about 300 W m-2) in concurrence with the minimum air temperature; both variables increased dramatically after the time of sunrise. The incoming longwave radiation varied little throughout the night, showing a fairly steady value of about 250 W m-2. It is interesting to note that the wind directions showed a remarkable shift, namely from the westerlies to easterlies at roughly one hour after sunset; subsequently, the along-slope wind velocity component ( U) indicated airflows persistently from the east sector during the night ( U < 0), except for occasional, transient turbulent mixing events. These observed nocturnal downslope flows could even persist for about two hours after sunrise.

Figure 4 Time series of 5-min measurements at the EL site from 1600 MST ( 22 October 2006) to 1030 MST ( 23 October 2006): (a) incoming and outgoing longwave radiation; (b) air temperature at 0.5 m; (c) wind direction at 0.5 m; and (d) along-slope wind velocity component at 0.5 m.

In Fig. 4d, the downslope wind velocities are quite low overall, mostly falling into the range of 0.2-0.8 m s-1. It should be noted that the downslope flow was significantly diminished around 2300 MST during a very short period with gradual increases in the air temperature (roughly 3 K within one hour or so); however, following such a transient warming episode, the downslope flow resumed its strength, with velocities rising above 1 m s-1. Concurrent warming episodes were detected below a height of 3.0 m at both the WL and EL sites and below a height of 5.0 m at the FLR site, which were not found at the upper sidewalls (figures omitted). During the warming episode mentioned above, the incoming and outgoing components of the longwave radiation did not show discernible variations. Therefore, the observed gradual increases in air temperature were not likely to result from possible modulations of the radiative energy budget, but some other causes, such as vertical mixing in the stably-stratified nocturnal boundary layer over the crater and its surroundings (further examinations to be presented in a future investigation).

4 Conclusions

Meteorological measurements available from METCRAX were used to analyze the characteristics of the thermally- induced near-surface flows inside and adjacent to an enclosed crater (Arizona, USA). Despite its relatively small size, this crater was found to act as an effective wind shelter; and, daytime upslope winds and nocturnal downslope flows were regularly observed on its west/east sidewalls. It is also interesting to note the slightly mismatched times of diurnal wind-direction shift (i.e., around sunrise and sunset) between the sites based on the east (EL and EU) and west (WL and WU) sidewalls, which suggest spatially non-uniform heating due to the crater’s shadowing effects. Under conditions of relatively weak synoptic forcing, the nocturnal downslope flows were found to occur throughout the night, though they typically have quite a low wind velocity (< 1 m s-1 near the surface) and presumably constitute a shallow layer. The nocturnal downslope flows do not occur concurrently on the west/ east sidewalls of this crater, when synoptic forcings are enhanced.

Reference
1 Haiden T. , C. D. Whiteman, 2005: Katabatic flow mechanisms on a low-angle slope, J. Appl. Meteor. , 44, 113-126.
2 Manins P. C. , 1992: Vertical fluxes in katabatic flows, Bound. -Layer Meteor. , 60, 169-178.
3 Mahrt L. , 1982: Momentum balance of gravity flows, J. Atmos. Sci. , 39, 2701-2711.
4 Oltmanns M. , F. Straneo, K. Moore, et al. , 2013: Strong downslope wind events in Ammassalik, SE Greenland , J. Climate, in press, doi: 10.1175/JCLI-D-13-00067.1.
5 Papadopoulos K. H. , C. G. helmis, 1999: Evening and morning transition of Katabolic flows, Bound. -Layer Meteor. , 92, 195-227.
6 Savage, L. C. III, S. Zhong, W. Yao, et al. , 2008: An observational and numerical study of a reginal-scale downslope flow in Northern Arizona, J. Geophys. Res. , 113, D14114, doi: 10.1029/2007JD009623.
7 Soler M. R. , C. Infante, P. Buenesttado, et al. , 2002: Observation of nocturnal drainage flow in a shallow gully, Bound. -Layer Meteor. , 105, 253-273.
8 Whiteman C. D. , S. Zhong, 2008: Downslope flows on a low angle slope and their interations with valley inversions. Part I: Observation, J. Appl. Meteor. Climate, 47, 2023-2038.
9 Whiteman C. D. , A. Muschinski, S. Zhong, et al. , 2008: METCRAX 2006—meteorological experiment in Arizona’s meteor crater, Bull. Amer. Meteor. Soc. , 89, 1665-1680.
10 Yao W. , S. Zhong, 2009: Nocturnal temperature inversions in a small, enclosed basin and their relationship to ambient atmospheric conditions, Meteor. Atmos. Phys. , 103, 195-210.
11 Zhong S. , J. Li, C. B. Clements, et al. , 2008: Forcing mechanisms for washoe zephyr—a daytime downslope wind system in the lee of the Sierra Nevada, J. Appl. Meteor. Climate, 47, 339-305.