UBC Faculty Research and Publications

Boundary Layer Experiment 1996 (BLX96). Berg, Larry K.; Santoso, Edi; Stull, Roland B.; Hacker, Joshua P. 1997-06-30

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata


Stull_AMS_1996_BAMS1149.pdf [ 195.64kB ]
JSON: 1.0041838.json
JSON-LD: 1.0041838+ld.json
RDF/XML (Pretty): 1.0041838.xml
RDF/JSON: 1.0041838+rdf.json
Turtle: 1.0041838+rdf-turtle.txt
N-Triples: 1.0041838+rdf-ntriples.txt
Original Record: 1.0041838 +original-record.json
Full Text

Full Text

1149Bulletin of the American Meteorological Society1. IntroductionDuring 15 July–13 August 1996, The Universityof British Columbia (UBC) conducted a boundarylayer field experiment (BLX96) in Oklahoma andKansas. BLX96 consisted of three subexperimentswithin the daytime convective mixed layer (ML):• Convective transport theory (CTT),• Radix layer (RxL), and• Boundary-layer cumulus (BLCu).The University of Wyoming (UW) King Airaircraft (N2UW) was the primary instrument plat-form for this experiment, based at Ponca City, Okla-homa (36°44.0′N, 97°06.0′W).BLX96 coincided with a routine U.S. Depart-ment of Energy (DOE) Atmospheric RadiationMeasurement (ARM) Program intensive operationperiod, conducted from 15 July to 4 August 1996at the Cloud and Radiation Testbed (CART) regionin the southern Great Plains (SGP). The central fa-cility (36.605°N, 97.485°W, elevation 318 m) ofCART is near Lamont, Oklahoma, although the full143 000-km2 CART region spans parts of Kansasand Oklahoma. Collaborators from Argonne Na-tional Laboratory (ANL) and the National Centerfor Atmospheric Research (NCAR) broughtground-based remote sensors into the CART re-gion, near the BLX96 flight tracks.a. Convective transport theoryDuring free convective conditions of light windsand statically unstable air, vertical transport iscaused primarily by thermals. The convective trans-port velocity (Stull 1994) wB = (βzi∆θv)1/2 is a mea-sure of this transport rate, where β = g/Tv is thebuoyancy parameter, g is gravitational acceleration,Tv is virtual absolute temperature in the mid-ML, zi isthe average ML depth, and ∆θv is the virtual potentialtemperature difference between the surface skin andthe mid-ML. Surface fluxes under these conditions areexpected to be proportional to this buoyancy velocitytimes the difference in mean condition between thesurface skin and the mid-ML; for example, w′θ ′ sBoundary Layer Experiment 1996(BLX96)Roland Stull, Edi Santoso, Larry Berg, and Joshua HackerAtmospheric Science Programme, Department of Geography,The University of British Columbia, Vancouver, British Columbia, CanadaABSTRACTThe University of Wyoming King Air aircraft was the primary instrument platform for turbulence measurements in thebottom half of the convective boundary layer during 15 July–13 August 1996. A total of 12 successful researchflights were made, each of about 4.5-h duration. Crosswind (east–west) flight patterns were flown in Oklahoma andKansas over three sites of different land use: forest, pasture, and crops.Measurements of mean values, turbulent deviations, and turbulent fluxes of temperature, moisture, and momentumwere made to test theories of convective transport, the radix layer, and cumulus potential. Additional portions of each flightincluded slant soundings and near-surface horizontal flights in order to determine mixed layer (ML) scaling variables suchas ML depth zi, Deardorff velocity w*, and buoyancy velocity wB. While the ML was shallower and the ground wetter thananticipated based on climatology, a high-quality dataset was obtained.Corresponding author address: Roland B. Stull, AtmosphericScience Programme, Dept. of Geography, UBC, 1984 West Mall,Vancouver, BC V6T 1Z2, Canada.E-mail: rstull@geog.ubc.caIn final form 12 December 1996.©1997 American Meteorological Society1150 Vol. 78, No. 6, June 1997= bHwB∆θ, where empirical constant bH ≈ 0.0005,and ∆θ = θskin − θmidML.Convective transport theory (CTT) predicts thatsurface fluxes should be independent of roughnesslength for free convection, where free convectionoccurs approximately when the mixed layerRichardson number, R* = (wB/MML)2, is greater than3, where MML is mid-ML wind speed. It also suggestsa diagnostic way to estimate ML depth from surfacefluxes and skin–ML temperature differences.Extensions to this theory allow for nonzero mean wind.To test CTT and its empirical parameters, BLX96 in-cluded measurements of eddy-correlation fluxes overthree sites with different aerodynamic roughness. Adownward-looking infrared (IR) radiation thermom-eter on the aircraft remotely measured surface skin tem-perature. The aircraft also made in situ measurementsof wind, temperature, and humidity in the mid-ML.b. Radix layerFor free convection, the ML can be divided intolayers identified by shapes of the mean profiles.Traditionally, the surface layer (SL) is defined asthe region where 1) turbulent fluxes are roughlyconstant with height near the surface, 2) Monin–Obukhov similarity theory applies, 3) wind profilesare nearly logarithmic, and 4) mechanically gen-erated turbulence is important. In the interior of theML is a uniform layer (UL) where wind speed andpotential temperature are approximately constantwith height and where buoyantly generated turbu-lence dominates. However, there is often a largegap between the top of the SL and the bottom ofthe UL, which is also the region where only a lim-ited amount of field data exists. Within this gap, freeconvection scaling can apply (Holtslag andNieuwstadt 1986; Panofsky 1978).A radix layer (RxL) has been identified bySantoso (1993) and Santoso and Stull (1997a,manuscript submitted to J. Atmos. Sci.) as the wholeregion between the surface and the base of the UL(Fig. 1). The classic SL and classic free convectionscaling layer are subsets of the RxL. At the top ofthe RxL, the wind speed becomes tangent to ULwind speed. Typical RxL depths are on the orderof 100–300 m for wind speed and 20–70 m forpotential temperature. The Latin name “radix,”meaning root or base, was suggested because theRxL is at the root of thermals.To investigate the RxL, a vertical zig-zag flightpattern was flown to provide vertical profiles ofmean wind, temperature, and humidity betweenaltitudes of about 5 and 700 m above ground level(AGL). This spans the SL, the RxL, and the bottomof the UL. In addition, higher sounding legs wereflown to get zi, and near-surface level legs were flownto get eddy-correlation fluxes from which theDeardorff {w* = [(g/Tv) zi w′θv′ s]1/3} and buoyancy(wB) velocities could be found, where w′θv′ s is thevertical eddy flux of virtual potential temperatureat the surface. To test dependency of the RxL tosurface roughness and land use, three regions ofdiffering land use were chosen for the flight tracks.c. Boundary layer cumulusHeterogeneous land surfaces cause small differ-ences in near-surface air temperature and humid-ity, which modulate characteristics of risingthermals and control the formation of boundarylayer cumulus (BLCu). Air parcels born over avariety of surfaces tend to rise to a range of heights,depending on their buoyancies. Buoyancy can bequantified by the virtual potential temperature (θv)difference between the parcel and the ambient θvprofile. The probability of a parcel to form a cu-mulus cloud also depends on its individual liftingcondensation level (LCL) height (zLCL).The joint frequency distribution (JFD) of θvversus zLCL can be used as a predictor of BLCu(Schrieber et al. 1996). Such a JFD consists of threesectors (Fig. 2). Parcels that have a θv less than theFIG. 1. Layers within the convective ML, where M is wind speedand G is geostrophic wind speed.1151Bulletin of the American Meteorological Societyenvironment comprise sector one; they are notbuoyant and will not form BLCu. Sector two in-cludes parcels that have θv greater than the envi-ronment; they are buoyant but will stop rising attheir height of neutral buoyancy before makingBLCu. Buoyant parcels in sector three are moistenough to reach their LCLs before reaching theirheight of neutral buoyancy and thus will formclouds. By tracking each parcel, one can diagnosethe range of cloud-base heights and cloud thickness.The portion of JFD within sector three gives fractionalforced and active cloudiness (Stull 1985) associatedwith rising thermals.BLX96 investigated how the JFD changes withheight and land use. To investigate land use effectswe flew over three different locations that were pre-dominantly pasture, forest, and cultivated farmfields, respectively. Fast response temperature,humidity, and altitude were measured to calculatethe JFDs. These JFDs were measured at five alti-tudes within the RxL.2. Sitea. OverviewRelatively flat topography, large areas of uniformland use, frequent fair weather, numerous airports, andexisting field instrumentation motivated the selectionof the U.S. southern Great Plains for the field site.Major cities include Wichita, Kansas, to the north;Tulsa, Oklahoma, to the east; and Oklahoma City,Oklahoma, to the south. These cities were distantenough to cause no major alteration of the weatherover the field site. The terrain elevation was roughly300 m above mean sea level (MSL) and generallysloped up toward the west–northwest with a mean gra-dient of 1/900.b. Land use under each flight trackFigure 3 shows the locations of flight tracks innorth-central Oklahoma and south-central Kansas.These east–west tracks are crosswind relative to theprevailing southerly winds. Ends of flight tracks arenamed after nearby villages, and the underlinedname is used as a generic site identifier.1) LAMONTThe Lamont flight track is over an area that waspredominantly agricultural with a few small towns.Each of the towns covers less than 3 km2. The ag-ricultural land was mainly nonirrigated wheat andpasture. A few of the farm fields were planted withsorghum, alfalfa, and hay. Land use ranged fromFIG. 3. Locations of flight tracks (dark lines) in north-centralOklahoma and south-central Kansas. White box in insert showssite location within North America. Smoothed elevation (thin gray)is contoured every 50 m.FIG. 2. Cumulus cloud triggering as a function of the ambientsounding (heavy line) and the joint frequency distribution (ovalindicates JFD envelope) of LCL vs virtual potential temperature.1152 Vol. 78, No. 6, June 199760%–80% wheat fields to 40%–20% pasture–othercrops. Roughly 40% of the cultivated fields wererecently plowed at the time of the experiment, leav-ing the reddish-brown soil bare. There were a smallnumber of trees, predominantly along fence rowsand drainage areas. Most of the trees were less than10 m tall. The terrain under the track was rather flatbut gently rising to the west, with elevations rang-ing from 320 to 425 m. The eastern edge of thisflight track was 3 km west of the ARM CART“Lamont” Central Facility site.2) WINFIELDThe Winfield flight track was over an area ofrangeland and woods. The track passed by severalsmall towns, each less than 3 km2. There was alarge quarry near the middle of the track. Tree cov-erage ranged from 50% to about 10%, with the eastend being more wooded. Pasture cov-erage ranged from 60% to 30%. Otheragricultural uses, such as hay mead-ows, sorghum, soy beans, alfalfa, andcotton fields covered 0%–30%. Mostof the trees, even in the forested areas,were less than 10 m tall. Elevations atthe west end of the track were near400 m, while those at the east were250 m. Small hills near the center ofthe flight track ranged from 70 to100 m above local ground level.3) MEEKERThe Meeker flight track had moresmall rolling hills than the other tracks.The area was a mix of forest, range-land, and pasture. This track had thedensest forest and more lakes andponds. Common trees are hackberry,cottonwood, elm, cedar, osage orange,and oak. The track passed close to sev-eral small towns, most under 3 km2.Okemah Lake (about 5 km2) was un-der the eastern end of the track. Treecoverage was 40%–60%, cropland10%–30%, and pasture 40%–50%.Trees in the area were generally under10 m tall. The western end of the flighttrack had more cultivated fields thandid the east end. Crops grown in thearea included wheat, sorghum, alfalfa,cotton, and hay. The elevation at thewest end of the Meeker track was about 280 m,while the east end was near 250 m. Hills along theflight track ranged from 40 to 60 m AGL.3. Measurements and instrumentationa. Measurements and proceduresMeasured quantities include:• vertical profiles of mean variables includinghorizontal wind components, temperature, andhumidity from the surface to above the ML top;• ML depth, zi;• near-surface, eddy-correlation fluxes of heat, mois-ture, and momentum;• upward and downward short and longwaveradiation;FIG. 4. Sketch of typical east–west flight pattern. Solid lines indicate measurementof ML scaling variables, dashed lines indicate radix-layer measurements, long-shortdashed lines were to measure boundary-layer cumulus formation factors, and thedotted lines were turns made outside the measurement domain. Letters index keypoints during the flight. Sketch (b) is a continuation from (a).1153Bulletin of the American Meteorological Society• land use imagery;• surface radiometric skin temperature statistics;• cumulus cloud coverage; and• joint frequency distributions of virtual potentialtemperature and lifting condensation level.A single flight pattern was designed to satisfyall experimental goals. Each flight of roughly 4.5 hwas flown back and forth in a crosswind directionover a single, straight, 72-km horizontal track ofrelatively uniform land use. Except for course-reversal turns outside the measurement domain, theflight was conducted within a vertical plane, withtwo exceptions as described below. Figure 4sketches the standard vertical flight pattern.During the beginning, middle, and end ofthis pattern, a vertical sounding and a near-surfaceleg (solid lines) were flown to estimate the MLdepth zi, Deardorff velocity w*, and buoyancyvelocity wB. These are ML scaling variables usedfor all experiment goals. During the two intervalsbetween each pair of scaling legs, a vertical zig-zag pattern (dashed lines) was flown in the radixlayer. The horizontal length of zig-zag needed forreliable RxL statistics was previously deter-mined using a simulated flight pattern with a syn-thetic dataset, described by Santoso and Stull(1997b, manuscript submitted to Bound.-LayerMeteor.).Also during each of the the same two intervals,horizontal legs (long and short dashes) at two alti-tudes allowed observation of surface layer plumesmerging into convective thermals, a process im-portant to formation of BLCu. These legs were par-allel to the near-surface ground track but displaceddownwind such that the convective flux footprintfor that altitude was centered on the near-surfaceground track (Weil and Horst 1992). During thesecond interval, two more horizontal legs were in-terlaced between the previous two altitudes.Fair-weather MLs typically have a three-phasegrowth (Stull 1988): 1) slow growth through thenocturnal stable boundary layer in the early morn-ing, 2) rapid rise through the residual layer in latemorning or early afternoon, and 3) deep, nearlysteady ML in the afternoon where synoptic-scalesubsidence approximately counteracts entrain-ment. With take-off at noon, our flights occurredduring the first half of phase 3, when both theML depth and the solar heating were relativelyconstant.b. Instrumentation on King AirThe King Air was configured with the standardUniversity of Wyoming turbulence instrumentationpackage. Table 1 lists the instruments and their char-acteristics.Known errors are summarized here. Preliminarydata analysis indicates an erroneous maximumcutoff of 50°C for the radiometric skin temperatureduring flights 1 through 4 (see next section for anindex of flights). The data tape from flight 2 wasunreadable in the field, but data recovery might bepossible. The LI-COR fast-response humidity instru-ment was inoperative for all of flight 3. Infrequentintermittent LI-COR outages occurred duringflights 1 and 2. The downward-looking solar radi-ometer had short cutouts during flights 7 and 8, andwas inoperative during all of flight 6. Thanks to thevigilance of the University of Wyoming programmanagers during their routine postflight data in-spection, these problems were caught early andremedied. The bulk of the data looks reasonablein our preliminary analysis.c. Collaborators with surface-based sensorsANL brought a 915-MHz boundary layer windprofiler (detection altitudes of 100 m–4.5 km), ra-dio acoustic sounding system (RASS: 100 m–2 km),and Doppler mini-sodar (10–200 m). It was locatedat 37.63°N, 96.54°W, 525-m altitude, north of theWinfield flight tracks near the village of Beaumont,Kansas. This site is the first of three remote sensingsites that will compose the permanent ANL boundarylayer measurement facility. They plan to compile con-tinuous boundary layer data 24 h every day duringmany years. Measurements at this Beaumont site be-gan on 18 July 1996.NCAR brought their ARM Multiple AntennaProfiler (MAPR), which is part of their integratedsounding system, including a surface weatherstation, Cross-Chain Loran Atmospheric SoundingSystem (CLASS) rawinsonde balloon system, andVaisala ceilometer. Their 915-MHz MAPR windprofiler with RASS was an experimental devicethat could use either spaced-antenna windfindingor Doppler beam swinging. The site was locatedat 37°12.1′N, 96°39.77′W, 405-m altitude, nearDexter, Kansas, very close to the Winfield flighttrack. Their measurements began at roughly25 July 1996.As mentioned in the introduction, the ARMCART site was conducting an intensive operation1154 Vol. 78, No. 6, June 1997period (IOP) from 15 July to 4 August 1996. In fact,this IOP was the reason for conducting our flightsat this site and during these weeks. Their observa-tions include an array of surface weather stations,Bowen ratio flux stations, rawinsonde stations withincreased launching frequency, radiometric sta-tions, and numerous remote sensors, including ra-dars, wind profilers, RASS, lidars, atmosphericemitted-radiation interferometer spectrometers,and others (Peppler et al. 1996).4. Case-study daysa. CriteriaOptimal conditions consisted of sunny after-noons, with little or no high and midlevel clouds,and neither thunderstorms nor precipitation. Windsof 0–10 m s−1 were acceptable, although light tocalm winds were most desired for the CTT goals.We hoped to fly half the flights with no fair-weather cumulus clouds, and half with. To fore-cast BLCu formation and ML growth, a simpleone-dimensional ML model based on the thermo-dynamic method (Stull 1988) was initialized withthe early morning rawinsonde sounding from theCART site.With a total of 60 research flight hours allottedto BLX96, we planned for a total of about 13flights. The goal was four flights at each location,with half of those flights during clear conditions andhalf during BLCu cloudy. Because the weather wasunusually wet during this field campaign, BLCuwere present during most of the flights.Air temperature Rosemount 102 −50° to +50°C 0.5°C 0.006°C Left wing tip, belowAir temperature Reverse flow −50° to +50°C 0.5°C 0.006°C Right wing tip, below(Minco element)Air temperature Friehe −50° to +50°C 0.5°C 0.006°C Nose, belowDewpoint temp. Cambridge Sys −50° to +50°C 1.0°C for T > 0°C 0.006°C Body near tail, rightModel 137C3 2.0°C for T < 0°CMag. heading King KPI 553 0°–360° 1° 0.02° NoseSperry C14-43Static pressure Rosemount 1501 0–108.0 kPa 0.05 kPa 0.0003 kPa Body near tail, rightand leftStatic pressure Rosemount 1201 0–103.4 kPa 0.05 kPa 0.006 kPa Body near tail, rightFA181A and leftGeometric alt. Stewart Warner 0–18 288 m 1% 0.073 m Underneath wing,APN159 Radar near body, rightAlt and leftPitot-static diff. Rosemount 1332 0–8.5 kPa 0.02 kPa 0.0005 kPa Nose boomLat/long Tremble 2000 GPS ±90° Lat 100 m (1.72 × 10−4)° Cockpit, above±180° LongLat/long Honeywell ±90° Lat 0.8 mm h−1 drift (1.72 × 10−4)° In middle cabinLaseref SM ±180° Long 1.65 mm h−1 drift (doesn’t needoutside sensor)TABLE 1. List of standard turbulence-package measurements on the University of Wyoming King Air aircraft.Parameter Instrument Range Accuracy Resolution Sensor mounted1155Bulletin of the American Meteorological Societyb. FlightsTable 2 shows the research flights during BLX96.The 11th flight on 4 August included only the firsthalf of the pattern (e.g., Fig. 4a, plus a last surface legand sounding) but was otherwise successful. The 12thflight on 9 August and the 14th flight on 14 Augustwere aborted after the first sounding and near-surfaceleg due to increasing cloud cover and are consideredunsuccessful flights. This leaves 12 successful flights.5. Weathera. Synoptic and climatic overviewTable 3 shows climatological conditions in theregion of northern Oklahoma, southern Kansas, andnortheastern Texas. The region was hot during Julyand August, with high temperatures frequently near40°C. The lee trough on the east side of the RockyMountains manifested itself as a dryline fromnorthern Texas through Kansas (Fig. 5a). Typicaleast–west diurnal oscillation of the dry line spannedthe length of the Oklahoma panhandle. The diur-nal oscillation was also apparent in the east–westpressure gradient, which strengthened daily andweakened nightly. Thunderstorms commonly oc-curred near the dryline, but most of the time theywere far enough west to remain inconsequential toour study. Cirrus and altocumulus from thunder-storm outflow were often observed to drift over thestudy area at night.With the dryline separating air masses, low-leveladvection of warm, humid air from the gulf ofMexico was unimpeded. Upper-level geopotentialfields were relatively flat this time of year, makingupper-level winds weak. Winds above 70 kPa gen-Ground velocity Honeywell 0–2106 m s−11 m s−1 after 0.0002 m s−1See aboveLaseref SM correctionVertical velocity Honeywell ±166.5 m s−11 m s−1 after 0.00016 m s−1See aboveLaseref SM correctionPitch–roll angle Honeywell ±90° pitch 0.05° (1.72 × 10−4)° See aboveLaseref SM ±180° rollTrue heading Honeyweel 180° 0.2° (1.72 × 10−4)° See aboveLaseref SMDifferential Rosemount ±1.5 kPa 0.02 kPa 0.000375 kPa Nose boompressure 858AJ/1332Pyranometer Eppley PSP 0–1400 W m−25 W m−20.08 W m−2Middle body, top(0.285–2.8µm) and bottomPyrgeometer Eppley PIR 0–700 W m−215 W m−20.04 W m−2Middle body, top(3.5–50µm) and bottomRadiation Heimann KT-19.85 −50–400°C 0.5°C 0.10° for 10 s Near tail, bottomPyrometer (9.6–11.5 µm)H2O LI-COR LI 6262 0–7.5 kPa 1% 0.01 kPa Pipe at top,concentration middle bodySurface and video camera — — — Cockpit (forward),cloud visual below left engine(downward)TABLE 1. Continued.Parameter Instrument Range Accuracy Resolution Sensor mounted1156 Vol. 78, No. 6, June 1997Lamont 21 July X 3 4.523 July X 5 4.427 July X 7 4.4  4 Aug X 11* 3.113 Aug X 13 4.6Winfield 15 July X 1 4.422 July X 4 4.525 July X 6 4.531 July X 9 4.5  9 Aug X 12** 0.8Meeker 16 July X 2 4.728 July X 8 4.8  2 Aug X 10 4.914 Aug X 14** 2.1*First half of flight pattern only.**Flight aborted early due to bad weather.TABLE 2. Record of flights during BLX96.Date FlightLocation 1996 BLCu Clear index HoursAverage temperature (°C) 26.0–28.4 25.1–27.6Fraction of possible sunshine (%) 73–79 72–79Number of clear sky days month−112.0–14.4 13.0–14.9Mean wind speeds (m s−1) 4.0–5.8 3.9–5.6Prevailing wind direction South SouthPrecipitation (cm) 6.9–9.2 6.1–7.6Tornadoes 1–2 1–2TABLE 3. Climatological conditions for the BLX96 region.Condition July Augusterally flowed from the west or northwest, and thejet stream remained at higher latitudes.b. Synoptic evolutionApproximately biweekly, a frontal system dis-rupted the lee trough and the oscillation of the dryline. By the time the front reached south of the fieldsite, much of its energy was spent, causing it tomove slowly through the study area (Fig. 5b). Suchwas the case during the first week, when the frontdisplayed a north–south diurnal variation, movingfarther south during night hours. A weak tempera-ture gradient coupled with a significant dewpointgradient characterized this front, with dewpointsaround 23°C south of the front and 19°C north.At the end of the third week and into the fourth,a front characterized by a stronger temperature gradi-ent and weaker dewpoint gradient was present nearthe site that displayed similar behavior. High tem-peratures were near 38°C south and 29°C north ofthe front. Although wind shifts were apparent nearthe fronts, the low-level winds were generally fromthe south, both south of the front and farther than100 km north of the front. This occurred becausethe fronts were weak.As the fronts slowly moved south, dewpointtemperatures in the study area normally droppedfor 1 or 2 days. Subsidence contributed to a de-crease in the depth of the boundary layer and lapserates in the early morning lower troposphere wereclose to moist adiabatic. The stable, drier air andthe thinner boundary layer combined to producemostly clear conditions and virtually no chance ofthunderstorms. Though slightly drier, occasionallythe post-frontal air was cool enough to form radia-tion fog over night. Whenever the front was com-pletely out of the region, the flow at lower levelsstrengthened from the south as the lee trough anddryline reestablished. This brought lapse ratescloser to dry adiabatic. The moisture returned andthe boundary layer was often deeper, increasing theoccurrence of observed cumulus and the chancefor thunderstorms.During the last 2 weeks of the experiment, adeep longwave trough over the eastern half ofNorth America dominated the weather. As short-waves propagated rapidly along the trough, thun-derstorm conditions were optimal. Consequently,the weather was unseasonably poor, as a series ofsurface fronts passed through the study area withvariable speeds and strengths. They displayed simi-1157Bulletin of the American Meteorological SocietyFIG. 5. Sketch of typical synoptic situation, showing idealizedisobars (solid lines), fronts and drylines (thick lines), pressurecenters, and experiment domain (shaded). (a) Climatologicallyexpected and (b) typically observed during July 1996.lar characteristics as those discussed, includingstagnation periods near the study area.Conditions were favorable for flights on thosedays when a frontal system was not directly overthe experiment site. To choose the particular flightlocation each day, we considered the chance of thun-derstorms, morning cloudiness from thunderstormoutflow, fog, and the development of BLCu. Weneeded morning sunshine to promote boundary-layer development as early as possible in the dayand no precipitation or thick mid- and high-levelclouds. Since we did not fly close to passing fronts,winds were usually from the south.Table 4 summarizes the weather conditions forthose days that the aircraft flew.6. Discussion and conclusionsA total of 53.3 of the 60 allotted research flighthours were flown during the 12 successful flights.Two other flights (numbers 12 and 14) wereaborted, consuming a total of 2.9 flight hours. Thelast flight (number 14, not listed in some of theprevious tables) was unsuccessful, because dete-riorating weather prompted us to change locationsand then terminate the flight early.Because of a stalled front that vacillated throughthe field site, the ML was generally shallower, andthe ground moister and cooler, than would have beenclimatologically expected. The infrequency of anticy-clones over the site also reduced the number of flightdays experiencing light winds and clear skies.These conditions affect all three goals. The CTTanalysis will be refocused on combined thermal andmechanical turbulence generation. The RxL analy-sis will be extended from pure free convection tocases of combined (free plus forced) convection.The BLCu study will add cloud shading–inducedheterogeneity to factors that affect the JFD.Detailed data analysis is just beginning.Available now is a technical report (Berg et al.1997) containing the airborne scientist flight logs.A table of key ML scales, mean meteorologicalvalues, and fluxes will be produced in about a year.The raw data is available to outside investigators now,and the calibrated and documented digital data willbe made available as soon as it is produced.Acknowledgments. BLX96 was sponsored by the National Sci-ence Foundation (NSF) under Grant ATM-9411467. We appreci-ate the willingness of Greg Tripoli at the University of Wisconsinto serve as U.S. liaison and co-primary investigator for this project.The experiment would have been impossible without outstand-ing support from the University of Wyoming (UW) King Air facil-ity, also sponsored by the NSF. We thank pilots Ernie Gasawayand Mark Hoshor, program managers Glenn Gordon and LarryOolman, and the aircraft support crew Larry Irving and Ken Endsley.Preliminary research, funded by the Department of Energy(DOE) Grant DE-FG02-92ER61361, led to the definition of goalsfor the experiment. The continued close cooperation with the ARMProgram, also sponsored by the DOE, was greatly appreciated. Wegive special thanks to ARM researchers and site managers, includ-ing Richard Cederwall, Doug Sisterson, and Jim Teske.Richard Coulter, Tim Martin, Donna Holdridge, and JerryKlazura of Argonne National Laboratory contributed data from thewind profiler–RASS–sodar site near Beaumont, Kansas. Coulteralso coordinated ARM CART 915-MHz wind profiler–RASS mea-surements near Lamont, Oklahoma. Steve Cohn and Dave Parsonsof the National Center for Atmospheric Research, who collaboratedwith a 915-MHz multiple antenna wind profiler near Dexter, Kan-sas, were funded by DOE via ARM Grant DE-AI05-90ER61070.Lawrence Fleck of the Oklahoma State University ExtensionCenter helped identify vegetation and crops from the aerial photo-graphs. We also thank Rich Clark of Millersville University, Penn-sylvania, who shared the aircraft to conduct his experiments on the1158 Vol. 78, No. 6, June 1997nocturnal jet. Our cordial interaction with him made administrationof the experiment a pleasure.ReferencesBerg, L. K., R. B. Stull, E. Santoso, and J. P. Hacker, 1997: Bound-ary Layer Experiment–1996 (BLX96) airborne scientist flight log.Boundary Layer Research Team Tech. Note BLRT-97-1, 116 pp.[Available from Roland B. Stull, Atmospheric ScienceProgramme, Dept. of Geography, UBC, 1984 West Mall,Vancouver, BC V6T 1Z2, Canada.]Holtslag, A. A. M., and F. T. M. Nieuwstadt, 1986: Scaling the at-mospheric boundary layer. Bound.-Layer Meteor., 36, 201–209.Panofsky, H. A., 1978: Matching in the convective planetary bound-ary layer. J. Atmos. Sci., 35, 272–276.Peppler, R. A., P. J. Lamb, and D. L. Sisterson, 1996: Site scien-tific mission plan for the southern Great Plains CART site, Janu-ary–June 1996. Tech. Memo. ARM-96-001, 86 pp. [Available15 July 2.5–5 m s−1, 180°–210° 25% BLCu Frontal passage 13 July, little pressure gradient16 July 5 m s−1, 160°–180° 25% BLCu High to SE blocking approaching low21 July Variable southerly Clear High to SE prevented widespread CB22 July 2.5 m s−1, 180°–200° Clear MCC in Nebraska with approaching front23 July Variable southerly 25% BLCu CB in north-central OK previous night with frontal passage25 July Variable northerly 25% BLCu CB rained heavily on 24 July, small low-level ridge in regionallowed fair wx27 July Variable northerly 15% BLCu CB rained heavily on 26 July, small low-level ridge in regionallowed fair wx28 July Variable southwesterly 20% BLCu Midlevel ridge gave fair wx over much of the region31 July Variable southeasterly 5% BLCu; Thunderstorms along approaching front kept to10%–20% Ci the north during the day  2 Aug Variable southeasterly Clear Front to the south washed out. Mid- and upper-level ridgingkept disturbances to the north  4 Aug 7.5–10 m s−1, Clear Strong low-level pressure gradient developed160°–180° strong winds but kept it clear  9 Aug Variable easterly Patchy Cc Flight aborted due to cloud cover13 Aug Variable Cloudless Ridging at all levels, subsidence kept skies clear;and hazy very moist BL contributed to haze14 Aug Variable southerly > 25% BLCu Flight aborted due to cloud coverTABLE 4. Daily weather (wx) conditions during flight days.Date Wind (Speed, dir.) Clouds Commentsfrom Office of Scientific and Technical Information, P.O. Box62, Oak Ridge, TN 37831.]Santoso, E., 1993: A wind-profile relationship for the unstable sur-face-layer mixed-layer system. M. S. thesis, Dept. of Atmosphericand Oceanographic Sciences, University of Wisconsin, 59 pp.Schrieber, K., R. Stull, and Q. Zhang, 1996: Distributions of sur-face-layer buoyancy versus lifting condensation level over a het-erogeneous land surface. J. Atmos. Sci., 53, 1086–1107.Stull, R. B., 1985: A fairweather cumulus cloud classification schemefor mixed-layer studies. J. Climate Appl. Meteor., 24, 49–56.——, 1988: An Introduction to Boundary Layer Meteorology.Kluwer, 666 pp.——, 1994: A convective transport theory for surface fluxes. J. Atmos.Sci., 51, 3–22.Weil, J. C., and T. W. Horst, 1992: Footprint estimates for atmo-spheric flux measurements in the convective boundary layer.Precipitation Scavenging and Atmosphere–Surface Exchange,Vol. 2, S. E. Schwartz and W. G. N. Slinn, Eds., HemisphericPublishing, 717–728.


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics

Country Views Downloads
United States 13 28
China 5 5
Russia 4 0
Germany 1 10
City Views Downloads
Ashburn 13 0
Shenzhen 5 5
Saint Petersburg 4 0
Unknown 1 10

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}
Download Stats



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items