Doppler Lidar Estimation of Mixing Height Using Turbulence, Shear, and Aerosol Profiles

Sara C. Tucker Cooperative Institute for Research in Environmental Sciences, University of Colorado, and NOAA/Earth System Research Laboratory, Boulder, Colorado

Search for other papers by Sara C. Tucker in
Current site
Google Scholar
PubMed
Close
,
Christoph J. Senff Cooperative Institute for Research in Environmental Sciences, University of Colorado, and NOAA/Earth System Research Laboratory, Boulder, Colorado

Search for other papers by Christoph J. Senff in
Current site
Google Scholar
PubMed
Close
,
Ann M. Weickmann Cooperative Institute for Research in Environmental Sciences, University of Colorado, and NOAA/Earth System Research Laboratory, Boulder, Colorado

Search for other papers by Ann M. Weickmann in
Current site
Google Scholar
PubMed
Close
,
W. Alan Brewer NOAA/Earth System Research Laboratory, Boulder, Colorado

Search for other papers by W. Alan Brewer in
Current site
Google Scholar
PubMed
Close
,
Robert M. Banta NOAA/Earth System Research Laboratory, Boulder, Colorado

Search for other papers by Robert M. Banta in
Current site
Google Scholar
PubMed
Close
,
Scott P. Sandberg NOAA/Earth System Research Laboratory, Boulder, Colorado

Search for other papers by Scott P. Sandberg in
Current site
Google Scholar
PubMed
Close
,
Daniel C. Law NOAA/Earth System Research Laboratory, Boulder, Colorado

Search for other papers by Daniel C. Law in
Current site
Google Scholar
PubMed
Close
, and
R. Michael Hardesty NOAA/Earth System Research Laboratory, Boulder, Colorado

Search for other papers by R. Michael Hardesty in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

The concept of boundary layer mixing height for meteorology and air quality applications using lidar data is reviewed, and new algorithms for estimation of mixing heights from various types of lower-tropospheric coherent Doppler lidar measurements are presented. Velocity variance profiles derived from Doppler lidar data demonstrate direct application to mixing height estimation, while other types of lidar profiles demonstrate relationships to the variance profiles and thus may also be used in the mixing height estimate. The algorithms are applied to ship-based, high-resolution Doppler lidar (HRDL) velocity and backscattered-signal measurements acquired on the R/V Ronald H. Brown during Texas Air Quality Study (TexAQS) 2006 to demonstrate the method and to produce mixing height estimates for that experiment. These combinations of Doppler lidar–derived velocity measurements have not previously been applied to analysis of boundary layer mixing height—over the water or elsewhere. A comparison of the results to those derived from ship-launched, balloon-radiosonde potential temperature and relative humidity profiles is presented.

Corresponding author address: Sara Tucker, NOAA/ESRL/CSD, 325 Broadway, Boulder, CO 80305. Email: sara.tucker@noaa.gov

Abstract

The concept of boundary layer mixing height for meteorology and air quality applications using lidar data is reviewed, and new algorithms for estimation of mixing heights from various types of lower-tropospheric coherent Doppler lidar measurements are presented. Velocity variance profiles derived from Doppler lidar data demonstrate direct application to mixing height estimation, while other types of lidar profiles demonstrate relationships to the variance profiles and thus may also be used in the mixing height estimate. The algorithms are applied to ship-based, high-resolution Doppler lidar (HRDL) velocity and backscattered-signal measurements acquired on the R/V Ronald H. Brown during Texas Air Quality Study (TexAQS) 2006 to demonstrate the method and to produce mixing height estimates for that experiment. These combinations of Doppler lidar–derived velocity measurements have not previously been applied to analysis of boundary layer mixing height—over the water or elsewhere. A comparison of the results to those derived from ship-launched, balloon-radiosonde potential temperature and relative humidity profiles is presented.

Corresponding author address: Sara Tucker, NOAA/ESRL/CSD, 325 Broadway, Boulder, CO 80305. Email: sara.tucker@noaa.gov

1. Importance of mixing height

In studies of air quality, information about the depth and dynamics of the atmospheric boundary layer (BL) is essential to interpreting in situ measurements of atmospheric species. Likewise, dynamics information is necessary for improvement of weather and air quality forecasting and modeling. (White et al. 1999) To understand the processes that affect concentrations of species emitted within the surface layer, one needs knowledge of transport and mixing conditions including mean horizontal wind speed and direction profiles, strength of turbulence, and depth of the atmospheric BL. The BL is defined here as the layer of atmosphere in turbulent connection with the surface of the earth and the height of the BL, referred to in this article as the mixing height (MH), defines the volume of atmosphere in which gas-phase or aerosol chemical species, emitted within the BL, are mixed and dispersed. Based on surface-level in situ measurements of aerosol properties and size distributions, knowledge about the height to which particles may be mixed can also improve assumptions about aerosol properties aloft for the purpose of aerosol–cloud interaction studies. The combination of MH, updrafts, wind speed and direction, and other meteorological information is crucial to understanding of in situ atmospheric chemistry measurements made during air quality studies.

This article describes a new method for estimating MH that relies primarily on profiles of velocity variance, but also incorporates high-resolution mean horizontal wind speed and direction profiles, and aerosol backscattered lidar intensity data, Ib. Application of the method is demonstrated on profiles derived from wind velocity and Ib data collected using the National Oceanic and Atmospheric Administration (NOAA) coherent High-Resolution Doppler lidar (HRDL; Grund et al. 2001) in August and September 2006 as part of the second Texas Air Quality Study (TexAQS 2006). This study was carried out by the Texas Commission on Environmental Quality (TCEQ), NOAA, National Aeronautical and Space Administration (NASA), and various universities during summer 2006 to study air quality conditions throughout Texas and to improve understanding of the effects of meteorology, topography, aerosol and gas-phase chemistry on air quality and climate change. The study included measurements from land, aircraft, and ship-based platforms. The ship-based intensive operation period of the campaign took place late July through mid-September 2006. During this period, scientists on board the NOAA Research Vessel Ronald H. Brown (R/V Brown) deployed a complement of remote sensing [Doppler and ozone Differential Absorption Lidar (DIAL) lidars, radar wind profilers, and C-band radar] and in situ (aerosol and gas phase chemistry and flux tower) instruments to study air quality, pollution, and dynamics conditions over water in the Houston/Galveston and Gulf Coast regions. Measurements were concentrated in the Gulf of Mexico, Galveston Bay, and Houston ship channel areas, with the ship moving to different locations to sample and study several pollution events (Bates et al. 2008).

During the campaign, the NOAA Earth System Research Laboratory (ESRL), Chemical Sciences Division operated HRDL 24 h a day, performing continuous, range-resolved measurements of boundary layer winds, and lidar intensity Ib from the aft deck of the R/V Brown. Line of sight (LOS) wind observations were processed into mean horizontal wind profiles and velocity variance συ2 profiles, both posted in near–real time during the campaign to provide context, transport, and turbulence information for the Ib profiles and in situ aerosol and gas-phase chemistry measurements. Because the ship-based instruments were taking measurements both close to land and out over open waters, 24 h a day, it was determined that classical methods to determine MH, such as Ib profiles or balloon-radiosonde launches, were not sufficient to estimate MH under all the conditions encountered because they did not always reflect turbulent mixing or they were too infrequent, as was the case with the radiosondes. This study demonstrates the use of coherent Doppler lidar data to estimate MH once every 15 min for all locations and boundary layer conditions encountered during the ship-based portion of TexAQS 2006.

2. Atmospheric boundary layer and mixing height definition

The boundary layer is the region of the atmosphere that interacts directly with the surface of the earth via turbulent mixing processes. The notion of interaction time scale has been introduced, as reflected in the definition given by Stull (1988): “that part of the troposphere that is directly influenced by the presence of the Earth’s surface, and responds to surface forcings with a timescale of about an hour or less.” In the literature, definitions of the BL depth vary according to application and/or instrumentation. The various definitions may converge, but in spatially complex terrain (e.g., coastal zones), during transition times, or in stable conditions (such as at night), the definitions may yield different results. Variations in BL definitions have also arisen because of different methods of measurement; for example, sodar measurement of the acoustic refractive index structure parameter Cn2(λsodar), radar wind profiler measurement of radar refractive index Cn2(λradar), radiosonde measurement of potential temperature (θ) or relative humidity (RH), lidar measurement of aerosol backscattered intensity Ib, and flux tower measurement of turbulence, etc., may all produce different estimates of the BL depth (Seibert et al. 2000).

Properties of the surface over which the BL forms can affect the stability and thus the measurements made in this layer. Well-mixed BLs often occur over/near land in the unstable daytime convective boundary layer (CBL), typically as a result of surface heating. A characteristic of the well-mixed BL is that the θ profile is approximately constant with height, although other constituents may achieve similar profiles. Stable boundary layer (SBL) conditions may be observed over land, typically at night where, in the absence of surface heating, the BL is in general not well mixed, and profiles of θ and other quantities are not constant with height (Mahrt 1999; Banta et al. 2002, 2003, 2006; Frehlich et al. 2006). SBL conditions are also observed over cold oceans (Smedman et al. 1997; Skyllingstad et al. 2005). Very stable boundary layers (vSBL), typically observed over land, exhibit weak shear turbulence and strong temperature gradients near the surface (Mahrt and Vickers 2006; Banta et al. 2007). Weakly unstable conditions are likely to occur over the Gulf of Mexico since surface heating and cooling cycles do not generate obvious diurnal patterns in convection (Hanna et al. 2006).

Defining the BL as the layer connected to the surface via turbulent mixing, and MH as the height of that layer allows profiles of turbulence quantities, such as velocity variance or turbulence kinetic energy (TKE), to be used to determine BL depth in all stability cases ranging from SBL conditions to strong CBL conditions. This layer, defined by turbulence, has previously been referred to as the mixing layer (Stull 1988; Beyrich 1997; Seibert et al. 2000) and sometimes, in the case of a CBL, as the “well-mixed” layer (Seibert et al. 2000). Consistent with the definition of BL, Seibert et al. (2000) provide a general definition of the MH term as “…the height of the layer adjacent to the ground over which pollutants or any constituents emitted within this layer or entrained into it become vertically dispersed by convection or mechanical turbulence within a time scale of about an hour.” This work demonstrates how Doppler lidar measurements of winds and turbulent parameters in convective or stable BLs, enable straightforward estimation of MH.

3. A history of boundary layer and mixing height estimation using lidar data

Myriad papers have discussed the use of lidar data for estimation of BL depth, MH, and entrainment zones. Most lidar-derived MH estimates involve using Sb profiles (range-function-corrected versions of Ib profiles) acquired from ground (Boers et al. 1984), aircraft (Melfi et al. 1985; Flamant et al. 1997; Menut et al. 1999), or ship-based platforms (Hennemuth and Lammert 2005). Typical analysis of Sb profiles assumes that the boundary layer has higher aerosol concentrations than the free troposphere above. Detection of the height of the top of this aerosol layer, referred to here as hA, involves searching for the height of the first strong negative gradient in aerosol backscatter. Turbulent mixing tends to produce uniformity in aerosol/species concentration so if turbulent mixing extends to a certain height, then aerosol concentration and thus Sb profiles will not likely exhibit strong negative gradients below that height. Three common methods for estimation of hA include thresholding Sb to determine hA for each profile (Melfi et al. 1985), finding the steepest negative gradient in the logarithmic derivative of the Sb profiles (Senff et al. 1996; White et al. 1999; Steyn et al. 1999; Haegeli et al. 2000; Hennemuth and Lammert 2005; Lammert and Bösenberg 2006), and wavelet analysis of the negative gradients in Sb as first mentioned by Davis et al. (1997) and later described by Davis et al. (2000). The wavelet method has since been developed and used in various forms by Cohn and Angevine (2000) and Brooks (2003).

In most cases, Sb profiles are generated using what is referred to as “direct detection” lidars that use photo-multiplier tubes or avalanche-photo diodes to measure the amount of light backscattered from a volume of atmosphere. This study uses data from a coherent Doppler lidar that uses heterodyne detection to measure wind-induced Doppler shifts in the frequency of laser pulses backscattered off aerosol particles in a volume of atmosphere, and thus estimate LOS velocity of the wind carrying those particles. The strength of the backscattered and heterodyne-detected signal from zenith stare periods may be processed to estimate Sb profiles. When not calibrated, as was the case for this HRDL dataset, Sb profiles from zenith stare periods can still be used to detect aerosol gradients. These profiles will be henceforth referred to as relative aerosol backscatter profiles, denoted by SRAB, since the profiles provide the relative intensity information needed to detect gradients, but do not represent absolute measurements of aerosol backscatter.

Data from Doppler lidars have been used in boundary layer depth estimation, but still with respect to Sb. Lothon et al. (2006) used the total variance σ2 on velocity measurements acquired using the HRDL in a ground-based, zenith staring configuration to estimate the MH in the unstable daytime CBL by searching for the lowest altitude where the σ2 exceeded a large threshold value. The total variance on velocity measurements includes both atmospheric variations and instrument noise, and so this value was set to correspond to a level where the velocity variance estimates are dominated by low-signal-level instrument effects. Although this method uses velocity data, the total variance of velocity estimates is strongly dependent on Sb according to the theoretical lower bound on velocity estimates (the Cramér–Rao Lower Bound; Cramér 1946; Rye and Hardesty 1993a,b), and thus the method effectively corresponds to locating an aerosol backscatter intensity threshold. Doppler lidar velocity data have also been used for numerous boundary layer structure studies including an investigation by Banta et al. (2006) in which the authors used wind data to estimate heights during low-level jets to characterize the stable nocturnal boundary layer. Likewise, Frehlich, et al. (2006) advocated the use of Doppler lidar data to estimate mixing heights in a study of nocturnal boundary layers. The data and methods presented here demonstrate utility under all stability conditions and during all times of day throughout a six-week period.

4. Ship-based Doppler lidar

Knowledge of the properties, qualities, uncertainties, and limitations of an instrument such as HRDL is important for understanding the quality of its measurements and derived products. HRDL is a 2.022-μm (invisible) wavelength, eyesafe, coherent Doppler lidar used to measure boundary layer winds from land, ship, and aircraft platforms. The system runs at a 200-Hz pulse repetition frequency (PRF) with a pulse-to-pulse frequency stability of approximately 1 MHz (Wulfmeyer et al. 2000). With 100-pulse averaging, HRDL produces a range-resolved beam of velocity estimates once every half-second with velocity estimate precision of 0.15 to 0.30 m s−1, depending on signal strength (Rye and Hardesty 1993a,b).

HRDL laser pulses are 200 ns long, resulting in approximately 30-m range resolution for independent velocity measurements. This resolution imposes a 30-m resolution limit on the MH estimates derived from zenith-stare data, which works well for most applications. Scanning, however, allows profiles of mean horizontal wind with finer vertical resolution. The scan sequence and processing parameters used during the TexAQS 2006 study enabled a vertical resolution of 5 m near the surface, increasing to about 30 m at 3 km altitude. Because scattering of the outgoing laser pulse off lidar-system optics renders very near-field signal returns unusable, the system’s minimum range is restricted to 180 m. For zenith staring, this minimum range implies a minimum altitude for the profiles, but scanning the beam provides low-level data for the horizontal wind profiles, reducing the minimum altitude for those profiles to a few meters off the surface.

Although backscatter for the 2.022-μm wavelength relies on Mie scattering with a preference to size distributions with modes that correspond to a 2-μm backscatter kernel (Feingold and Grund 1994), HRDL has operated with adequate signal even when smaller particles (i.e., less than 0.2-μm diameter) dominated the aerosol size distributions and total number concentrations were less than 100 cm−3. Because the low-energy 2-μm wavelength of HRDL’s laser pulses does not penetrate optical depths greater than about two (typical of thick clouds), HRDL cannot detect vertical velocities or aerosol layers very far into or above such clouds. The presence of clouds is a known difficulty for lidar-based MH estimation because, although they may form at the top of an updraft, turbulence may extend through the cloud, potentially dispersing surface-emitted pollutants to altitudes higher than cloud base. Note, however, that one may observe vertical velocity variance σw2 and aerosol layers above cloud height during cloud breaks. The ship’s location over water meant that fewer cloud cover conditions were encountered than may have been encountered at a land-based site because less surface heating meant less convective turbulence available to lift air to condensation heights. The MH estimation was performed at times when the cloud fraction was less then 50% of the observation period or when mixing heights were clearly below cloud base. For the TexAQS 2006 data, the lidar-estimated MH during cloud-topped mixing layers was typically close to the cloud base height (i.e., within 60 m) and close to the MH derived using θ and RH profiles.

a. Ship-based data collection

NOAA has used the HRDL system in various land, aircraft, and ship-based experiments to study boundary layer transport, dynamics, and small-scale meteorology (e.g., Banta et al. 2002, 2003, 2006; Lothon et al. 2006; Wolfe et al. 2007; Fairall et al. 2006; Brewer et al. 2005; Kiemle et al. 2007; Pichugina et al. 2008). During the ship-based studies, NOAA’s unique motion-stabilized, low-elevation-angle scans have enabled HRDL to provide high vertical and temporal resolution measurements of mean horizontal wind speed and direction. Because HRDL uses a diffraction limited “spotlight” beam that suffers no sidelobes, it can scan down to the surface to generate wind profiles at heights where radar profilers typically fail because of sea clutter (Wolfe et al. 2007). Starting with TexAQS 2006, improvements to HRDL’s stabilization and scanner control software allowed automatic incorporation of motion-compensated zenith stare periods into periodic HRDL scan sequences. Such zenith stares provided measurement of the vertical wind field from 180 m altitude up to the top of the aerosol layer (typically 2–3 km) with 30-m vertical and 0.5-s temporal resolution.

Scanning Doppler lidar operation from a moving platform is accomplished using a GPS-based motion compensation system that enables active pointing stabilization and removal of the ship’s mean velocity, in x, y, and z directions, from velocity measurements along the lidar’s line-of-sight. The system provides pointing stability with a 0.1° precision in a static situation (i.e., on land) and with better than 0.5° precision even under heavy seas. A demonstration of the efficacy of this system under heavy seas was presented in Hill et al. (2008).

b. HRDL measurements during TexAQS 2006

HRDL operated 24-h a day during the TexAQS 2006 ship-based operational periods, taking over 955 h of data during the six-week study. The system was typically programmed to perform 15-min scan sequences consisting of constant elevation, scanning azimuth scans (conical scans) at 1°, 7°, and 45° elevation angles, followed by a 5–8 min zenith stare period, and then by vertical slice scans of fixed azimuth, varying elevation (elevation scans). See Fig. 1 for a display of the typical scan patterns. All scans were set to operate in an earth-fixed (north–east–down) coordinate system, compensating for midscan changes in the ship’s orientation by adjusting scanner azimuth, elevation, and/or tilt angle to maintain the appropriate orientation in the world frame.

The data from conical scans were used to automatically generate BL mean horizontal wind speed and direction profiles, but they also provide information about horizontal and vertical turbulence intensities and other surface dynamics, such as thunderstorm outflows, ship/tower wakes, etc. The elevation scans provide supplemental information about mixing, layering, and wind shear. The 2-Hz aerosol Sb and velocity data from zenith stare files were used to derive average zenith SRAB and σw2 profiles. Other instruments on board the R/V Brown during TexAQS 2006 that provide data usable to estimate MH include balloon radiosondes, a 915-MHz radar profiler, a flux-tower package, and an aerosol and ozone profiling lidar; however, the Doppler lidar was the only instrument to provide continuous profiles of lower-tropospheric dynamics throughout the entire experiment.

5. Estimation of MH using HRDL data

The following paragraphs demonstrate the use of the 2-μm Doppler lidar-generated profiles for estimating heights of various types of lower-tropospheric layers. These heights include the turbulence height derived from velocity variance profiles, hσ, horizontal wind shear heights, hdir, hspd, and hvec, and the height of the aerosol layer(s) hA derived from SRAB profiles. With the MH defined in terms of turbulence, using the σw2 profiles is preferred for estimation of MH, when available, but all of these heights may be considered candidates for the MH estimate. When the MH is below the minimum zenith range, or when other scanning priorities do not allow time for zenith profiling, one may make use of other types of profile, derived from the Doppler lidar data, that reflect either the effects of turbulence (as in the case of SRAB profiles) or are related to turbulence, as is the case for wind shear. The available profiles and their corresponding layer height names are listed in Table 1 and will be further described in the following subsections. For each type of profile, example measurements and the method for estimating the associated layer height are presented.

The last part of this section describes the procedures for selection of the appropriate MH from among the various candidate heights to produce MH estimates for each 24-h period of the experiment. At the end of section 5, results obtained using the various methods for estimation of MH will be compared to MH estimates found using the gradient method for radiosonde-derived θ and RH profiles (Hennemuth and Lammert 2005).

a. Vertical velocity variance σw2 profiles

The maximum height of surface-based turbulence hσ can be estimated using vertical velocity w data to calculate σw2 turbulence profiles, assuming that the surface-based turbulence extends above the 180-m minimum range of the zenith staring lidar. Once every 15 min during TexAQS 2006, w profiles were acquired at a 2-Hz rate during 5-to-11-min-long, motion-stabilized, zenith stare portions of the scan sequence (cf. Fig. 1). Even at low relative (to the ship) horizontal mean wind speeds of 1–3 m s−1, a 5-min stare corresponds to a 300-to-900-m fetch, providing sufficient sampling of turbulent structure for estimating the local (in time and space) mixing height even if the sampling interval is not long enough for quantitative estimates of TKE. Large convective eddies often present in the CBL may modulate the MH over scales longer than the zenith observations indicating that hσ may be a local minimum or a maximum in MH. An example of w profiles, from a zenith stare period on 27 August 2006, is shown in the top left of Fig. 2a. Wind speed profiles around this time indicate a fetch of 2–3.5 km (altitude dependent) for the approximately 6-min sampling period.

Estimation of σw2 profiles starts by creating a velocity time series from the data in each 30 m range gate in the zenith stare period, removing the mean (in order to reduce any bias introduced by possible combinations of scanner pointing errors, strong horizontal mean winds, and uncompensated ship motion), and calculating the variance σ2. Note that the data are not detrended. This total variance includes both correlated and uncorrelated components. Although it is understood that the uncorrelated component of σ2 may be both instrument and atmospheric in origin, most of the values are close to the Cramér–Rao lower bound on the precision of the velocity estimates for the corresponding instrument parameters and signal strength levels (Rye and Hardesty 1993a). Thus, for simplicity, the uncorrelated component is referred to here as “instrument variance” σinst2. Next, σinst2 is estimated (Lenschow et al. 2000) and the resulting difference between σ2 and σinst2 gives the estimate of σw2. Figure 2b (top, right) illustrates the results of this method with profile plots of the total velocity variance σ2 (red line) for the example data from the left panel, the uncorrelated portion of this profile σinst2, (green) and the difference between these, which represents the atmospheric σw2 profile (thick blue line).

To demonstrate diurnal variation in the σw2 profiles observed when the ship was close to land, Fig. 2c (bottom) contains a time–height image of σw2 profiles, indicated using color, for a five-day period during the first half of the TexAQS 2006 experiment. In these images, colors from white to red indicate stronger σw2 (i.e., above 0.03 m2 s−2) corresponding to stronger turbulent mixing. The change to cooler colors with height is typically indicative of the top of the mixing layer.

Experimenting with several different methods for using the σw2 profile information to estimate hσ demonstrated that a method based on thresholding worked best. This method defines significant turbulence as any σw2 value greater than a given threshold. By comparing MH estimates from σw2 profiles to those derived from radiosonde RH and θ data, the optimum threshold for this dataset was empirically determined to be 0.04 m2 s−2 (0.20 m s−1 standard deviation) for most conditions except over the open ocean, where 0.03 m2 s−2 (0.17 m s−1) was more appropriate. If the σw2 values in the lowest few range gates of a profile, starting at 180 m, qualify as turbulent (i.e., they are greater than the threshold), then the height at which the σw2 values drop below this threshold is assigned to hσ. Most of the variations in vertical velocity above this height are on the order of ±0.05 m s−1 or less in the TexAQS 2006 data unless a shear layer is present at a higher altitude.

One possible reason for the lower, empirically determined, threshold value over the open ocean is that weak mixing over the water may demonstrate local minima in the σw2 profile, triggering the threshold at too low an altitude. However, applying the different thresholds produced no change in the MH estimate about 50% of the time and 80% of the time it produced a change in the estimates of less than 20%. The differences between the land and marine boundary layer structures, as observed with these σw2 profiles and resulting MH, are the focus of a separate study.

b. Composite velocity variance profiles συ2

A limitation of using zenith profile data is the lack of data below the 180-m minimum range. It is important to test whether σw2 values above threshold in the lowest range gates of a zenith profile are connected to the surface via turbulent processes in order to verify that the zenith estimate of hσ is the surface-based MH. In addition, near shore at night when the land-influenced boundary layer tends to become stable to very stable and may be ≤200 m deep, estimation of hσ becomes impossible with zenith data. To address these issues, data from HRDL elevation scans, typically repeated in the scan pattern in orthogonal azimuth planes, were used to provide velocity and turbulence data in the near-horizontal dimension, extending the availability of turbulence information down to the surface. Each of these scans provides a 2D visualization of the wind field in the vertical dimension along a particular azimuth angle. Mean horizontal wind profiles derived from conical azimuth scan data are used to estimate and remove the mean wind field from each velocity estimate in the elevation scan. This leaves a residual wind field used to estimate horizontal velocity variance, σH2, free from biases induced by vertical wind shear. Figure 3a (top, left) contains an example of a residual wind field with velocity variations of ±1 m s−1 from the surface up to about 700 m.

To estimate σH2 profiles, the horizontal component of the LOS residual winds is first estimated by dividing by the cosine of the elevation angle for each point in an elevation scan. The data are then divided into altitude bins of 30-m depth (to match the zenith resolution) with the data in each bin used to estimate velocity variance in a method similar to that discussed in section 5. This technique is applied to several scans performed along the same azimuth direction during a 2–3-min period and the results are averaged to form a single profile, σx2. The process is repeated for the remaining scans, performed at an azimuth angle orthogonal to the first azimuth angle, to obtain σy2. The x and y directions are not necessarily parallel and perpendicular to the mean surface wind, but are defined according to the ship’s axis at the beginning of the scan period. The results are averaged to form the final profile σH2 = 0.5(σx2 + σy2). Although horizontal motion is likely to be the largest component of the LOS velocity measurements used from the vertical-slice scans, these calculations are not attempts to estimate horizontal TKE, but rather to look for turbulence gradients in the lowest 200–300 m, where data from the σw2 profiles are unavailable or unreliable.

Composite profiles (συ2) are produced using σH2 data from the surface to 300 m and then using σw2 data from 330 m up. Figure 3b (top, right) contains an example of the two kinds of profiles acquired during convective conditions when turbulence is more isotropic. In such a situation, the overlapping portions of the profiles match very well. During stable conditions observed during a low-level jet (LLJ), turbulence is not isotropic and σH2 values below 200 m are generally stronger than the σw2 values at and above that height, but between 250 and 300 m, where the amplitude of the wind shear (and thus also of the turbulence) is reduced, the profiles often agree.

Determination of hσ from the composite συ2 profiles follows the same procedure, with the same thresholds, described in the previous section for σw2. If the boundary layer extended above the 180-m minimum range, then the estimates for hσ will be the same as those determined using just σw2 profiles. Figure 3c (bottom) contains an example of the composite profiles, with the estimate of hσ, for data acquired 31 August 2006. The composite profiles sometimes demonstrate a discontinuity in the data around 330 m (where συ2 estimates based on elevation scan data change to συ2 based on zenith data); however, the altitude at which the variance (whether derived from horizontal or vertical velocity data) drops below threshold is not typically affected by this discontinuity.

c. Mean horizontal wind shear profiles

Wind shear generates mechanical or “forced” turbulence, especially during neutral or stable atmospheric conditions (Brown 1972; Stull 1988), although shear may also be present at the top of a convective boundary layer. In a shear-driven boundary layer the vertical shear profile, expressed in terms of speed shear, directional shear, or vector shear, drives the turbulent mixing and helps to define the shear-based MH candidates. This section describes the methods used for estimation of surface-based heights hspd, hdir, and hvec, using the respective shear profiles.

Balsley et al. 2006 and Banta et al. 2006 have demonstrated that the point at which a strong shear profile exhibits a minimum corresponds to the top of the layer of significant surface-connected turbulence. For the purpose of estimating the speed shear height hspd, one first searches for mean wind profiles, such as the one shown in Fig. 4, where the speed shear at or just above the surface is greater than 0.02 s−1. If the profile qualifies, then the height at which the speed shear drops below a threshold of 0.005 s−1 is chosen for hspd, as indicated by a large “+” in Fig. 4. The threshold of 0.02 s−1 for shear near the surface represents the minimum value for which mechanically induced surface-connected turbulence, as determined using HRDL data, was consistently observed; however, near-surface shear levels of 0.05 s−1 (i.e., a 2.5 m s−1 change in wind speed over a 50-m change in altitude) or greater were commonly observed. The 0.005 s−1 threshold used to define the “top” of the shear was empirically chosen based on comparisons to radiosonde data and συ2 profiles. The resulting hspd values obtained from this method during a shear driven BL demonstrate good correlation with MH determined using other methods such as radiosonde RH and θ profiles or συ2 profiles, as will be demonstrated in section 5f. An example of hspd estimates found using 24 h of mean horizontal wind speed profiles is provided at the end of this section.

A similar algorithm determines hdir from wind direction shear profiles using a qualifying shear threshold of 1° m−1. The hdir is chosen to be the first height where the shear drops below 0.1° m−1. Empirically, the TexAQS data demonstrate that for each m s−1 of reference wind speed, 1° m−1 of direction shear corresponds to a crosswind speed shear of approximately 0.02 s−1 over 10 m, echoing the 0.02 s−1 threshold for hspd, while the 0.1° m−1 corresponds to observed shear profile minima. One may also combine the speed and directional profiles to generate vector shear profiles and then perform similar thresholding techniques to search for the surface-based height hvec. In most cases in the TexAQS 2006 dataset, however, the vector shear was dominated by speed or directional shear and so vector shear profiles were rarely used. Although speed, directional, and/or vector shear is always present in some form, when the strength of the shear (in speed, degrees, or vector amplitude) does not qualify the profile for an estimate of a shear-based mixing height candidate, the algorithm assigns a flag (i.e., “nan”) to the particular candidate, (hspd, hdir, or hvec) for that profile.

d. Relative aerosol backscatter profiles

The coherent Doppler lidar range function is more complicated than direct detection lidars (Frehlich and Kavaya 1991) and is strongly dependent on current atmospheric conditions (Rye 1981; Belmonte and Rye 2000). Thus HRDL data are not optimized for estimation of absolute aerosol backscatter profiles. This is especially true for the TexAQS dataset because the ship’s plume often blew over the lidar, affecting the zenith beam with its strong refractive turbulence. Nevertheless, SRAB profiles, such as those shown in Fig. 5, can be used for determination of hA because they still provide the gradient information. The discussion in section 3 referred to several methods used for estimating MH using aerosol profiles but the present analysis implements a Haar-wavelet method based on that described by Davis et al. (2000).

In general, most of the references use Sb (range corrected) profiles to determine hA, typically assuming that the BL is characterized by higher concentrations of backscattering aerosol particles than are found above this layer in the “cleaner and drier free troposphere” (Cohn and Angevine 2000). During the TexAQS 2006 experiment, however, the air near the surface was at times observed to be relatively cleaner, and sometimes drier, than air 500 m or so above the surface. Such situations require a more general approach to estimating the height of the surface-based layer by searching for any gradient (negative or positive) in the backscatter profiles.

First a wavelet transform of a 100-m scale is applied and then the lowest-altitude local maximum or minimum in the resulting profile is found. This maximum/minimum typically represents the first major change in aerosol concentration and thus defines hA. The 100-m scale was chosen according to the principles outlined in Davis et al. (2000) and Brooks (2003): 100 m covers three range gates, is large enough to smooth out noise but still catch most of the gradients of interest, and is still small enough to produce filtered estimates close to the profile’s minimum range. Searching for a maximum in the wavelet-filtered SRAB profile enables detection of clean surface-based layers capped by dirtier air aloft. Examples of hA estimates are plotted over their corresponding profiles in Fig. 5.

e. Combined mixing height

Using the methods described in the preceding sections, five different time series, hσ, hA, hspd, hdir, and hvec, were generated with 15-min resolution for each 24-h period of the experiment. Although each of these estimates may be considered a candidate for the MH, the definition of MH implies that σw2 profiles offer the best method to determine this parameter from this dataset. Other measurements, such as mean wind speed and direction profiles or συ2 profiles, generated from the elevation scans may provide MH estimates when σw2 data are unavailable or when the MH is around or below the 180-m zenith minimum range. Estimates for all five types of profile may not be available at times because of a lack of qualifying shear, extremely stable conditions, cloud interference with the SRAB-derived height, or times when certain scans were not performed. In these situations, those time stamps in the particular time series are flagged (as “nan”). Typically, at least one applicable estimate is available for each of the various atmospheric conditions and so, by combining the various methods, coverage for MH estimation during TexAQS 2006 is optimized resulting in over 90% coverage for the experiment.

Assuming all the applicable profiles and their corresponding MH candidate estimates are optimized, the algorithm used to select MH is described as follows: First, according to the definition of MH, all the MH on a given day were assigned to equal hσ as derived from σw2 profiles for that day. The combination process for the remaining estimates requires reviewing the data for each day to observe what types of data are available and to assess the general atmospheric conditions including lack or presence of strong convective mixing, lack or presence of an LLJ, location of the ship, etc. Those times when hσ was unavailable are assigned the next most applicable height (i.e., hA, hspd, or hσ from συ2 profiles) based on ship location and conditions. Aerosol profiles could be used during convective periods; but for more stable periods, συ2-based estimates were used. During nights when the ship was near land, mixing heights were often determined by the shape of the nocturnal LLJ, and so hspd was used for the MH estimates during those nights. In general, if strong wind shear was present, capping the surface mixing, the shear based values could be used during periods when hσ were unavailable. There were a few times when qualifying shear was observed near the surface but convective mixing, as observed in the συ2 profiles, reached through this shear, and so the hσ was a better MH estimate. Continuity should also play a role in choosing the type of heights to assign for MH. For example, if MH based on hσ measured near shore drops during the evening because of the formation of the nocturnal LLJ, then the choice for MH should switch to either hσ based on horizontal variance profiles or hspd. If periods of undetermined MH still exist, a third type of height may be applied if useful data are available for those times. A final manual pass through may be necessary to eliminate problematic data and/or unexplained discontinuities.

A typical scenario for combining complementary profiles and their corresponding MH candidates occurred on 11 August 2006 when the ship was in the Houston ship channel overnight. As shown in Fig. 6a, σw2 profiles provide hσ and thus MH estimates for most of the daylight hours. At night, however, a strong LLJ produced a stable boundary layer and a strong shear layer near the surface. Figure 6a shows that low σw2 values were observed at and above 180 m during much of the night (0300–1400 UTC) indicating that the MH fell below that level during this time. The corresponding mean wind speed profiles that extend down to within a few meters of the surface (displayed using color for wind speed in Fig. 6b) show the LLJ and resulting strong shear, which generates mechanical turbulence, during this period. In each of the LLJ profiles, the shear falls below threshold near the altitude at which radiosonde θ and RH profiles (when available) show the top of the nocturnal inversion layer as noted by Balsley et al. (2006). Similarly, Banta et al. 2006 demonstrated that the point at which LLJ shear exhibits a minimum typically corresponds to the top of the layer of increased surface-connected turbulence. The speed-shear-derived hspd values for this nocturnal LLJ are plotted over their corresponding mean wind speed profiles in Fig. 6b. To verify that these heights are valid estimates for the MH, they are plotted, along with the daytime hσ estimates, on top of composite συ2 profiles displayed in Fig. 6c. The συ2 profiles demonstrate weak velocity variance above the MH chosen from the shear heights hspd, and thus the combination of hσ and hspd values serves to provide all the MH values for this particular day. Later in the day (around 2100 UTC) qualifying shear is again observed near the surface but the συ2 profiles demonstrate continuous turbulence through that shear height, and so hσ is used for the MH during that period.

Similar nocturnal LLJ scenarios were frequently observed during the ship-based portion of TexAQS 2006. When MH cannot be estimated using σw2 profiles, the low-level speed shear and σH2 profiles may both be used to estimate the MH, but our confidence is typically higher in the shear profiles because they are less noisy. Estimates of hA based on SRAB profiles occasionally proved useful when the ship was in the Gulf of Mexico and σw2 profiles were unreliable.

f. Validation comparisons

Continuous validation of the ship-based Doppler-lidar derived MH is not possible because of the lack of comparable instruments on the ship. Fortunately, during the R/V Brown portion of TexAQS 2006, radiosondes were launched at 0500, 1100, 1700, and 2300 UTC. The data provided by these radiosondes were used to derive θ profiles that were used with the corresponding RH profile to hand-pick the MH based on the gradient method (Hennemuth and Lammert 2005; Martucci et al. 2007). As noted by Beyrich (1997) and Balsley et al. (2006), turbulence in a stable boundary layer does not have to be uniform and MH is not necessarily the same as the inversion height, as may be determined using radiosonde data. These radiosonde-based estimates of MH are compared to the corresponding HRDL-derived MH by finding the HRDL-based MH within 15 min of each radiosonde profile time stamp and interpolating these onto the radiosonde’s time base. Figure 7 contains a plot of RH-and-θ-derived MH against HRDL-derived MH. The correlation between these two time series is approximately 0.91 (best fit slope is 1.09 with an intercept of −16 m). The largest differences were observed during convective periods when the lateral displacement of the radiosonde from the ship location was at least 1 km by the time the radiosonde reached 1 km altitude, indicating that the radiosonde could be sampling a different air mass from that observed by HRDL from the ship.

Correlations performed between corresponding radiosonde RH-and-θ-derived MH and the individual hσ, hA, hspd, hdir, and hvec estimates for the same 1 August to 11 September time period found a correlation of r = 0.87 with the hσ (with 99 points for comparison), r = 0.37 with hA (93 points), r = 0.01 with hspd and hvec (34 points each), and r = −0.01 with hdir (27 points). The number of points for comparison changes because there was not an MH candidate for each type of lidar profile within 15 min of each of the 158 sonde profile launches. The differences between the each type of HRDL-derived MHs and radiosonde-derived MHs inevitably raise questions about the causes for each difference. If sonde locality and drift can be ruled out, and in most cases during TexAQS 2006 they cannot, the differences likely represent situations that require careful analysis and consideration of all available measurements in the form of case studies. Incorporating all available measurements from different instruments, while keeping in mind the assumptions and limitations of each, results in a better understanding and characterization of the boundary layer at any given time.

The results of applying these methods to data from the ship-based portion of TexAQS 2006 are summarized in Fig. 8a as an image of MH estimates displayed using intensity versus time of day (x axis) and day of the experiment (y axis). This display highlights the diurnal cycles in mixing height observed when the ship was close to land. Figure 8b shows a similar image showing the type of HRDL profile used for the estimation of MH. Note that most of the MH estimates were obtained using σw2 profile data. Wind shear (direction, speed, or vector shear) were often used when the ship was near land, especially during nocturnal LLJs that typically occurred between 0200 and 1500 UTC. Aerosol estimates were most useful when the ship was in the Gulf of Mexico, when σw2 gradients were weak or σw2 was poorly defined because of the weakly unstable conditions over the water.

6. Summary and conclusions

A new composite method for estimation of the mixing height using various forms of Doppler lidar data has been presented. The method has demonstrated wide applicability via application to Doppler lidar data acquired under various atmospheric conditions and in various locations during the ship-based portion of TexAQS 2006. Having a motion-stabilized scanning coherent Doppler lidar, such as HRDL, that can perform azimuth scans, elevation scans, and zenith stares to obtain velocity and relative aerosol backscatter data in three dimensions enables measurement of profiles of horizontal and vertical turbulent wind statistics, aerosol backscatter, and mean horizontal wind speed and direction from a moving platform. Methods for estimating surface-based heights hσ, hA, hspd, hdir, and hvec using such profiles have been developed and these heights have been combined to produce a single time series of mixing height (MH). Profiles of vertical velocity variance are, by the definition of the MH, the most useful data for estimation of MH, especially during convective conditions. Comparison of the HRDL-derived estimates with radiosonde RH and θ-derived estimates demonstrates both the quality of the estimates as well as the variability of mixing heights between radiosonde launches.

The MH for the ship-based portion of TexAQS 2006 was determined more than 90% of the time using these techniques applied to HRDL data. Several studies that incorporate these mixing height estimates to better understand aerosol and gas-phase chemistry, coastal meteorological processes, aerosol transport, and the causes of high pollution events in Houston–Galveston region are currently underway (Bates et al. 2008).

Acknowledgments

The authors would like to acknowledge support from the NOAA Health of the Atmosphere program, the Texas Commission on Environmental Quality, the officers and crew of the R/V Ronald H. Brown. The authors also thank Dan Wolfe of NOAA/ESRL for providing the radiosonde data.

REFERENCES

  • Balsley, B. B., Frehlich R. G. , Jensen M. L. , and Meillier Y. , 2006: High-resolution in situ profiling through the stable boundary layer: Examination of the SBL top in terms of minimum shear, maximum stratification, and turbulence decrease. J. Atmos. Sci., 63 , 12911307.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Banta, R. M., Newsom R. K. , Lundquist J. K. , Pichugina Y. L. , Coulter R. , and Mahrt L. , 2002: Nocturnal low-level jet characteristics over Kansas during CASES-99. Bound.-Layer Meteor., 105 , 221252.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Banta, R. M., Pichugina Y. L. , and Newsom R. K. , 2003: Relationship between low-level jet properties and turbulence kinetic energy in the nocturnal stable boundary layer. J. Atmos. Sci., 60 , 25492555.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Banta, R. M., Pichugina Y. L. , and Brewer W. A. , 2006: Turbulent velocity-variance profiles in the stable boundary layer generated by a nocturnal low-level jet. J. Atmos. Sci., 63 , 27002719.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Banta, R. M., Mahrt L. , Vickers D. , Sun J. , Balsley B. B. , Pichugina Y. L. , and Williams E. J. , 2007: The very stable boundary layer on nights with weak low-level jets. J. Atmos. Sci., 64 , 30683090.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bates, T. S., and Coauthors, 2008: Boundary layer aerosol chemistry during TexAQS/GoMACCS 2006: Insights into aerosol sources and transformation processes. J. Geophys. Res., 113 , D00F01. doi:10.1029/2008JD010023.

    • Search Google Scholar
    • Export Citation
  • Belmonte, A., and Rye B. J. , 2000: Heterodyne lidar returns in the turbulent atmosphere: Performance evaluation of simulated systems. Appl. Opt., 39 , 24012411.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Beyrich, F., 1997: Mixing height estimation from sodar data: A critical discussion. Atmos. Environ., 31 , 39413953.

  • Boers, R., Eloranta E. W. , and Coulter R. L. , 1984: Lidar observations of mixed layer dynamics: Tests of parameterized entrainment models of mixed layer growth rate. J. Appl. Meteor., 23 , 247266.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Brewer, W. A., and Coauthors, 2005: Performance and results from recent ship-borne and ground-based boundary layer measurements using NOAA/ETL coherent Doppler lidars. Proc. 13th Coherent Laser Radar Conf., Kamakura, Japan, 92–95.

    • Search Google Scholar
    • Export Citation
  • Brooks, I., 2003: Finding boundary layer top: Application of a wavelet covariance transform to lidar backscatter profiles. J. Atmos. Oceanic Technol., 20 , 10921105.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Brown, R., 1972: On the inflection point instability of a stratified Ekman boundary layer. J. Atmos. Sci., 29 , 850859.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cohn, S. A., and Angevine W. M. , 2000: Boundary layer height and entrainment zone thickness measured by lidars and wind-profiling radars. J. Appl. Meteor., 39 , 12331247.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cramér, H., 1946: Mathematical Methods of Statistics. Princeton University Press, 575 pp.

  • Davis, K. J., Lenschow D. H. , Oncley S. P. , Kiemle C. , Ehret G. , and Giez A. , 1997: Role of entrainment in surface–atmosphere interactions over a boreal forest. J. Geophys. Res., 102 , 2921929230.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Davis, K. J., Gamage N. , Hagelberg C. R. , Kiemle C. , Lenschow D. H. , and Sullivan P. P. , 2000: An objective method for deriving atmospheric structure from airborne lidar observations. J. Atmos. Oceanic Technol., 17 , 14551468.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Fairall, C. W., and Coauthors, 2006: Turbulent bulk transfer coefficients and ozone deposition velocity in the International Consortium for Atmospheric Research into Transport and Transformation. J. Geophys. Res., 111D , D23S20. doi:10.1029/2006JD007597.

    • Search Google Scholar
    • Export Citation
  • Feingold, G., and Grund C. J. , 1994: On the feasibility of using multi-wavelength lidar measurements to measure cloud condensation nuclei. J. Atmos. Oceanic Technol., 11 , 15431558.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Flamant, C., Pelon J. , Flamant P. H. , and Durand P. , 1997: Lidar determination of the top of the entrainment zone thickness at the top of the unstable marine atmospheric boundary layer. Bound.-Layer Meteor., 83 , 247284.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Frehlich, R. G., and Kavaya M. J. , 1991: Coherent laser radar performance for general atmospheric refractive turbulence. Appl. Opt., 30 , 53255352.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Frehlich, R. G., Meillier Y. , Jensen M. L. , Balsley B. , and Sharman R. , 2006: Measurements of boundary layer profiles in an urban environment. J. Appl. Meteor. Climatol., 45 , 821837.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Grund, C. J., and Coauthors, 2001: High-resolution Doppler lidar for boundary layer and cloud research. J. Atmos. Oceanic Technol., 18 , 376393.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Haegeli, P., Steyn D. G. , and Strawbridge K. B. , 2000: Spatial and temporal variability of mixed-layer depth and entrainment zone thickness. Bound.-Layer Meteor., 97 , 4771.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hanna, S. R., MacDonald C. P. , Lilly M. , Knoderer C. , and Huang C. H. , 2006: Analysis of three years of boundary layer observations over the Gulf of Mexico and its shores. Estuarine Coastal Shelf Sci., 70 , 541550.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hennemuth, B., and Lammert A. , 2005: Determination of the atmospheric boundary layer height from radiosonde and lidar backscatter. Bound.-Layer Meteor., 120 , 181200.

    • Search Google Scholar
    • Export Citation
  • Hill, R. J., Brewer W. A. , and Tucker S. , 2007: Platform-motion correction of velocity measured by Doppler lidar. J. Atmos. Oceanic Technol., 25 , 13691382.

    • Search Google Scholar
    • Export Citation
  • Kiemle, C., and Coauthors, 2007: Latent heat flux profiles from collocated airborne water vapor and wind lidars during IHOP_2002. J. Atmos. Oceanic Technol., 24 , 627639.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lammert, A., and Bösenberg J. , 2006: Determination of the convective boundary-layer height with laser remote sensing. Bound.-Layer Meteor., 199 , 159170.

    • Search Google Scholar
    • Export Citation
  • Lenschow, D. H., Wulfmeyer V. , and Senff C. , 2000: Measuring second-through-fourth-order moments in noisy data. J. Atmos. Oceanic Technol., 17 , 13301347.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lothon, M., Lenschow D. H. , and Mayor S. D. , 2006: Coherence and scale of vertical velocity in the convective boundary layer from a Doppler lidar. Bound.-Layer Meteor., 121 , 521536.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mahrt, L., 1999: Stratified atmospheric boundary layers. Bound.-Layer Meteor., 90 , 375396.

  • Mahrt, L., and Vickers D. , 2006: Extremely weak mixing in stable conditions. Bound.-Layer Meteor., 119 , 1939.

  • Martucci, G., Matthey R. , Mitev V. , and Richner H. , 2007: Comparison between backscatter lidar and radiosonde measurements of the diurnal and nocturnal stratification in the lower troposphere. J. Atmos. Oceanic Technol., 24 , 12311244.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Melfi, S. H., Spinhirne J. D. , Chou S-H. , and Palm S. P. , 1985: Lidar observations of vertically organized convection in the planetary boundary layer over the ocean. J. Climate Appl. Meteor., 24 , 806821.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Menut, L., Flamant C. , Pelon J. , and Flamant P. H. , 1999: Urban boundary-layer height determination from lidar measurements over the Paris area. Appl. Opt., 38 , 945954.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Pichugina, Y. L., and Coauthors, 2008: Horizontal-velocity and variance measurements in the stable boundary layer using Doppler lidar: Sensitivity to averaging procedures. J. Atmos. Oceanic Technol., 25 , 13071327.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rye, B. J., 1981: Refractive-turbulence contribution to incoherent backscatter heterodyne lidar returns. J. Opt. Soc. Amer., 71 , 687691.

  • Rye, B. J., and Hardesty R. M. , 1993a: Discrete spectral peak estimation in incoherent backscatter heterodyne lidar. I. Spectral accumulation and the Cramer–Rao lower bound. IEEE Trans. Geosci. Remote Sens., 31 , 1627.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rye, B. J., and Hardesty R. M. , 1993b: Discrete spectral peak estimation in incoherent backscatter heterodyne lidar. II. Correlogram accumulation. IEEE Trans. Geosci. Remote Sens., 31 , 2835.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Seibert, P., Beyrich F. , Gryning S. E. , Joffre S. , Rasmussen A. , and Tercier P. , 2000: Review and intercomparison of operational methods for the determination of the mixing height. Atmos. Environ., 34 , 10011027.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Senff, C., Bösenberg J. , Peters G. , and Schaberl T. , 1996: Remote sensing of turbulent ozone fluxes and the ozone budget in the convective boundary layer with DIAL and Radar-RASS: A case study. Beitr. Phys. Atmos., 69 , 161176.

    • Search Google Scholar
    • Export Citation
  • Skyllingstad, E. D., Samelson R. M. , Mahrt L. , and Barbour P. , 2005: A numerical modeling study of warm offshore flow over cool water. Mon. Wea. Rev., 133 , 345361.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Smedman, A-S., Bergström H. , and Grisogano B. , 1997: Evolution of stable internal boundary layers over a cold sea. J. Geophys. Res., 102 , 10911099.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Steyn, D. G., Baldi M. , and Hoff R. M. , 1999: The detection of mixed layer depth and entrainment zone thickness from lidar backscatter profiles. J. Atmos. Oceanic Technol., 16 , 953959.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Stull, R. B., 1988: An Introduction to Boundary Layer Meteorology. Kluwer Academic, 664 pp.

  • White, A. B., Senff C. J. , and Banta R. M. , 1999: A comparison of mixing depths observed by ground-based wind profilers and an airborne lidar. J. Atmos. Oceanic Technol., 16 , 584590.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wolfe, D., and Coauthors, 2007: Shipboard multi-sensor wind profiles from New England Air Quality Study 2004: Radar wind profiler, high resolution Doppler lidar, GPS rawindsonde. J. Geophys. Res., 112 , D10S15. doi:10.1029/2006JD007344.

    • Search Google Scholar
    • Export Citation
  • Wulfmeyer, V., Randall M. , Brewer A. , and Hardesty R. M. , 2000: 2-μm Doppler lidar transmitter with high frequency stability and low chirp. Opt. Lett., 25 , 12281230.

    • Crossref
    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

Typical scan schedule for the HRDL on board R/V Brown during TexAQS 2006.

Citation: Journal of Atmospheric and Oceanic Technology 26, 4; 10.1175/2008JTECHA1157.1

Fig. 2.
Fig. 2.

(a) HRDL zenith staring velocity estimates and (b) corresponding estimated σw2 profile from 1830 to 1845 UTC 27 Aug 2006 during TexAQS 2006. The data were acquired while the R/V Brown was station keeping in Barbour’s Cut (Port of Houston). (b) The red line indicates total variance of the velocity estimates. The green line indicates the uncorrelated or instrument portion of the variance and the difference between the two gives the atmospheric σw2 profile (thick blue line). (a),(b) The thick black horizontal line indicates the height at which the atmospheric variance falls below the 0.04 m2 s−2 threshold. (c) HRDL zenith staring small-scale (5–10-min variance estimates) σw2 profiles for 6 days of the first leg. The black line on the image indicates hσ determined using the σw2 profiles, when there is qualifying variance.

Citation: Journal of Atmospheric and Oceanic Technology 26, 4; 10.1175/2008JTECHA1157.1

Fig. 3.
Fig. 3.

(a) LOS velocity in the residual wind field (mean horizontal wind removed) for one sweep of an elevation scan with constant azimuth. Data from such scans can be used to estimate σH2 profiles from the surface up to 330 m as shown in (b) the plot. The same σw2 profile plotted in Fig. 2b is plotted here in blue. The corresponding σH2 profile that reaches down to the surface is plotted in green. The red circle indicates the mixing height candidate hσ = 1.2 km. (c) Composite συ2 profiles for a 24-h period on 31 Aug 2006. The ship was in Barbour’s Cut overnight (red bar at top), and then crossed Galveston Bay (blue bar) before heading out into the Gulf of Mexico (white bar). Data in the lowest 330 m are σH2 profiles derived from elevation scans. Above 330 m, the profiles are of σw2. The black line indicates the MH estimates determined using συ2 with those values determined from σw2 data highlighted with a red circle and those values determined from σH2 data highlighted with a blue “+.” The white-filled, black squares indicate the MH determined using data from radiosonde launches. Sunrise was just before 1200 UTC.

Citation: Journal of Atmospheric and Oceanic Technology 26, 4; 10.1175/2008JTECHA1157.1

Fig. 4.
Fig. 4.

An example wind speed shear profile from the TexAQS 2006 data. The circle indicates the maximum shear and the “+” at about 190 m indicates hspd, where the shear drops below the 0.005 m s−1 threshold (drawn as a dashed vertical line).

Citation: Journal of Atmospheric and Oceanic Technology 26, 4; 10.1175/2008JTECHA1157.1

Fig. 5.
Fig. 5.

HRDL zenith staring 2-μm SRAB values (log-10 scale) for the same time period shown in Fig. 2c (10–16 Aug 2006). The black line connects adjacent estimates of the aerosol height hA indicated by x’s. No estimates are made when a cloud (high-intensity return) is present.

Citation: Journal of Atmospheric and Oceanic Technology 26, 4; 10.1175/2008JTECHA1157.1

Fig. 6.
Fig. 6.

(a) The σw2 profiles, (b) horizontal mean wind speed profiles, and (c) συ2 profiles for a 24-h period on 11 Aug 2006. The ship was in the Houston ship channel overnight.

Citation: Journal of Atmospheric and Oceanic Technology 26, 4; 10.1175/2008JTECHA1157.1

Fig. 7.
Fig. 7.

Correlation plot of ship-based lidar-derived mixing heights to balloon-sonde-derived heights (based on θ and RH profiles) for the 1 Aug–11 Sep 2006. Each comparison point represents a comparison between a sonde launch and a lidar-derived MH measured within 15 min of the sonde launch; x’s, diamonds, and circles represent heights estimated using συ2, SRAB, and shear profiles, respectively. The thick line has a slope of 1.09 and an intercept of −16 m, which is less than the vertical resolution of the HRDL σw2 and SRAB data. Overall correlation is 0.91. There are 137 total points in the correlation.

Citation: Journal of Atmospheric and Oceanic Technology 26, 4; 10.1175/2008JTECHA1157.1

Fig. 8.
Fig. 8.

(a) The MH shown as color brightness vs time of day (UTC) and day of the year for the entire ship-based portion of the TexAQS 2006 experiment. These data represent all locations encountered during the experiment. The days that demonstrate a clear diurnal cycle (e.g., 10–15 Aug 2006) represent times when the ship was close to shore for most of the day. Gray periods indicate times when no MH estimate is available. (b) Type of HRDL data used to estimate MH. “None” indicates that no height was estimated, usually when the ship was in port. Black outlines indicate those times when the ship was docked or in port.

Citation: Journal of Atmospheric and Oceanic Technology 26, 4; 10.1175/2008JTECHA1157.1

Table 1.

Doppler lidar data derived profiles with corresponding minimum range and surface-based height names.

Table 1.
Save
  • Balsley, B. B., Frehlich R. G. , Jensen M. L. , and Meillier Y. , 2006: High-resolution in situ profiling through the stable boundary layer: Examination of the SBL top in terms of minimum shear, maximum stratification, and turbulence decrease. J. Atmos. Sci., 63 , 12911307.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Banta, R. M., Newsom R. K. , Lundquist J. K. , Pichugina Y. L. , Coulter R. , and Mahrt L. , 2002: Nocturnal low-level jet characteristics over Kansas during CASES-99. Bound.-Layer Meteor., 105 , 221252.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Banta, R. M., Pichugina Y. L. , and Newsom R. K. , 2003: Relationship between low-level jet properties and turbulence kinetic energy in the nocturnal stable boundary layer. J. Atmos. Sci., 60 , 25492555.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Banta, R. M., Pichugina Y. L. , and Brewer W. A. , 2006: Turbulent velocity-variance profiles in the stable boundary layer generated by a nocturnal low-level jet. J. Atmos. Sci., 63 , 27002719.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Banta, R. M., Mahrt L. , Vickers D. , Sun J. , Balsley B. B. , Pichugina Y. L. , and Williams E. J. , 2007: The very stable boundary layer on nights with weak low-level jets. J. Atmos. Sci., 64 , 30683090.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bates, T. S., and Coauthors, 2008: Boundary layer aerosol chemistry during TexAQS/GoMACCS 2006: Insights into aerosol sources and transformation processes. J. Geophys. Res., 113 , D00F01. doi:10.1029/2008JD010023.

    • Search Google Scholar
    • Export Citation
  • Belmonte, A., and Rye B. J. , 2000: Heterodyne lidar returns in the turbulent atmosphere: Performance evaluation of simulated systems. Appl. Opt., 39 , 24012411.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Beyrich, F., 1997: Mixing height estimation from sodar data: A critical discussion. Atmos. Environ., 31 , 39413953.

  • Boers, R., Eloranta E. W. , and Coulter R. L. , 1984: Lidar observations of mixed layer dynamics: Tests of parameterized entrainment models of mixed layer growth rate. J. Appl. Meteor., 23 , 247266.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Brewer, W. A., and Coauthors, 2005: Performance and results from recent ship-borne and ground-based boundary layer measurements using NOAA/ETL coherent Doppler lidars. Proc. 13th Coherent Laser Radar Conf., Kamakura, Japan, 92–95.

    • Search Google Scholar
    • Export Citation
  • Brooks, I., 2003: Finding boundary layer top: Application of a wavelet covariance transform to lidar backscatter profiles. J. Atmos. Oceanic Technol., 20 , 10921105.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Brown, R., 1972: On the inflection point instability of a stratified Ekman boundary layer. J. Atmos. Sci., 29 , 850859.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cohn, S. A., and Angevine W. M. , 2000: Boundary layer height and entrainment zone thickness measured by lidars and wind-profiling radars. J. Appl. Meteor., 39 , 12331247.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cramér, H., 1946: Mathematical Methods of Statistics. Princeton University Press, 575 pp.

  • Davis, K. J., Lenschow D. H. , Oncley S. P. , Kiemle C. , Ehret G. , and Giez A. , 1997: Role of entrainment in surface–atmosphere interactions over a boreal forest. J. Geophys. Res., 102 , 2921929230.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Davis, K. J., Gamage N. , Hagelberg C. R. , Kiemle C. , Lenschow D. H. , and Sullivan P. P. , 2000: An objective method for deriving atmospheric structure from airborne lidar observations. J. Atmos. Oceanic Technol., 17 , 14551468.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Fairall, C. W., and Coauthors, 2006: Turbulent bulk transfer coefficients and ozone deposition velocity in the International Consortium for Atmospheric Research into Transport and Transformation. J. Geophys. Res., 111D , D23S20. doi:10.1029/2006JD007597.

    • Search Google Scholar
    • Export Citation
  • Feingold, G., and Grund C. J. , 1994: On the feasibility of using multi-wavelength lidar measurements to measure cloud condensation nuclei. J. Atmos. Oceanic Technol., 11 , 15431558.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Flamant, C., Pelon J. , Flamant P. H. , and Durand P. , 1997: Lidar determination of the top of the entrainment zone thickness at the top of the unstable marine atmospheric boundary layer. Bound.-Layer Meteor., 83 , 247284.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Frehlich, R. G., and Kavaya M. J. , 1991: Coherent laser radar performance for general atmospheric refractive turbulence. Appl. Opt., 30 , 53255352.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Frehlich, R. G., Meillier Y. , Jensen M. L. , Balsley B. , and Sharman R. , 2006: Measurements of boundary layer profiles in an urban environment. J. Appl. Meteor. Climatol., 45 , 821837.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Grund, C. J., and Coauthors, 2001: High-resolution Doppler lidar for boundary layer and cloud research. J. Atmos. Oceanic Technol., 18 , 376393.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Haegeli, P., Steyn D. G. , and Strawbridge K. B. , 2000: Spatial and temporal variability of mixed-layer depth and entrainment zone thickness. Bound.-Layer Meteor., 97 , 4771.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hanna, S. R., MacDonald C. P. , Lilly M. , Knoderer C. , and Huang C. H. , 2006: Analysis of three years of boundary layer observations over the Gulf of Mexico and its shores. Estuarine Coastal Shelf Sci., 70 , 541550.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hennemuth, B., and Lammert A. , 2005: Determination of the atmospheric boundary layer height from radiosonde and lidar backscatter. Bound.-Layer Meteor., 120 , 181200.

    • Search Google Scholar
    • Export Citation
  • Hill, R. J., Brewer W. A. , and Tucker S. , 2007: Platform-motion correction of velocity measured by Doppler lidar. J. Atmos. Oceanic Technol., 25 , 13691382.

    • Search Google Scholar
    • Export Citation
  • Kiemle, C., and Coauthors, 2007: Latent heat flux profiles from collocated airborne water vapor and wind lidars during IHOP_2002. J. Atmos. Oceanic Technol., 24 , 627639.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lammert, A., and Bösenberg J. , 2006: Determination of the convective boundary-layer height with laser remote sensing. Bound.-Layer Meteor., 199 , 159170.

    • Search Google Scholar
    • Export Citation
  • Lenschow, D. H., Wulfmeyer V. , and Senff C. , 2000: Measuring second-through-fourth-order moments in noisy data. J. Atmos. Oceanic Technol., 17 , 13301347.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lothon, M., Lenschow D. H. , and Mayor S. D. , 2006: Coherence and scale of vertical velocity in the convective boundary layer from a Doppler lidar. Bound.-Layer Meteor., 121 , 521536.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mahrt, L., 1999: Stratified atmospheric boundary layers. Bound.-Layer Meteor., 90 , 375396.

  • Mahrt, L., and Vickers D. , 2006: Extremely weak mixing in stable conditions. Bound.-Layer Meteor., 119 , 1939.

  • Martucci, G., Matthey R. , Mitev V. , and Richner H. , 2007: Comparison between backscatter lidar and radiosonde measurements of the diurnal and nocturnal stratification in the lower troposphere. J. Atmos. Oceanic Technol., 24 , 12311244.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Melfi, S. H., Spinhirne J. D. , Chou S-H. , and Palm S. P. , 1985: Lidar observations of vertically organized convection in the planetary boundary layer over the ocean. J. Climate Appl. Meteor., 24 , 806821.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Menut, L., Flamant C. , Pelon J. , and Flamant P. H. , 1999: Urban boundary-layer height determination from lidar measurements over the Paris area. Appl. Opt., 38 , 945954.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Pichugina, Y. L., and Coauthors, 2008: Horizontal-velocity and variance measurements in the stable boundary layer using Doppler lidar: Sensitivity to averaging procedures. J. Atmos. Oceanic Technol., 25 , 13071327.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rye, B. J., 1981: Refractive-turbulence contribution to incoherent backscatter heterodyne lidar returns. J. Opt. Soc. Amer., 71 , 687691.

  • Rye, B. J., and Hardesty R. M. , 1993a: Discrete spectral peak estimation in incoherent backscatter heterodyne lidar. I. Spectral accumulation and the Cramer–Rao lower bound. IEEE Trans. Geosci. Remote Sens., 31 , 1627.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rye, B. J., and Hardesty R. M. , 1993b: Discrete spectral peak estimation in incoherent backscatter heterodyne lidar. II. Correlogram accumulation. IEEE Trans. Geosci. Remote Sens., 31 , 2835.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Seibert, P., Beyrich F. , Gryning S. E. , Joffre S. , Rasmussen A. , and Tercier P. , 2000: Review and intercomparison of operational methods for the determination of the mixing height. Atmos. Environ., 34 , 10011027.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Senff, C., Bösenberg J. , Peters G. , and Schaberl T. , 1996: Remote sensing of turbulent ozone fluxes and the ozone budget in the convective boundary layer with DIAL and Radar-RASS: A case study. Beitr. Phys. Atmos., 69 , 161176.

    • Search Google Scholar
    • Export Citation
  • Skyllingstad, E. D., Samelson R. M. , Mahrt L. , and Barbour P. , 2005: A numerical modeling study of warm offshore flow over cool water. Mon. Wea. Rev., 133 , 345361.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Smedman, A-S., Bergström H. , and Grisogano B. , 1997: Evolution of stable internal boundary layers over a cold sea. J. Geophys. Res., 102 , 10911099.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Steyn, D. G., Baldi M. , and Hoff R. M. , 1999: The detection of mixed layer depth and entrainment zone thickness from lidar backscatter profiles. J. Atmos. Oceanic Technol., 16 , 953959.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Stull, R. B., 1988: An Introduction to Boundary Layer Meteorology. Kluwer Academic, 664 pp.

  • White, A. B., Senff C. J. , and Banta R. M. , 1999: A comparison of mixing depths observed by ground-based wind profilers and an airborne lidar. J. Atmos. Oceanic Technol., 16 , 584590.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wolfe, D., and Coauthors, 2007: Shipboard multi-sensor wind profiles from New England Air Quality Study 2004: Radar wind profiler, high resolution Doppler lidar, GPS rawindsonde. J. Geophys. Res., 112 , D10S15. doi:10.1029/2006JD007344.

    • Search Google Scholar
    • Export Citation
  • Wulfmeyer, V., Randall M. , Brewer A. , and Hardesty R. M. , 2000: 2-μm Doppler lidar transmitter with high frequency stability and low chirp. Opt. Lett., 25 , 12281230.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Fig. 1.

    Typical scan schedule for the HRDL on board R/V Brown during TexAQS 2006.

  • Fig. 2.

    (a) HRDL zenith staring velocity estimates and (b) corresponding estimated σw2 profile from 1830 to 1845 UTC 27 Aug 2006 during TexAQS 2006. The data were acquired while the R/V Brown was station keeping in Barbour’s Cut (Port of Houston). (b) The red line indicates total variance of the velocity estimates. The green line indicates the uncorrelated or instrument portion of the variance and the difference between the two gives the atmospheric σw2 profile (thick blue line). (a),(b) The thick black horizontal line indicates the height at which the atmospheric variance falls below the 0.04 m2 s−2 threshold. (c) HRDL zenith staring small-scale (5–10-min variance estimates) σw2 profiles for 6 days of the first leg. The black line on the image indicates hσ determined using the σw2 profiles, when there is qualifying variance.

  • Fig. 3.

    (a) LOS velocity in the residual wind field (mean horizontal wind removed) for one sweep of an elevation scan with constant azimuth. Data from such scans can be used to estimate σH2 profiles from the surface up to 330 m as shown in (b) the plot. The same σw2 profile plotted in Fig. 2b is plotted here in blue. The corresponding σH2 profile that reaches down to the surface is plotted in green. The red circle indicates the mixing height candidate hσ = 1.2 km. (c) Composite συ2 profiles for a 24-h period on 31 Aug 2006. The ship was in Barbour’s Cut overnight (red bar at top), and then crossed Galveston Bay (blue bar) before heading out into the Gulf of Mexico (white bar). Data in the lowest 330 m are σH2 profiles derived from elevation scans. Above 330 m, the profiles are of σw2. The black line indicates the MH estimates determined using συ2 with those values determined from σw2 data highlighted with a red circle and those values determined from σH2 data highlighted with a blue “+.” The white-filled, black squares indicate the MH determined using data from radiosonde launches. Sunrise was just before 1200 UTC.

  • Fig. 4.

    An example wind speed shear profile from the TexAQS 2006 data. The circle indicates the maximum shear and the “+” at about 190 m indicates hspd, where the shear drops below the 0.005 m s−1 threshold (drawn as a dashed vertical line).

  • Fig. 5.

    HRDL zenith staring 2-μm SRAB values (log-10 scale) for the same time period shown in Fig. 2c (10–16 Aug 2006). The black line connects adjacent estimates of the aerosol height hA indicated by x’s. No estimates are made when a cloud (high-intensity return) is present.

  • Fig. 6.

    (a) The σw2 profiles, (b) horizontal mean wind speed profiles, and (c) συ2 profiles for a 24-h period on 11 Aug 2006. The ship was in the Houston ship channel overnight.

  • Fig. 7.

    Correlation plot of ship-based lidar-derived mixing heights to balloon-sonde-derived heights (based on θ and RH profiles) for the 1 Aug–11 Sep 2006. Each comparison point represents a comparison between a sonde launch and a lidar-derived MH measured within 15 min of the sonde launch; x’s, diamonds, and circles represent heights estimated using συ2, SRAB, and shear profiles, respectively. The thick line has a slope of 1.09 and an intercept of −16 m, which is less than the vertical resolution of the HRDL σw2 and SRAB data. Overall correlation is 0.91. There are 137 total points in the correlation.

  • Fig. 8.

    (a) The MH shown as color brightness vs time of day (UTC) and day of the year for the entire ship-based portion of the TexAQS 2006 experiment. These data represent all locations encountered during the experiment. The days that demonstrate a clear diurnal cycle (e.g., 10–15 Aug 2006) represent times when the ship was close to shore for most of the day. Gray periods indicate times when no MH estimate is available. (b) Type of HRDL data used to estimate MH. “None” indicates that no height was estimated, usually when the ship was in port. Black outlines indicate those times when the ship was docked or in port.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 1333 422 41
PDF Downloads 1162 352 25