Abstract

The accuracy of the method of azimuth structure function for estimation of the dissipation rate of the kinetic energy of turbulence from an array of radial velocities measured by low-energy micropulse coherent Doppler lidars with conical scanning by a probing beam around the vertical axis has been studied numerically. The applicability of the method in dependence on the turbulence intensity and the signal-to-noise ratio has been determined. The method of azimuth structure function was applied for estimation of the turbulent energy dissipation rate from radial velocities measured by the lidar in the experiments on the coast of Lake Baikal. Two dimensional time–height patterns of the wind turbulence energy dissipation rate were obtained. Part of them were obtained in presence of the atmospheric internal waves (AIWs) and low-level jet streams. It is observed that the wind turbulence in the area occupied by jet streams is very weak. In the process of dissipation of AIWs the wind turbulence strength increases.

© 2017 Optical Society of America

1. Introduction

The rate of dissipation of the kinetic energy of turbulence ε is one of the key parameters of the turbulent wind field. In the inertial range of spatial scales of inhomogeneities of the velocity field, the cascade process of turbulent energy transfer from large-scale inhomogeneities to small-scale ones is fully determined by this parameter and obeys the Kolmogorov 2/3 law [1]. The dissipation rate characterizes the intensity of turbulent processes in the atmosphere, and it is important to know how this parameter changes in space and time. Many papers are devoted to development of methods for remote and, in particular, lidar measurement of the dissipation rate (see monograph [2] and review [3] and references therein). There are also more recent publications concerning this issue [4–7]. Thus, in [6] it was proposed to estimate the dissipation rate within the inertial range of turbulence in the atmosphere from the transverse (azimuth) structure function of the radial velocity measured by a pulsed coherent Doppler wind lidar with conical scanning by a probing beam around the vertical axis. It was shown in [6] that estimates of the dissipation rate ε^L obtained by this method from data of the Halo Photonics Stream Line lidar and values of the dissipation rate ε^S measured simultaneously by a sonic anemometer at the same height are in good agreement. However, in the general case, the error of estimation of the dissipation rate by this method from wind data obtained with micropulse lidars of the class of Stream Line lidar under various atmospheric conditions was not analyzed yet.

The point is that fiber micropulse coherent Doppler lidars (CDL) similar Stream Line lidar are characterized by the very low signal-to-noise ratio at the background concentration of atmospheric aerosol. In [8], a method was developed for estimation of the wind speed and direction with acceptable accuracy from data measured by such lidars which can contain up to 50% of “bad” [9] estimates of the radial velocity. This method uses the procedure of filtering of good estimates of the radial velocity. There are no similar filtering procedures, which could be used in estimation of wind turbulence parameters. Therefore, for determination of turbulence parameters, only slightly noised data, in which the probability of bad estimate of the radial velocity is close to zero, can be used. As a result, the height limit for reconstruction of profiles of turbulent parameters from lidar data is much lower than the maximal height of reconstruction of wind speed and direction profiles, and it is necessary to know previously which data can be used for estimation of turbulence parameters.

Below, in the next part of this paper, we analyze the accuracy of the method of azimuth structure function (MASF) in dependence on the atmospheric conditions: concentration of aerosol (signal-to-noise ratio) and turbulence intensity (dissipation rate). In third section of the paper, we present the results of lidar studies of wind turbulence in the atmospheric boundary layer in the shore zone of Lake Baikal with this method.

2. Theoretical foundation

2.1 Basic equations

During the measurement of the radial velocity by a lidar with the use of conical scanning by the probing beam around the vertical axis, the beam elevation angle φ is fixed, and the azimuth angle θ with time t varies from 0° to 360° with the angular scanning rate ωc. The raw data of measurements by the Stream Line lidar are arrays of estimates of correlation functions of the complex signal C^(lTs;Rk,θm,n), where l=0,1,...,6; Ts=1/B; B = 50 MHz is the receiver bandwidth; Rk=R0+kΔR is the distance from the lidar to the center of the sensing volume; ΔR is the range step, which is usually taken equal to 18 or 30 m; k=0,1,2,...,K; θm=mΔθ is the azimuth angle; m=1,2,3,...,M; Δθ is the resolution in the azimuth angle, and n=1,2,3,...,N is the scan number. For every correlation function C^(lTs;Rk,θm,n), the accumulation of data with the same number Na of probing pulses is used. In the case of the Stream Line lidar, the pulse repetition frequency isfP = 15 kHz. If Na = 1500 (the minimal possible value Na for this lidar), the time for measurement of the correlation function of the complex lidar signal is TC=Na/fP = 0.1 s. Then, of the scanning rate ωc = 10°/s, the resolution in the azimuth angle is Δθ=ωcTC = 1°, the number of rays for one full scan is M = 360, and the duration of one scan is Tscan=TCM = 36 s.

For estimating turbulence parameters from wind data, it is important to conduct measurements with the high resolution in the azimuth angle. At the same time, the tangential velocity of motion of the sensing volume Vt=ωcRkcosφ should exceed significantly the average wind velocity at the height of measurement.

The estimates of the correlation functions C^(lTs;Rk,θm,n) are used to calculate the Doppler spectra S^D(f;Rk,θm,n) [10], and the positions of their maximum are then used to obtain the array of estimates of the radial velocities V^r(Rk,θm,n). In addition, the array of estimates of the signal-to-noise ratio is determined from the array C^(lTs;Rk,θm,n) as SN^R(Rk,θm,n)=C^(0;Rk,θm,n)/PN1, where PN is the noise power. The further procedure of processing of lidar data in order to obtain estimates of the dissipation rate ε^(hk) at heights hk=Rksinφ consists in the following.

For every height hk, the vector of average wind velocity V={Vz,Vx,Vy} is determined from the array of estimates of radial velocities V^r(Rk,θm,n) with the aid of sine-wave fitting [2], and fluctuations of the radial velocity are calculated as

V^r(Rk,θm,n)=V^r(Rk,θm,n)SmV(hk),
where Sm={sinφ,cosφcosθm,cosφsinθm}. Then, estimates of the azimuth structure function of the radial velocity measured by the lidar are determined with the following equation:

D^L(mΔθ)=N1n=1N(Mm)1m=1Mm[V^r(Rk,θm+mΔθ,n)V^r(Rk,θm,n)]2.

In Eq. (2), m=1,2,...,M and MΔθ<<π/2 (here Δθ is measured in radians). The condition MΔθ<<π/2 allows the azimuth structure function DL(mΔθ) to be replaced with the transverse [11] structure function of the radial velocity D(mΔyk), where Δyk=ΔθRkcosφ is transverse (with respect to the probing beam optical axis) size of the sensing volume. The estimate of the turbulence energy dissipation rate can be found from the difference of lidar estimates of the azimuth structure function ΔD^L(MΔθ)=D^L(MΔθ)D^L(Δθ) [6], 1) if MΔy does not exceed the upper boundary of the inertial range of turbulence, 2) the probability of bad estimate [9] of the radial velocity Pb is zero, and 3) V^r(Rk,θm,n) can be represented in the form [2]

V^r(Rk,θm,n)=V¯r(Rk,θm,n)+Ve(Rk,θm,n).
In Eq. (3), V¯r(Rk,θm,n) is the radial velocity averaged over the sensing volume, and Ve(Rk,θm,n) is the random instrumental error having properties of white noise with zero mean and the variance σe2=<Ve2>.

The probability of bad estimate of the radial velocity from lidar data depends on atmospheric conditions, mostly, on the concentration of aerosol particles, which manifests itself from the signal-to-noise ratio. Thus, to determine the conditions of applicability of MASF for estimation of the dissipation rate, it is necessary to find the minimal value of the signal-to-noise ratio SNR, at which the probability of bad estimate Pb in the array of estimates V^r(Rk,θm,n) obtained from measurements micropulse lidar is close to zero (does not exceed an arbitrarily small preset value). Toward this end, we should know the dependence Pb(SNR). This dependence can be found on the basis of the theory developed in Ref [12], or using numerical simulation of random realizations of lidar estimates of radial velocity.

Using the results of numerical simulation of lidar signals at different values of signal-to-noise ratio SNR(description of the simulation algorithm is given in [2,9]), we found the dependence Pb(SNR) for the Stream Line lidar with the following parameters: laser wavelength λ = 1.5 μm, probing pulse duration τp = 170 ns, and the width of the time window used to estimate the Doppler spectra τw = 120 ns. In simulation we set that the noise component of the lidar signal is a white noise with Gaussian probability density function (PDF) and neglected by homogeneity of the radial velocity within the sensing volume. It is valid in the case of weak wind turbulence and absence of a strong wind shear. For spectral accumulation we used Na = 1500 laser shots. For each SNR value, we obtained 106 statistically independent radial velocity estimates V^r. From the obtained array of simulated radial velocity estimates, we calculated the PDF of the radial velocity estimate p(V^r). Then we determined the probability (or fraction) Pb of a bad estimate of the radial velocity by minimizing the mean square difference between the PDF p(V^r) obtained in the numerical experiment and the model PDF pM(V^r) described by Eq. (34) in Ref [9].

From the obtained dependence Pb(SNR)at Na = 1500 it follows that in the case of M = 360 and N = 20 the probability Pb is less than (MN)1, if SNR0.03 (−15.2 dB). We also obtained dependencies Pb(SNR) for other values of accumulation number Na. Analysis of the obtained results has shown that the probability of a bad estimate of the radial velocity is a function of the product of SNRand Na, at least for Na1000. Therefore, if for Na = 1500 (TC = 0.1 s) the condition Pb<(MN)1 (M = 360 and N = 20) is fulfilled, when SNR0.03 (−15.2 dB), then for Na = 15000 (TC = 1 s) this condition is realized at SNR0.03/10 (−20.2 dB).

Thus, obtained relations determine the measurement conditions at which the azimuth structure function of the radial velocity can be calculated from the Stream Line lidar array of radial velocity estimates without using a special data filtration procedure.

2.2 Azimuth structure function of the radial velocity

The condition MΔykLI, where LI is the outer scale of the inertial range, allows the turbulence to be considered as isotropic in the calculation of the azimuth structure function of the radial velocity DL(mΔθ)=<D^L(mΔθ)>. Then, assuming horizontal statistical homogeneity and stationarity of the turbulent wind field, with allowance for Eq. (3), for DL(mΔθ) we obtain

DL(mΔθ)=D¯(mΔθ)+2σe2,
where D¯(mΔθ)=<[V¯r(Rk,θm+mΔθ,n)V¯r(Rk,θm,n)]2> is the azimuth structure function of the radial velocity averaged over the sensing volume and V¯r=V¯r<Vr>. For calculation of the structure function D¯(mΔθ) in Eq. (4) at σe2 = 0, we used method Monte Carlo based on the algorithm of computer simulation of random realizations of two-dimensional distributions of the radial velocity Vr(R+z,θ) [2,13].

Figure 1 shows an example of simulation of random realizations of the dependence on the azimuth angle θm for the radial velocity Vr(R,θm) (blue curve) and the lidar estimate of the radial velocity from positions of maxima of the Doppler spectra V^r(R,θm) (red curve). In simulation we set the average wind speed 10 m/s, the elevation angle φ = 60°, the distance to the center of the sensing volume R = 344 m (height h 300 m), resolution in the azimuth angle Δθ = 1° (Δyk = 3 m), the turbulence energy dissipation rate ε = 10−2 m2/s3, and the integral scale of correlation of wind velocity fluctuations LV = 250 m. Since σe2 = 0, according to Eq. (3) the lidar estimate of the radial velocity is V^r(R,θm)=V¯r(R,θm). One can see that the red curve in Fig. 1 is smoother in comparison with the blue curve, that is, the averaging over the sensing volume takes place. In the case of the Stream Line lidar, the longitudinal dimension of the sensing volume is Δz = 30 m.

 figure: Fig. 1

Fig. 1 Azimuth dependence of the average radial velocity (black curve), radial velocity at the center of the sensing volume moving over the scanning cone base (blue curve), and lidar estimate of the radial velocity (red curve).

Download Full Size | PPT Slide | PDF

According to Eq. (4), the difference structure function ΔDL(mΔθ) = DL(mΔθ)DL(Δθ) is independent of the instrumental error of the estimate of radial velocity σe and is a difference structure function averaged over sensing volume ΔD¯(y)=D¯(mΔθ)D¯(Δθ), where y=mΔyk. Figure 2 demonstrates the effect of averaging over the sensing volume on the difference structure function.

 figure: Fig. 2

Fig. 2 Difference structure functions ΔD(y) (blue curve), ΔD¯max(y) (red curve), and ΔD¯cnt(y) (green curve) at ε = 10−3 m2/s3 and LV = 300 m.

Download Full Size | PPT Slide | PDF

Figure 2 shows the calculated by Monte Carlo method difference structure functions ΔD(y)=D(mΔθ)D(Δθ), where D(mΔθ)=<[Vr(R,θm+mΔθ)Vr(R,θm)]2> is the structure function of the radial velocity (without averaging over the sensing volume), and ΔD¯max(y), and ΔD¯cnt(y), where the subscripts “max” and “cnt” indicate that the radial velocity was estimated, respectively, from the position of the spectral maximum and from the centroid of the spectral distribution. It can be seen that, due to the averaging of the radial velocity over the sensing volume, ΔD¯cnt(y) and ΔD¯max(y) are less than ΔD(y) in the considered range of y variation. Therefore, when estimating the dissipation rate from the lidar data, one should necessarily take into account the averaging over the sensing volume.

2.3 Relative error of lidar estimate of the dissipation rate

The relative error of lidar estimate of the dissipation rate Eε=<(ε^/ε1)2> was calculated for Na = 1500, N = 20, M = 360, Δyk = 3 m, and M = 7 (y = 21 m) in dependence the signal-to-noise ratio SNR and magnitude of the dissipation rate ε.

For the ratio of the estimate of dissipation rate ε^ to the given (true) value ε, the following is valid: ε^/ε=[ΔD^L(y)/ΔD¯(y)]3/2, where ΔD^L(y)=D^L(y)D^L(Δyk). We assume that Eε<<1. Then the error Eε can be represented in the form

Eε=32<[ΔD^L(y)ΔD¯(y)]2>ΔD¯(y).
Using Eqs. (2) and (3) with allowance for the inequality M>>1 and the following properties of the random instrumental error of estimation of the radial velocity: <Ve>=<V¯rVe>=<Ve3>=0, <Ve4>=3<Ve2>2 (the probability density function of Ve is Gaussian one [12]), and <Ve(mΔθ,n)Ve(mΔθ,n)>=σe2δmmδnn (δα is the Kronecker delta) in Eq. (5), after carrying out rather cumbersome analytical calculations, we derived the following equation for Eε:
Eε=(3/2)σD¯2+σI2,
where σD¯2 is the relative variance of ΔD^L(y) at σe2 = 0 and

σI2=8(MM)N(1+σe2ΔD¯(y))σe2ΔD¯(y).

The value of σD¯ was calculated from the model data of a pulsed lidar with the use of numerical simulation of random realizations Vr(R+z,θ) and estimation of the radial velocity from the positions of maxima of the Doppler spectra. At N = 20, M = 360, Δyk = 3 m, and M = 7, we have obtained σD¯ = 0.04. This value is independent of the dissipation rate ε. The model difference transverse structure function averaged over sensing volume ΔD¯(y) in accordance with the theory [13] is represented in a form

ΔD¯(y)=[A(mΔyk)A(Δyk)]ε2/3,
where
A(z)=43Ckz2/3G(z),
G(z)=[3πΓ(5/6)]10dξ1(1cosξ1)0dξ2(ξ12+ξ23)43[1+83ξ12ξ12+ξ22]H(ξ1,ξ2;z),
H(ξ1,ξ2;z)=exp[2(Δpξ22z)2]sinc2(Δqξ22z)sinc2(Δykξ12z).
In Eqs. (9)-(11) Ck=2 is the Kolmogorov constant, Δp=cσp/2, σp=τp/(2ln2), Δq=cτw/2, and Γ(x) is a gamma function.

To calculate the instrumental error of estimation of the radial velocity σe as a function of SNR, we used the data of numerical simulation of the lidar signal based on the approach [2,8,9,13]. The simulation was performed under assumption of homogenous wind without taking into account the turbulent fluctuations of wind velocity. It follows from the obtained numerical results that the dependence σe(SNR) for the Stream Line lidar parameters (τp = 170 ns, τw = 120 ns and the sampling interval of the complex lidar signal 20 ns) can be approximated well by the simple formula:

σe=(1+1SNR)ΔvNa,
where Δv = 0.4 m/s is an empirical parameter of fitting.

Rough estimates demonstrate that for such small values of the pulse duration and the time window, the turbulent broadening of the Doppler spectrum is negligible as compared to the instrumental broadening, at least in the case of weak and moderate wind turbulence (ε< 0.001 m2/s3). Thus, at least in the range of the turbulent intensity determined by the condition ε< 0.001 m2/s3, Eq. (12) can be used for estimation of σe.

Figure 3 shows the calculated dependence of the relative error in the lidar estimate of dissipation rate of the kinetic energy of turbulence Eε×100% on the signal-to-noise ratio at the different intensity of wind turbulence. It can be seen that the dissipation rate can be estimated with the acceptable accuracy from measurements by the Stream Line lidar even at the very weak turbulence, when ε = 10−6 m2/s3, if SNR is no less than 0.24.

 figure: Fig. 3

Fig. 3 Relative error in the lidar estimate of dissipation rate as a function of the signal-to-noise ratio for ε = 10−6 (1), 10−5 (2) 10−4 (3), and 10−3 m2/s3 (4).

Download Full Size | PPT Slide | PDF

The data of numerical simulation were used to determine, by Eqs. (6), (7), and (12) the values of ε and SNR, at which σI0.1. With allowance for σD¯ = 0.04, the relative error Eε×100% in this case should not exceed 16%. Figure 4 shows the dependence of the signal-to-noise ratio on the turbulence energy dissipation rate, at which the relative error Eε×100% = 16%.

 figure: Fig. 4

Fig. 4 Signal-to-noise ratio as a function of the turbulence energy dissipation rate at Eε×100% = 16% (solid red curve).

Download Full Size | PPT Slide | PDF

In Fig. 4, the grey color indicate an area of ε and SNR values, at which the error of lidar estimate of dissipation rate exceed 16%. The dashed line in the figure shows the level SNR = 0.03, below which the estimation of the dissipation rate is impossible at any value of ε because of the presence of bad estimates in the measured array of radial velocities. Figure 4 allows us to judge the representativeness of estimation of the dissipation rate ε^ from the measured signal-to-noise ratio SN^R.

3. Experiment

Wind turbulence in the boundary layer of atmosphere was studied on the coast of the Lake Baikal in summer campaigns of 2015 and 2016 with the use of a Stream Line lidar. The lidar was mounted 340 m away from the coast of the lake, on the territory of the Baikal Astrophysical Observatory of Institute of Solar-Terrestrial Physics, Siberian Branch, Russian Academy of Sciences (51°50′47.17”N, 104°53′31.21”E); the lidar altitude above the lake level was 180 m. Using the initial data of these measurements (arrays of radial velocities and signal-to-noise ratios) we retrieved the time–height patterns of the turbulence kinetic energy dissipation rate and calculated the relative error of assessment of the dissipation rate from the lidar data.

3.1 Methodology

Stream Line lidar measurements were carried out under the following parameters: elevation angle φ = 60°; number of pulses used for data accumulation Na = 3000; angular speed of probing beam scanning ωc = 10°/s; measurement time per scanning 36 s; resolution in the azimuth angle Δθ = 2°, and the number of rays per scanning M = 180.

To retrieve the vertical profiles of the wind speed U, the wind direction angle θV, and the vertical component of the wind velocity Vz, the method of filtered sine wave fitting was used [2,8]. As shown in [14] for Na = 3000, the minimal signal to noise ratio is 0.005 (–23 dB), under which the wind speed estimation error does not exceed 0.1 m/s.

To assess the turbulence energy dissipation rate ε, we used the method of transverse (azimuth) structure function of the radial velocity measured within the turbulence inertial interval by the scanning lidar. The time–height patterns of the turbulence energy dissipation rate ε(h,t) were calculated with a height step of 26 m and a time step of 36 s based on the array of radial velocities V^r(Rk,θm,n) measured with the use of data received for N = 25 scannings (measurement time is 15 min) for each vertical profile of ε. For each height h, the SNR values from the array of SN^R(Rk,θm,n) estimates were averaged over all the azimuth angles θm and over N = 25 scannings. Then the relative error of the dissipation rate Eε was calculated by Eqs. (6) and (7). The error of the radial velocity estimate σe, which enters into Eq. (7), was determined from the experimental data by two methods: 1) calculation of σe by Eq. (12), using measured SNR and 2) estimation of σe, using Eq. (22) given in [15]. The second method is applicable for arbitrary intensity of the wind turbulence.

3.2 Results

The analysis of the lidar data processing results revealed several AIW events occurred in the presence of jet streams. The measurements on August 23, 2015, showed an AIW with a period of 9 min and longitudinal and vertical components of the velocity of about 1 m/s and 0.3 m/s, respectively, which was propagating through the region of experiment for almost 6 h in the presence of two jet streams. The initial data measured by the lidar from 11:00 to 17:00 on August 23, 2015, satisfied the condition SNR> 0.05 up to 500 m. We limited to this altitude in the retrieval of the vertical profile of the dissipation rate, while the profiles of the wind velocity components were retrieved up to height of about 900 m.

Figure 5 shows the time–height patterns of the wind speed U(h,t), vertical component of the wind velocity Vz(h,t), ε(h,t), SNR(h,t), σe(h,t), and Eε(h,t). The SNR is below a threshold of 0.005 in the black regions. The values for the error in estimating the radial velocity σe(h,t) presented in Fig. 5 were obtained by the second method (using Eq. (22) in [15]), since in many cases the dissipation rate ε(h,t) exceeded 10−3 m2/s3. These values were used to determine Eε(h,t). SNR varied from 0.05 to 0.2 in the 100–500 m layer during that measurement. The radial velocity error does not exceed 0.16 m/s. The data on the wind speed in Fig. 5 clearly show the presence of two jet streams with speed maxima at altitudes of about 250 m and 750 m, respectively (hereinafter, the altitudes are given relative to the lidar location point). The directions of these jet streams are almost perpendicular to each other (the bottom jet stream was directed from north to south through mountains, and the top jet stream was directed from east to west, i.e., from the lake side) [14]. Near-harmonic oscillations of U(h,t) and Vz(h,t) are clearly seen in Fig. 5 from 14:20 to 15:00. The data on the turbulence energy dissipation rate in Fig. 5 show that ε(h,t) takes values from a wide range, attaining a maximum of about 0.006 m2/s3 before the arising of a jet stream. Relatively strong wind turbulence is also observed under the jet stream. The turbulence is quite weak inside the jet stream, including during an AIW event.

 figure: Fig. 5

Fig. 5 Time–height patterns of the wind speed U, vertical component of the wind velocity Vz, turbulence energy dissipation rate ε, SNR, relative error of the dissipation rate Eε, and the radial velocity error σe retrieved from the Stream Line lidar measurements on the coast of the Lake Baikal on August 23, 2015.

Download Full Size | PPT Slide | PDF

Figures 6 and 7 show the vertical profiles of ε, SNR, Eε, and σe measured with an interval of 1 h (the data are taken from Fig. 5).The turbulence energy dissipation rate obviously can be retrieved from lidar measurements with an error of no larger than 30% at ε~5106m2/s3. The SNR should be no less than 0.075 in this case. The turbulence spatial structure should include an inertial interval with the upper boundary of no less than a half of the longitudinal size of the sensing volume Δz (Δz = 30 m for the Stream Line lidar).

 figure: Fig. 6

Fig. 6 Vertical profiles of the turbulence energy dissipation rate (a), SNR (b), relative error of the dissipation rate estimate (c), and the radial velocity error (d) retrieved from the lidar measurements at 11:30 (black curve), 12:30 (red curve), and 13:30 (blue curve). Data are taken from Fig. 5. Dashed curves show results of calculation by Eqs. (12) and (6).

Download Full Size | PPT Slide | PDF

 figure: Fig. 7

Fig. 7 Vertical profiles of the turbulence energy dissipation rate (a), SNR (b), relative error of the dissipation rate estimate (c), and the radial velocity error (d) retrieved from the lidar measurements at 14:30 (black curve), 15:30 (red curve), and 16:30 (blue curve). Data are taken from Fig. 5. Dashed curves show results of calculation by Eqs. (12) and (6).

Download Full Size | PPT Slide | PDF

In Figs. 6 and 7, the dashed curves show the results for σe and Eε calculated by Eq. (12) and Eq. (6) (the first method). Using all the data for σe(h,t)=σ2 given in Fig. 5 (the second method) and the data for σe(h,t)=σ1 found by the first method, we calculated values b=<η> and d=<η2>, where η=2(σ1σ2)/(σ1+σ2) and angular brackets mean spatial-temporal averaging over all obtained data for σ1 and σ2. As a result, we found that b = −0.07 and d = 0.13. It means that the errors of radial velocity estimate (and, correspondingly, the errors of the turbulent energy dissipation rate estimate) calculated by these two methods differ insignificantly.

In August of 2016, in contrast to 2015, when the concentration of aerosol particles was increased due to forest fires, the measurements were conducted at the rather weak lidar echo signal. Long rains washing out the aerosol from the atmosphere were often falling. Therefore, we succeeded in conducting adequate measurements in the absence of rain only for three days. To increase the signal-to-noise ratio at the height of localization of the low-level jet flow, we often had to refocus the probing beam to the closer distance F = 300 m. The elevation angle φ, as in 2015, was taken equal to 60°, the scanning rate was 5°/s (duration of one conical scan of 72 s), and Na = 3000 (Δθ = 1°, M = 360). Despite unfavorable weather conditions, we still succeeded in revealing one case of appearance of an atmospheric internal wave from the lidar data.

Figure 8 depicts the two-dimensional distributions of the wind speed U(h,t), wind direction angle θV(h,t), and vertical component of the wind vector Vz(h,t) obtained from raw data of the Stream Line lidar as measured on August 6 of 2016. During the measurements, slightly overcast weather was observed approximately to 01:00 Local Time. Nevertheless, a jet flow was observed, in which the height of maximum of the wind velocity decreased from 350 m to 200 m with time. After 5:30 LT, the jet flow practically disappeared, and wind oscillations took place, which especially clearly seen in the figures for the wind direction angle and the vertical component. We believe that these oscillations are associated with appearance of an atmospheric internal wave early in the morning, when the stratification of air temperature is most stable.

 figure: Fig. 8

Fig. 8 Results of spatiotemporal visualization of the wind speed (a), wind direction angle (b), and vertical component (c) from measurements by the Stream Line lidar on shore of Lake Baikal on 06.08.2016.

Download Full Size | PPT Slide | PDF

Since the measurements were conducted under conditions of weak echo signal, it appeared impossible to obtain two-dimensional distributions of the dissipation rate ε(h,t). Only in the thin layer of 250–350 m, owing to the focusing of the probing beam to 300 m, the signal-to-noise ratio met the requirement of very low probability of bad estimate of the radial velocity. To obtain estimates of the dissipation rate in this layer, we used the array of radial velocities measured by the lidar for 25 scans (measurement duration of 30 min).

Figure 9 shows the dynamics of the wind speed, wind direction angle, vertical component of the wind velocity vector, turbulence energy dissipation rate, and relative error of lidar estimate of the dissipation rate at a height of 300 m (480 m above the surface of Lake Baikal). Every point in Fig. 9(a), Fig. 9(b), Fig. 9(c) corresponds to the measurement for 72 s (one full scan). The error of estimation of the dissipation rate Eε was determined as described above in section 2.3. At this height till approximately 07:30 LT, the wind direction was from the shore to the lake. Then, for the relatively short time, the wind direction alternated by approximately 90°.

 figure: Fig. 9

Fig. 9 Dynamics of the wind speed (a), wind direction angle (b), vertical component of the wind vector (c), turbulence energy dissipation rate (d), and relative error of estimation of the dissipation rate (e) at a height of 300 m as obtained from measurements by the Stream Line lidar on shore of Lake Baikal on 06.08.2016.

Download Full Size | PPT Slide | PDF

As can be seen from Fig. 9, since 01:30 LT for five hours at a height of 300 m the vertical component of wind Vz Fig. 9(c) took only negative values (downward flow). Starting from 04:40 LT, the time profile of Vz began to oscillate. The period of these oscillations and their amplitude decreased, on average, with time. According to Fig. 9(c), since 05:00 till 07:30 LT the amplitude of oscillations of the vertical velocity was approximately 0.6 m/s. At the same time, the amplitude of oscillations of the horizontal velocity UFig. 9(a) was, at least, two times smaller. This behavior of the internal wave differs significantly from the results of the experiment in 2015 [14], when we observed nearly harmonic oscillations of the three components of the wind velocity vector with the amplitudes, on average, of 1 m/s for the horizontal speed and 0.3 m/s for the vertical one.

We succeeded in obtaining estimates for the turbulence energy dissipation rate with an acceptable accuracy only for time interval, whose total duration is about 30% of the total measurement time (see Fig. 9(e), Fig. 9(f)). The data of Fig. 9 demonstrate that the turbulence inside the jet flow is very weak. It is very similar that the inertial range of turbulence is completely absent, and the flow in the jet is nearly laminar, low frequency wind variations in the jet are mostly caused by mesoscale processes in the atmosphere. We can also suppose that observed increase in the turbulence intensity after 07:30 LT is caused by decay of the internal wave and transfer of its energy to turbulization of the wind flow.

4. Conclusions

The paper has proposed the method for calculation of the relative error in lidar estimate of the dissipation rate of the kinetic energy of turbulence. The accuracy of the method of azimuth structure function for estimation of the dissipation rate from the array of radial velocities measured by micropulse coherent Doppler lidars at the conical scanning by the probing beam around the vertical axis has been analyzed numerically. It was shown that the validity of this method depends on the magnitude of the dissipation rate measured and the signal to noise ratio. Obtained in the paper quantitative results allow one to determine beforehand the representativeness of lidar estimation of the turbulence energy dissipation rate from the measured signal to noise ratios. It is especially important when using the low energy micropulse CDL such as Halo Photonics commercial lidar system Stream Line.

The method of azimuth structure function was applied for estimation of the turbulent energy dissipation rate from radial velocities measured by the scanning micropulse CDL Stream Line in the boundary layer of atmosphere on the coast of Lake Baikal. Two dimensional time–height patterns of the wind turbulence energy dissipation rate were obtained and accuracy of lidar estimate of the dissipation rate was determined. Few events of the internal atmospheric waves were registered during the experimental campaigns. All the registered AIWs were observed in the presence of low level jet streams. It is observed that the wind turbulence in the area occupied by jet streams is very weak. Possibly, it is absent at all, and wind velocity variations in the area of jet streams are caused by mesoscale atmospheric processes. In the process of dissipation of AIWs the wind turbulence strength increases. These results are important for understanding the turbulence - wave interaction processes in the atmosphere and evaluation of wind energy variations [7,16].

Funding

Russian Science Foundation (No. 14-17-00386-П).

References and links

1. A. N. Kolmogorov, “Local structure of turbulence in incompressible viscous fluid at very large Reynolds numbers,” Doklady AN SSSR 30, 299–303 (1941).

2. V. Banakh and I. Smalikho, Coherent Doppler Wind Lidars in a Turbulent Atmosphere (Artech House, 2013).

3. A. Sathe and J. Mann, “A review of turbulence measurements using ground-based wind lidars,” Atmos. Meas. Tech. 6(11), 3147–3167 (2013). [CrossRef]  

4. A. Sathe, R. Banta, L. Pauscher, K. Vogstad, D. Schlipf, and S. Wylie, “Estimating Turbulence Statistics and Parameters from Groundand Nacelle-Based Lidar Measurements:IEA Wind Expert Report,” (2015).

5. A. Sathe, J. Mann, N. Vasiljevic, and G. Lea, “A six-beam method to measure turbulence statistics using ground-based wind lidars,” Atmos. Meas. Tech. 8(2), 729–740 (2015). [CrossRef]  

6. I. N. Smalikho, V. A. Banakh, A. V. Falits, and Yu. A. Rudi, “Estimation of the turbulent energy dissipation rate from the data measured by the “StreamLine” lidar in the surface layer of atmosphere,” Atmos. Oceanic Opt. 28(10), 901–905 (2015).

7. S. Wu, B. Liu, J. Liu, X. Zhai, C. Feng, G. Wang, H. Zhang, J. Yin, X. Wang, R. Li, and D. Gallacher, “Wind turbine wake visualization and characteristics analysis by Doppler lidar,” Opt. Express 24(10), A762–A780 (2016). [CrossRef]   [PubMed]  

8. V. A. Banakh and I. N. Smalikho, “Wind sensing in an atmospheric boundary layer by means of micropulse coherent Doppler lidars,” Opt. Spectrosc. 121(1), 152–159 (2016). [CrossRef]  

9. R. G. Frehlich and M. J. Yadlowsky, “Performance of mean-frequency estimators for Doppler radar and lidar,” J. Atmos. Ocean. Technol. 11(5), 1217–1230 (1994). [CrossRef]  

10. V. A. Banakh, V. A. Brewer, E. L. Pichugina, and I. N. Smalikho, “Measurements of wind velocity and direction with coherent Doppler lidar in conditions of a weak echo signal,” Atmos. Oceanic Opt. 23(5), 381–388 (2010). [CrossRef]  

11. N. L. Byzova, V. N. Ivanov, and E. K. Garger, Turbulence in Atmospheric Boundary Layer (Gidrometeoizdat, 1989) [in Russian]

12. B. J. Rye and R. M. Hardesty, “Discrete spectral peak estimation in incoherent backscatter heterodyne lidar. I: Spectral accumulation and Cramer-Rao lower bound,” IEEE Trans. Geosci. Remote Sens. 31(1), 16–27 (1993). [CrossRef]  

13. I. N. Smalikho and V. A. Banakh, “Accuracy of estimation of the turbulent energy dissipation rate from wind measurements with a conically scanning pulsed coherent Doppler lidar,” Part I. Algorithm of data processing, Atmos. Oceanic Opt. 26(5), 404–410 (2013). [CrossRef]  

14. V. A. Banakh and I. N. Smalikho, “Lidar observations of atmospheric internal waves in the boundary layer of atmosphere on the coast of Lake Baikal,” Atmos. Meas. Tech. 9(10), 5239–5248 (2016). [CrossRef]  

15. I. N. Smalikho and V. A. Banakh, “Measurements of wind turbulence parameters by a conically scanning coherent Doppler lidar in the atmospheric boundary layer,” Atmos. Meas. Tech. Discuss., https://doi.org/10.5194/amt-2017-140 (2017). [CrossRef]  

16. I. N. Smalikho, V. A. Banakh, Y. L. Pichugina, W. A. Brewer, R. M. Banta, J. K. Lundquist, and N. D. Kelley, “Lidar investigation of atmosphere effect on a wind turbine wake,” J. Atmos. Ocean. Technol. 30(11), 2554–2570 (2013). [CrossRef]  

References

  • View by:
  • |
  • |
  • |

  1. A. N. Kolmogorov, “Local structure of turbulence in incompressible viscous fluid at very large Reynolds numbers,” Doklady AN SSSR 30, 299–303 (1941).
  2. V. Banakh and I. Smalikho, Coherent Doppler Wind Lidars in a Turbulent Atmosphere (Artech House, 2013).
  3. A. Sathe and J. Mann, “A review of turbulence measurements using ground-based wind lidars,” Atmos. Meas. Tech. 6(11), 3147–3167 (2013).
    [Crossref]
  4. A. Sathe, R. Banta, L. Pauscher, K. Vogstad, D. Schlipf, and S. Wylie, “Estimating Turbulence Statistics and Parameters from Groundand Nacelle-Based Lidar Measurements:IEA Wind Expert Report,” (2015).
  5. A. Sathe, J. Mann, N. Vasiljevic, and G. Lea, “A six-beam method to measure turbulence statistics using ground-based wind lidars,” Atmos. Meas. Tech. 8(2), 729–740 (2015).
    [Crossref]
  6. I. N. Smalikho, V. A. Banakh, A. V. Falits, and Yu. A. Rudi, “Estimation of the turbulent energy dissipation rate from the data measured by the “StreamLine” lidar in the surface layer of atmosphere,” Atmos. Oceanic Opt. 28(10), 901–905 (2015).
  7. S. Wu, B. Liu, J. Liu, X. Zhai, C. Feng, G. Wang, H. Zhang, J. Yin, X. Wang, R. Li, and D. Gallacher, “Wind turbine wake visualization and characteristics analysis by Doppler lidar,” Opt. Express 24(10), A762–A780 (2016).
    [Crossref] [PubMed]
  8. V. A. Banakh and I. N. Smalikho, “Wind sensing in an atmospheric boundary layer by means of micropulse coherent Doppler lidars,” Opt. Spectrosc. 121(1), 152–159 (2016).
    [Crossref]
  9. R. G. Frehlich and M. J. Yadlowsky, “Performance of mean-frequency estimators for Doppler radar and lidar,” J. Atmos. Ocean. Technol. 11(5), 1217–1230 (1994).
    [Crossref]
  10. V. A. Banakh, V. A. Brewer, E. L. Pichugina, and I. N. Smalikho, “Measurements of wind velocity and direction with coherent Doppler lidar in conditions of a weak echo signal,” Atmos. Oceanic Opt. 23(5), 381–388 (2010).
    [Crossref]
  11. N. L. Byzova, V. N. Ivanov, and E. K. Garger, Turbulence in Atmospheric Boundary Layer (Gidrometeoizdat, 1989) [in Russian]
  12. B. J. Rye and R. M. Hardesty, “Discrete spectral peak estimation in incoherent backscatter heterodyne lidar. I: Spectral accumulation and Cramer-Rao lower bound,” IEEE Trans. Geosci. Remote Sens. 31(1), 16–27 (1993).
    [Crossref]
  13. I. N. Smalikho and V. A. Banakh, “Accuracy of estimation of the turbulent energy dissipation rate from wind measurements with a conically scanning pulsed coherent Doppler lidar,” Part I. Algorithm of data processing, Atmos. Oceanic Opt. 26(5), 404–410 (2013).
    [Crossref]
  14. V. A. Banakh and I. N. Smalikho, “Lidar observations of atmospheric internal waves in the boundary layer of atmosphere on the coast of Lake Baikal,” Atmos. Meas. Tech. 9(10), 5239–5248 (2016).
    [Crossref]
  15. I. N. Smalikho and V. A. Banakh, “Measurements of wind turbulence parameters by a conically scanning coherent Doppler lidar in the atmospheric boundary layer,” Atmos. Meas. Tech. Discuss., https://doi.org/10.5194/amt-2017-140 (2017).
    [Crossref]
  16. I. N. Smalikho, V. A. Banakh, Y. L. Pichugina, W. A. Brewer, R. M. Banta, J. K. Lundquist, and N. D. Kelley, “Lidar investigation of atmosphere effect on a wind turbine wake,” J. Atmos. Ocean. Technol. 30(11), 2554–2570 (2013).
    [Crossref]

2016 (3)

S. Wu, B. Liu, J. Liu, X. Zhai, C. Feng, G. Wang, H. Zhang, J. Yin, X. Wang, R. Li, and D. Gallacher, “Wind turbine wake visualization and characteristics analysis by Doppler lidar,” Opt. Express 24(10), A762–A780 (2016).
[Crossref] [PubMed]

V. A. Banakh and I. N. Smalikho, “Wind sensing in an atmospheric boundary layer by means of micropulse coherent Doppler lidars,” Opt. Spectrosc. 121(1), 152–159 (2016).
[Crossref]

V. A. Banakh and I. N. Smalikho, “Lidar observations of atmospheric internal waves in the boundary layer of atmosphere on the coast of Lake Baikal,” Atmos. Meas. Tech. 9(10), 5239–5248 (2016).
[Crossref]

2015 (2)

A. Sathe, J. Mann, N. Vasiljevic, and G. Lea, “A six-beam method to measure turbulence statistics using ground-based wind lidars,” Atmos. Meas. Tech. 8(2), 729–740 (2015).
[Crossref]

I. N. Smalikho, V. A. Banakh, A. V. Falits, and Yu. A. Rudi, “Estimation of the turbulent energy dissipation rate from the data measured by the “StreamLine” lidar in the surface layer of atmosphere,” Atmos. Oceanic Opt. 28(10), 901–905 (2015).

2013 (3)

A. Sathe and J. Mann, “A review of turbulence measurements using ground-based wind lidars,” Atmos. Meas. Tech. 6(11), 3147–3167 (2013).
[Crossref]

I. N. Smalikho, V. A. Banakh, Y. L. Pichugina, W. A. Brewer, R. M. Banta, J. K. Lundquist, and N. D. Kelley, “Lidar investigation of atmosphere effect on a wind turbine wake,” J. Atmos. Ocean. Technol. 30(11), 2554–2570 (2013).
[Crossref]

I. N. Smalikho and V. A. Banakh, “Accuracy of estimation of the turbulent energy dissipation rate from wind measurements with a conically scanning pulsed coherent Doppler lidar,” Part I. Algorithm of data processing, Atmos. Oceanic Opt. 26(5), 404–410 (2013).
[Crossref]

2010 (1)

V. A. Banakh, V. A. Brewer, E. L. Pichugina, and I. N. Smalikho, “Measurements of wind velocity and direction with coherent Doppler lidar in conditions of a weak echo signal,” Atmos. Oceanic Opt. 23(5), 381–388 (2010).
[Crossref]

1994 (1)

R. G. Frehlich and M. J. Yadlowsky, “Performance of mean-frequency estimators for Doppler radar and lidar,” J. Atmos. Ocean. Technol. 11(5), 1217–1230 (1994).
[Crossref]

1993 (1)

B. J. Rye and R. M. Hardesty, “Discrete spectral peak estimation in incoherent backscatter heterodyne lidar. I: Spectral accumulation and Cramer-Rao lower bound,” IEEE Trans. Geosci. Remote Sens. 31(1), 16–27 (1993).
[Crossref]

1941 (1)

A. N. Kolmogorov, “Local structure of turbulence in incompressible viscous fluid at very large Reynolds numbers,” Doklady AN SSSR 30, 299–303 (1941).

Banakh, V. A.

V. A. Banakh and I. N. Smalikho, “Wind sensing in an atmospheric boundary layer by means of micropulse coherent Doppler lidars,” Opt. Spectrosc. 121(1), 152–159 (2016).
[Crossref]

V. A. Banakh and I. N. Smalikho, “Lidar observations of atmospheric internal waves in the boundary layer of atmosphere on the coast of Lake Baikal,” Atmos. Meas. Tech. 9(10), 5239–5248 (2016).
[Crossref]

I. N. Smalikho, V. A. Banakh, A. V. Falits, and Yu. A. Rudi, “Estimation of the turbulent energy dissipation rate from the data measured by the “StreamLine” lidar in the surface layer of atmosphere,” Atmos. Oceanic Opt. 28(10), 901–905 (2015).

I. N. Smalikho, V. A. Banakh, Y. L. Pichugina, W. A. Brewer, R. M. Banta, J. K. Lundquist, and N. D. Kelley, “Lidar investigation of atmosphere effect on a wind turbine wake,” J. Atmos. Ocean. Technol. 30(11), 2554–2570 (2013).
[Crossref]

I. N. Smalikho and V. A. Banakh, “Accuracy of estimation of the turbulent energy dissipation rate from wind measurements with a conically scanning pulsed coherent Doppler lidar,” Part I. Algorithm of data processing, Atmos. Oceanic Opt. 26(5), 404–410 (2013).
[Crossref]

V. A. Banakh, V. A. Brewer, E. L. Pichugina, and I. N. Smalikho, “Measurements of wind velocity and direction with coherent Doppler lidar in conditions of a weak echo signal,” Atmos. Oceanic Opt. 23(5), 381–388 (2010).
[Crossref]

Banta, R.

A. Sathe, R. Banta, L. Pauscher, K. Vogstad, D. Schlipf, and S. Wylie, “Estimating Turbulence Statistics and Parameters from Groundand Nacelle-Based Lidar Measurements:IEA Wind Expert Report,” (2015).

Banta, R. M.

I. N. Smalikho, V. A. Banakh, Y. L. Pichugina, W. A. Brewer, R. M. Banta, J. K. Lundquist, and N. D. Kelley, “Lidar investigation of atmosphere effect on a wind turbine wake,” J. Atmos. Ocean. Technol. 30(11), 2554–2570 (2013).
[Crossref]

Brewer, V. A.

V. A. Banakh, V. A. Brewer, E. L. Pichugina, and I. N. Smalikho, “Measurements of wind velocity and direction with coherent Doppler lidar in conditions of a weak echo signal,” Atmos. Oceanic Opt. 23(5), 381–388 (2010).
[Crossref]

Brewer, W. A.

I. N. Smalikho, V. A. Banakh, Y. L. Pichugina, W. A. Brewer, R. M. Banta, J. K. Lundquist, and N. D. Kelley, “Lidar investigation of atmosphere effect on a wind turbine wake,” J. Atmos. Ocean. Technol. 30(11), 2554–2570 (2013).
[Crossref]

Falits, A. V.

I. N. Smalikho, V. A. Banakh, A. V. Falits, and Yu. A. Rudi, “Estimation of the turbulent energy dissipation rate from the data measured by the “StreamLine” lidar in the surface layer of atmosphere,” Atmos. Oceanic Opt. 28(10), 901–905 (2015).

Feng, C.

Frehlich, R. G.

R. G. Frehlich and M. J. Yadlowsky, “Performance of mean-frequency estimators for Doppler radar and lidar,” J. Atmos. Ocean. Technol. 11(5), 1217–1230 (1994).
[Crossref]

Gallacher, D.

Hardesty, R. M.

B. J. Rye and R. M. Hardesty, “Discrete spectral peak estimation in incoherent backscatter heterodyne lidar. I: Spectral accumulation and Cramer-Rao lower bound,” IEEE Trans. Geosci. Remote Sens. 31(1), 16–27 (1993).
[Crossref]

Kelley, N. D.

I. N. Smalikho, V. A. Banakh, Y. L. Pichugina, W. A. Brewer, R. M. Banta, J. K. Lundquist, and N. D. Kelley, “Lidar investigation of atmosphere effect on a wind turbine wake,” J. Atmos. Ocean. Technol. 30(11), 2554–2570 (2013).
[Crossref]

Kolmogorov, A. N.

A. N. Kolmogorov, “Local structure of turbulence in incompressible viscous fluid at very large Reynolds numbers,” Doklady AN SSSR 30, 299–303 (1941).

Lea, G.

A. Sathe, J. Mann, N. Vasiljevic, and G. Lea, “A six-beam method to measure turbulence statistics using ground-based wind lidars,” Atmos. Meas. Tech. 8(2), 729–740 (2015).
[Crossref]

Li, R.

Liu, B.

Liu, J.

Lundquist, J. K.

I. N. Smalikho, V. A. Banakh, Y. L. Pichugina, W. A. Brewer, R. M. Banta, J. K. Lundquist, and N. D. Kelley, “Lidar investigation of atmosphere effect on a wind turbine wake,” J. Atmos. Ocean. Technol. 30(11), 2554–2570 (2013).
[Crossref]

Mann, J.

A. Sathe, J. Mann, N. Vasiljevic, and G. Lea, “A six-beam method to measure turbulence statistics using ground-based wind lidars,” Atmos. Meas. Tech. 8(2), 729–740 (2015).
[Crossref]

A. Sathe and J. Mann, “A review of turbulence measurements using ground-based wind lidars,” Atmos. Meas. Tech. 6(11), 3147–3167 (2013).
[Crossref]

Pauscher, L.

A. Sathe, R. Banta, L. Pauscher, K. Vogstad, D. Schlipf, and S. Wylie, “Estimating Turbulence Statistics and Parameters from Groundand Nacelle-Based Lidar Measurements:IEA Wind Expert Report,” (2015).

Pichugina, E. L.

V. A. Banakh, V. A. Brewer, E. L. Pichugina, and I. N. Smalikho, “Measurements of wind velocity and direction with coherent Doppler lidar in conditions of a weak echo signal,” Atmos. Oceanic Opt. 23(5), 381–388 (2010).
[Crossref]

Pichugina, Y. L.

I. N. Smalikho, V. A. Banakh, Y. L. Pichugina, W. A. Brewer, R. M. Banta, J. K. Lundquist, and N. D. Kelley, “Lidar investigation of atmosphere effect on a wind turbine wake,” J. Atmos. Ocean. Technol. 30(11), 2554–2570 (2013).
[Crossref]

Rudi, Yu. A.

I. N. Smalikho, V. A. Banakh, A. V. Falits, and Yu. A. Rudi, “Estimation of the turbulent energy dissipation rate from the data measured by the “StreamLine” lidar in the surface layer of atmosphere,” Atmos. Oceanic Opt. 28(10), 901–905 (2015).

Rye, B. J.

B. J. Rye and R. M. Hardesty, “Discrete spectral peak estimation in incoherent backscatter heterodyne lidar. I: Spectral accumulation and Cramer-Rao lower bound,” IEEE Trans. Geosci. Remote Sens. 31(1), 16–27 (1993).
[Crossref]

Sathe, A.

A. Sathe, J. Mann, N. Vasiljevic, and G. Lea, “A six-beam method to measure turbulence statistics using ground-based wind lidars,” Atmos. Meas. Tech. 8(2), 729–740 (2015).
[Crossref]

A. Sathe and J. Mann, “A review of turbulence measurements using ground-based wind lidars,” Atmos. Meas. Tech. 6(11), 3147–3167 (2013).
[Crossref]

A. Sathe, R. Banta, L. Pauscher, K. Vogstad, D. Schlipf, and S. Wylie, “Estimating Turbulence Statistics and Parameters from Groundand Nacelle-Based Lidar Measurements:IEA Wind Expert Report,” (2015).

Schlipf, D.

A. Sathe, R. Banta, L. Pauscher, K. Vogstad, D. Schlipf, and S. Wylie, “Estimating Turbulence Statistics and Parameters from Groundand Nacelle-Based Lidar Measurements:IEA Wind Expert Report,” (2015).

Smalikho, I. N.

V. A. Banakh and I. N. Smalikho, “Wind sensing in an atmospheric boundary layer by means of micropulse coherent Doppler lidars,” Opt. Spectrosc. 121(1), 152–159 (2016).
[Crossref]

V. A. Banakh and I. N. Smalikho, “Lidar observations of atmospheric internal waves in the boundary layer of atmosphere on the coast of Lake Baikal,” Atmos. Meas. Tech. 9(10), 5239–5248 (2016).
[Crossref]

I. N. Smalikho, V. A. Banakh, A. V. Falits, and Yu. A. Rudi, “Estimation of the turbulent energy dissipation rate from the data measured by the “StreamLine” lidar in the surface layer of atmosphere,” Atmos. Oceanic Opt. 28(10), 901–905 (2015).

I. N. Smalikho, V. A. Banakh, Y. L. Pichugina, W. A. Brewer, R. M. Banta, J. K. Lundquist, and N. D. Kelley, “Lidar investigation of atmosphere effect on a wind turbine wake,” J. Atmos. Ocean. Technol. 30(11), 2554–2570 (2013).
[Crossref]

I. N. Smalikho and V. A. Banakh, “Accuracy of estimation of the turbulent energy dissipation rate from wind measurements with a conically scanning pulsed coherent Doppler lidar,” Part I. Algorithm of data processing, Atmos. Oceanic Opt. 26(5), 404–410 (2013).
[Crossref]

V. A. Banakh, V. A. Brewer, E. L. Pichugina, and I. N. Smalikho, “Measurements of wind velocity and direction with coherent Doppler lidar in conditions of a weak echo signal,” Atmos. Oceanic Opt. 23(5), 381–388 (2010).
[Crossref]

Vasiljevic, N.

A. Sathe, J. Mann, N. Vasiljevic, and G. Lea, “A six-beam method to measure turbulence statistics using ground-based wind lidars,” Atmos. Meas. Tech. 8(2), 729–740 (2015).
[Crossref]

Vogstad, K.

A. Sathe, R. Banta, L. Pauscher, K. Vogstad, D. Schlipf, and S. Wylie, “Estimating Turbulence Statistics and Parameters from Groundand Nacelle-Based Lidar Measurements:IEA Wind Expert Report,” (2015).

Wang, G.

Wang, X.

Wu, S.

Wylie, S.

A. Sathe, R. Banta, L. Pauscher, K. Vogstad, D. Schlipf, and S. Wylie, “Estimating Turbulence Statistics and Parameters from Groundand Nacelle-Based Lidar Measurements:IEA Wind Expert Report,” (2015).

Yadlowsky, M. J.

R. G. Frehlich and M. J. Yadlowsky, “Performance of mean-frequency estimators for Doppler radar and lidar,” J. Atmos. Ocean. Technol. 11(5), 1217–1230 (1994).
[Crossref]

Yin, J.

Zhai, X.

Zhang, H.

Atmos. Meas. Tech. (3)

A. Sathe and J. Mann, “A review of turbulence measurements using ground-based wind lidars,” Atmos. Meas. Tech. 6(11), 3147–3167 (2013).
[Crossref]

A. Sathe, J. Mann, N. Vasiljevic, and G. Lea, “A six-beam method to measure turbulence statistics using ground-based wind lidars,” Atmos. Meas. Tech. 8(2), 729–740 (2015).
[Crossref]

V. A. Banakh and I. N. Smalikho, “Lidar observations of atmospheric internal waves in the boundary layer of atmosphere on the coast of Lake Baikal,” Atmos. Meas. Tech. 9(10), 5239–5248 (2016).
[Crossref]

Atmos. Oceanic Opt. (2)

V. A. Banakh, V. A. Brewer, E. L. Pichugina, and I. N. Smalikho, “Measurements of wind velocity and direction with coherent Doppler lidar in conditions of a weak echo signal,” Atmos. Oceanic Opt. 23(5), 381–388 (2010).
[Crossref]

I. N. Smalikho, V. A. Banakh, A. V. Falits, and Yu. A. Rudi, “Estimation of the turbulent energy dissipation rate from the data measured by the “StreamLine” lidar in the surface layer of atmosphere,” Atmos. Oceanic Opt. 28(10), 901–905 (2015).

Doklady AN SSSR (1)

A. N. Kolmogorov, “Local structure of turbulence in incompressible viscous fluid at very large Reynolds numbers,” Doklady AN SSSR 30, 299–303 (1941).

IEEE Trans. Geosci. Remote Sens. (1)

B. J. Rye and R. M. Hardesty, “Discrete spectral peak estimation in incoherent backscatter heterodyne lidar. I: Spectral accumulation and Cramer-Rao lower bound,” IEEE Trans. Geosci. Remote Sens. 31(1), 16–27 (1993).
[Crossref]

J. Atmos. Ocean. Technol. (2)

R. G. Frehlich and M. J. Yadlowsky, “Performance of mean-frequency estimators for Doppler radar and lidar,” J. Atmos. Ocean. Technol. 11(5), 1217–1230 (1994).
[Crossref]

I. N. Smalikho, V. A. Banakh, Y. L. Pichugina, W. A. Brewer, R. M. Banta, J. K. Lundquist, and N. D. Kelley, “Lidar investigation of atmosphere effect on a wind turbine wake,” J. Atmos. Ocean. Technol. 30(11), 2554–2570 (2013).
[Crossref]

Opt. Express (1)

Opt. Spectrosc. (1)

V. A. Banakh and I. N. Smalikho, “Wind sensing in an atmospheric boundary layer by means of micropulse coherent Doppler lidars,” Opt. Spectrosc. 121(1), 152–159 (2016).
[Crossref]

Part I. Algorithm of data processing, Atmos. Oceanic Opt. (1)

I. N. Smalikho and V. A. Banakh, “Accuracy of estimation of the turbulent energy dissipation rate from wind measurements with a conically scanning pulsed coherent Doppler lidar,” Part I. Algorithm of data processing, Atmos. Oceanic Opt. 26(5), 404–410 (2013).
[Crossref]

Other (4)

N. L. Byzova, V. N. Ivanov, and E. K. Garger, Turbulence in Atmospheric Boundary Layer (Gidrometeoizdat, 1989) [in Russian]

I. N. Smalikho and V. A. Banakh, “Measurements of wind turbulence parameters by a conically scanning coherent Doppler lidar in the atmospheric boundary layer,” Atmos. Meas. Tech. Discuss., https://doi.org/10.5194/amt-2017-140 (2017).
[Crossref]

V. Banakh and I. Smalikho, Coherent Doppler Wind Lidars in a Turbulent Atmosphere (Artech House, 2013).

A. Sathe, R. Banta, L. Pauscher, K. Vogstad, D. Schlipf, and S. Wylie, “Estimating Turbulence Statistics and Parameters from Groundand Nacelle-Based Lidar Measurements:IEA Wind Expert Report,” (2015).

Cited By

OSA participates in Crossref's Cited-By Linking service. Citing articles from OSA journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (9)

Fig. 1
Fig. 1 Azimuth dependence of the average radial velocity (black curve), radial velocity at the center of the sensing volume moving over the scanning cone base (blue curve), and lidar estimate of the radial velocity (red curve).
Fig. 2
Fig. 2 Difference structure functions ΔD(y) (blue curve), Δ D ¯ max (y) (red curve), and Δ D ¯ cnt (y) (green curve) at ε = 10−3 m2/s3 and L V = 300 m.
Fig. 3
Fig. 3 Relative error in the lidar estimate of dissipation rate as a function of the signal-to-noise ratio for ε = 10−6 (1), 10−5 (2) 10−4 (3), and 10−3 m2/s3 (4).
Fig. 4
Fig. 4 Signal-to-noise ratio as a function of the turbulence energy dissipation rate at E ε ×100% = 16% (solid red curve).
Fig. 5
Fig. 5 Time–height patterns of the wind speed U, vertical component of the wind velocity V z , turbulence energy dissipation rate ε, SNR, relative error of the dissipation rate E ε , and the radial velocity error σ e retrieved from the Stream Line lidar measurements on the coast of the Lake Baikal on August 23, 2015.
Fig. 6
Fig. 6 Vertical profiles of the turbulence energy dissipation rate (a), SNR (b), relative error of the dissipation rate estimate (c), and the radial velocity error (d) retrieved from the lidar measurements at 11:30 (black curve), 12:30 (red curve), and 13:30 (blue curve). Data are taken from Fig. 5. Dashed curves show results of calculation by Eqs. (12) and (6).
Fig. 7
Fig. 7 Vertical profiles of the turbulence energy dissipation rate (a), SNR (b), relative error of the dissipation rate estimate (c), and the radial velocity error (d) retrieved from the lidar measurements at 14:30 (black curve), 15:30 (red curve), and 16:30 (blue curve). Data are taken from Fig. 5. Dashed curves show results of calculation by Eqs. (12) and (6).
Fig. 8
Fig. 8 Results of spatiotemporal visualization of the wind speed (a), wind direction angle (b), and vertical component (c) from measurements by the Stream Line lidar on shore of Lake Baikal on 06.08.2016.
Fig. 9
Fig. 9 Dynamics of the wind speed (a), wind direction angle (b), vertical component of the wind vector (c), turbulence energy dissipation rate (d), and relative error of estimation of the dissipation rate (e) at a height of 300 m as obtained from measurements by the Stream Line lidar on shore of Lake Baikal on 06.08.2016.

Equations (12)

Equations on this page are rendered with MathJax. Learn more.

V ^ r ( R k , θ m ,n)= V ^ r ( R k , θ m ,n) S m V( h k ) ,
D ^ L ( m Δθ)= N 1 n=1 N (M m ) 1 m=1 M m [ V ^ r ( R k , θ m + m Δθ,n) V ^ r ( R k , θ m ,n) ] 2 .
V ^ r ( R k , θ m ,n)= V ¯ r ( R k , θ m ,n)+ V e ( R k , θ m ,n) .
D L ( m Δθ)= D ¯ ( m Δθ)+2 σ e 2 ,
E ε = 3 2 < [Δ D ^ L (y)Δ D ¯ (y)] 2 > Δ D ¯ (y) .
E ε =(3/2) σ D ¯ 2 + σ I 2 ,
σ I 2 = 8 (M M )N ( 1+ σ e 2 Δ D ¯ (y) ) σ e 2 Δ D ¯ (y) .
Δ D ¯ (y)=[A(mΔ y k )A(Δ y k )] ε 2/3 ,
A(z)= 4 3 C k z 2/3 G(z) ,
G(z)= [3 π Γ(5/6)] 1 0 d ξ 1 (1cos ξ 1 ) 0 d ξ 2 ( ξ 1 2 + ξ 2 3 ) 4 3 [ 1+ 8 3 ξ 1 2 ξ 1 2 + ξ 2 2 ]H( ξ 1 , ξ 2 ;z) ,
H( ξ 1 , ξ 2 ;z)=exp[ 2 ( Δp ξ 2 2z ) 2 ]sin c 2 ( Δq ξ 2 2z )sin c 2 ( Δ y k ξ 1 2z ).
σ e =( 1+ 1 SNR ) Δv N a ,

Metrics