Coherent Doppler Lidar Measurements of the Atmospheric Boundary Layer

1
Coherent Doppler Lidar Measurements of the
Atmospheric Boundary Layer

• R. Frehlich, Y. Meillier, M. Jensen
• University of Colorado, Boulder, CO

2
Acknowledgments

• Army Research Office (Walter Bach)
• NSF (Steve Nelson)
• CTI (Steve Hannon, Jerry Pelk, Mark Vercauteren)
• NREL (Neil Kelley)

3
Boundary Layer Profiles

• Required for atmospheric dynamics and numerical
  model parameterizations
• In situ sensors are required for high-resolution
  profiles of wind speed, temperature, and
  turbulence
• Scanning Doppler lidar provides spatially
  averaged profiles over a 3D measurement volume
• Optimal lidar design and processing algorithms

4
CIRES Tethered Lifting System (TLS)Hi-Tech
Kites or Aerodynamic Blimps
5
High Resolution Profiles from TLS Data

• TLS instrumentation required for accurate in situ
  turbulence profiles
• Hot-wire sensor for small scale velocity
• Cold-wire sensor for small scale temperature

6
Lafayette Campaign
7
CTI LIDAR VELOCITY MAP at 1o

• Radial velocity
• 100 range-gates along beam
• 180 beams (0.5 degree)
• 18 seconds per scan (Taylors frozen hypothesis is
  valid)

in situ prediction
8
Wind Speed and Direction

• Best-fit lidar radial velocity for wind speed and
  direction
• Spatial average reduces estimation error
• Smaller angular regions produce useful fits
• Fluctuations around best-fit are defined to be
  turbulence

in situ prediction
9
Turbulence Estimates at 0o

• Structure function of radial velocity in range a)
• Best fit produces estimates of ?u, ?u, L0u
• Structure function in azimuth b)
• Best fit produces estimates of ?v, ?v, L0v

in situ prediction
tropprof04
10
Turbulence Estimates H80 m

• Best fit for noise corrected structure function
  (o)
• Raw structure functions ()
• Radial velocity has small elevation angles (lt
  4o)
• Structure function in azimuth b)
• Best agreement in ?u and ?v (isotropy)

in situ prediction
tropprof04
11
Lidar and TLS Profiles
12
Large Turbulent Length Scale

• Difficult to separate turbulence and larger scale
  processes
• Similar to stable troposphere
• What is optimal methodology?

13
Filter Radial Velocity in Azimuth

• Raw data has large L0
• Filtered data has well defined length scale
• Good agreement in ?v
• Variance ?v2 is reduced by filtering

in situ prediction
tropprof04
14
Profiles with Filtered Data

• Filtered data has small effect on e
• Largest effect on L0
• Good agreement at low altitudes when L0 is small

in situ prediction
tropprof04
15
CU/NREL Wind Energy Research

• High turbulence conditions
• Height variation of ?, ?, and L0
• Ideal input for optimal wind farm operation

in situ prediction
tropprof04
16
Summary

• In situ (TLS) measurements are required for
  resolving small scale features and identifying
  spatial variability
• Doppler lidar produces high quality volume
  averaged measurements
• Small scale turbulence information (?) is most
  robust
• Simultaneous profiles of longitudinal and
  transverse velocity statistics is feasible
• More optimization is required especially for low
  turbulence conditions

17
Velocity Turbulence

• Along-stream velocity u(t)
• Spectrum Su(f)
• Taylors frozen hypothesis
• Energy dissipation rate ?

18
Temperature Turbulence

• Along-stream temperature T(t)
• Spectrum ST(f)
• Taylors frozen hypothesis
• Temperature structure constant CT2

19
LIDAR, SODAR AND TLS (BLIMP)
20
LIDAR AND TLS (BLIMP)
21
Low Turbulence Data

• Accurate corrections required
• Correct model for spatial statistics

e 5.0 10-6 m2/s3 ?v 0.090 m/s L0 140 m
22
LOW TURBULENCE
in situ prediction
23
LAFAYETTE DATA
24
CONVECTION
in situ prediction
25
SUMMARY

• In situ (TLS) and volume averaged (lidar)
  measurements are complementary
• Accurate lidar derived turbulence measurements
  are feasible
• More optimization is required especially for low
  turbulence
• High resolution TLS data essential
• Spatial statistics of the turbulent fields are
  not well known

26
SUMMARY

• In situ (TLS) and volume averaged (lidar)
  measurements are complementary
• Accurate lidar derived turbulence measurements
  are feasible
• More optimization is required especially for low
  turbulence
• High resolution TLS data essential
• Spatial statistics of the turbulent fields
  require more research, especially for stable
  conditions

27
DAILY SUMMARY PLOTS

• Zero elevation angle scan
• Typical daily pattern (low turbulence at night)
• Wind speed, direction, and ? are critical
  parameters for TD
• Low wind speeds can have erratic direction

in situ prediction
28
Velocity Turbulence

• Along-stream velocity u(t)
• Spectrum Su(f)
• Taylors frozen hypothesis
• Energy dissipation rate ?

29
Temperature Turbulence

• Along-stream temperature T(t)
• Spectrum ST(f)
• Taylors frozen hypothesis
• Temperature structure constant CT2

30
WIND PROFILES

• Urban environment
• Operational real-time profile for low altitudes

31
WIND PROFILES

• Larger domain
• Operational real-time profile for data
  assimilation

32
LAFAYETTE DATA
33
TURBULENCE STATISTICS

• Magnitude of turbulence is the standard deviation
  ?
• Spatial size of fluctuations is the length scale
  L0
• Magnitude of the small-scale fluctuations is ?

in situ prediction