ROB NEWSOM1, ROBERT BANTA2, YELENA PICHUGINA1

1
Observations of the Stable Boundary Layer and
Retrieval of Microscale Flow Structure from Lidar
Data

• ROB NEWSOM1, ROBERT BANTA2, YELENA PICHUGINA1
• 1CG/AR, CIRA, Colorado State University, Fort
  Collins, CO
• 2Environmental Technology Laboratory/NOAA,
  Boulder, CO

2
Doppler Lidar Data Analysis

• Objectives
• Develop methods to extract information from
  Doppler lidar data
• Velocity variance, wavenumber spectra, etc
• 3D wind and thermodynamic retrieval
• Scientific objectives
• Turbulent transport in the SBL
• Parameterizations for NWP
• Dispersion in an urban setting

3
Doppler Lidar Data Analysis Retrieval of
Microscale Flow Structures

• The Problem
• Doppler lidars measure radial velocity and
  aerosol backscatter no thermodynamics
• What flow field produced the observed radial
  velocity
• field?

4

• Approach (4DVAR)
• Assume the motion equations (model) are known.
• Fit the model output to the available
  observations using the adjoint of the model
• Initial conditions are adjustable parameters
• Initial conditions are iteratively adjusted to
  minimize the discrepancy between the model and
  the lidar data
• Analysis produces complete dynamics (4D wind and
  thermodynamic field)

5

• The Model
• Simulates dry shallow incompressible flow under
  the Boussinesq approximation
• Momentum
• Heat

6

• The Model
• Simulates dry shallow incompressible flow under
  the Bousinesq approximation

7
The Cost Function Measure of the error between
the model and the observations

• Essentially the mean squared difference between
  the model and the observations
• Radial velocity data is interpolated to the model
  grid
• Does not consider measurement error (or error
  covariance)

8

• The Cost Function
• Measure of the error between the model and the
  observations.

From the lidar
From the model
Measurement error
Observed Radial Velocity
9
Constraints (Lagrangian approach)
10
Derivation of the Adjoint Equations
Construct the Lagrangian
Weak constraint
Strong constraint
11
Derivation of the Adjoint Equations
Assuming
12
Derivation of the Adjoint Equations
Put it all together and impose the constraints
The Adjoint of the Model
The Model
Left overs
The answer
13
The Adjoint Equations
14
The Retrieval Algorithm
Exit if L is minimized
15

• Issues
• How good are the retrievals?
• What are the effects of
• Boundary conditions
• Turbulence parameterizations
• Data resolution vs model resolution
• Data preprocessing (objective analysis)
• Measurement errors

16
Application to Simulated Data convergence tests
17
Application of the Retrieval Algorithm
Volume scan of radial velocity CBL, CASES-99, 25
October, 1999
18
Application of the Retrieval Algorithm
Observed Radial Velocity
Retrieved Perturbation Potential Temperature
Retrieved Perturbation velocity
19
Application to Real Data Retrieval
20
Application of the Retrieval Algorithm
21
Application of the Retrieval Algorithm
Horizontal cross section of retrieved
perturbation velocity (Courtesy of Ching-Long Lin
and Tianfeng Chai, Univ. Iowa)
22
Application of the Retrieval Algorithm Comparison
with tower data
Vertical velocities from 60-m tower
Vertical velocities from 60-m tower
23

• Data Preprocessing and Error Analysis
• Measurement error covariance has not been
  considered
• If the measurements are ingested as is then
• No need to interpolate measurements to the model
  grid. Rather,
• Model output is interpolated to data coordinates.
• Error covariance matrix is diagonal.
• Cost function takes a simple form.

24

• Data Ingest and Error Analysis
• Careful error analysis must be incorporated.
• Objective analyses performed in preprocessing
  input data correlates measurement error.
• Better to ingest the data as is

25

• Error Analysis
• Formal definition of the cost function
• If the measurement errors are uncorrelated then
  E is diagonal.
• If the radial velocity measurements are used as
  is E is diagonal.

26

• Error Analysis and Data Ingest
• Doppler Lidar Velocity precision
• Reformulate the cost function

27

• Error Analysis and Data Ingest
• The new cost function
• Model output is evaluated at the data coordinates
• Then

28

• Deliverables
• Publications (see supplementary slide)
• Data set available at http//www.joss.ucar.edu/ca
  ses99/
• Web site (including lidar data products)
  http//www.etl.noaa.gov/et2/projects/cases/
• DoD Contacts
• 15th Symposium on Boundary Layers and Turbulence,
  Wageningen, The Netherlands, July 02
• Mesoscale Data Integration Workshop, Univ. N.
  Dakota, Sept. 02
• Dr. David Ligon (ARL)

29
Publications in 02
Chai, T., C. L. Lin, R. K. Newsom 2003 Retrieval
of Microscale Flow Structures from High
Resolution Doppler Lidar using and Adjoint
Model.. J. Atmos. Sci., submitted. Drobinski, P.,
P. Carlotti, R. Newsom, R. Foster, R. Banta, J.
Redelsperger, 2003 Review of near-surface flow
dynamics in the neutral planetary boundary-layer
from a CASES99 case study. J. Atmos. Sci.,
submitted. Newsom R.K. and R.M. Banta, 2002
Shear flow instability in the stable nocturnal
boundary layer as observed by Doppler lidar
during CASES-99. J. Atmos. Sci., 60, 16-33. Banta
R.M., R.K. Newsom, J. K. Lundquist, Y. L.
Pichugina, R. L. Coulter, and L. D. Mahrt, 2002
Nocturnal low-level jet characteristics over
Kansas during CASES-99. Boundary-Layer Meteor,
105, 221-252. Poulos, G.S., W. Blumen, , D.C.
Fritts, J.K. Lundquist , J. Sun, S. Burns, C.
Nappo, R.M. Banta, R.K. Newsom, J. Cuxart, E.
Terradellas, B. Balsley, M. Jensen, 2002
CASES-99 A comprehensive investigation of the
stable nocturnal boundary layer. Bull. Amer.
Meteorol. Soc., 83, 555-581. Sun J., D. H.
Lenschow, S. P. Burns, R.M. Banta, R.K. Newsom,
R. L. Coulter, S. Frasier, T. Ince, C. Nappo, B.
Balsley, M. Jensen, D. Miller, B. Skelly, J.
Cuxart, W. Blumen, X. Lee, and X. Z. Hu, 2002
Intermittent turbulence in stable boundary layers
and its relationship with density currents.
Boundary-Layer Meteor., 105, 199-219. Sun J., D.
H. Lenschow, S. P. Burns, R.M. Banta, R.K.
Newsom, R. L. Coulter, S. Frasier, T. Ince, C.
Nappo, B. Balsley, M. Jensen, D. Miller, B.
Skelly, J. Cuxart, W. Blumen, X. Lee, and X. Z.
Hu, 2002 Intermittent turbulence in stable
boundary layers and the processes that generate
it. Boundary-Layer Meteor., submitted. Banta R.
M., R. K. Newsom, Y. L. Pichugina , J. K.
Lundquist, 2002 Nocturnal LLJ evolution and its
relationship to turbulence and fluxes. 15th
Symposium on Boundary Layers and Turbulence,
Wageningen, The Netherlands, 15-19 July. Amer.
Meteor. Soc., Boston. Drobinski, P., R. K.
Newsom, P. Naveau, R. M. Banta, P. Carlotti, R.
C. Foster, 2002 Turbulence in a shear-driven
surface layer during the CASES-99 experiment,
15th Symposium on Boundary Layers and Turbulence,
Wageningen, The Netherlands, 15-19 July. Amer.
Meteor. Soc., Boston. Drobinski P., R.K. Newsom,
R.M. Banta, P. Carlotti, R.C. Foster, P. Naveau
and J.L. Redelsperger, 2002 Turbulence in a
shear-driven nocturnal surface layer as observed
by Doppler lidar, rawinsondes and sonic
anemometer during the CASES99 experiment. 21th
International Laser Radar Conference, Montreal,
Quebec, 8-12 July. Nappo C. J., R. K. Newsom, R.
M. Banta, 2002 Analysis techniques for
boundary-layer atmospheric gravity waves. 15th
Symposium on Boundary Layers and Turbulence,
Wageningen, The Netherlands, 15-19 July. Amer.
Meteor. Soc., Boston. Newsom R. K., R. M. Banta,
Y. L. Pichugina, 2002 Formation, evolution and
decay of a shear flow instability in the stable
nocturnal boundary layer. 15th Symposium on
Boundary Layers and Turbulence, Wageningen, The
Netherlands, 15-19 July. Amer. Meteor. Soc.,
Boston. Newsom R. K., R. M. Banta, 2002
Sensitivity of wind and temperature retrievals
from 4DVAR to prescribed eddy viscosity profiles.
15th Symposium on Boundary Layers and Turbulence,
Wageningen, The Netherlands, 15-19 July. Amer.
Meteor. Soc., Boston.
30

• Future Plans and Deliverables
• Stable Boundary Layer Analysis of CASES-99
  dataset
• Factors effecting the development and
  characteristics of the LLJ Gully surges
  low-speed steaks.
• Retrieval of microscale flow structure
• Efficient implementation of new data ingest
  scheme error analysis
• Use adjoint retrieval algorithm as a tool for
  interpreting Doppler lidar data (CASES-99, VTMX,
  Urban 03 ?, )
• Dual Doppler validation (Urban 03 ?)
• Investigate the potential of ensemble methods
• Conduct turbulence research using ABLE
  instrumentation.