THE NERC MST RADAR FACILITY AT ABERYSTWYTH

NERC MSTRF: file format for MST Radar Cardinal data files, version 4.0

processing scheme is the latest version. The v4.0 Cardinal data files described on this page should be used in preference to any data files produced by earlier processing schemes for ST-mode observations.

File contents

The files contain altitude profiles (from approximately 2 - 20 km) of the eastward, northward and upward components of the wind velocity and of the following MST Radar data products: (vertical beam) signal power, (vertical beam) spectral width, (beam-broadening) corrected spectral width, and aspect sensitivity. The time separation between the profiles is typically a few minutes. A radar-derived tropopause altitude and sharpness is given for each set of profiles.

Availability

Files are processed by version 4.0 software are currently available from 1st June 2011 onwards.

File naming convention:
nerc-mstrf-radar-mst_capel-dewi_YYYYMMDD_stRRR_cardinal_SSmin-smoothing_v4-0.nc

YYYY is a 4-digit year
MM is a 2-digit month [01 - 12]
DD is a 2-digit day [01 - 31]
RRR is the range resolution (m) [150 | 300 | 600 | 1200 | 2400 | 4800]
SS is the number of minutes smoothing in time
.nc represents that this is a netCDF file

i.e. nerc-mstrf-radar-mst_capel-dewi_20170327_st300_cardinal_33min-smoothing_v4-0.nc contains Cardinal data for 27th March 2017, st altitude coverage at 300 m range resolution.
Click here for the background to the file naming convention.

File location:
/badc/mst/data/nerc-mstrf-radar-mst/v4-0/st-mode/cardinal/

Instrument catalogue page:
http://catalogue.ceda.ac.uk/uuid/6fcd4e23841b4aa0af75b8bb783011ea

Dataset catalogue page:
http://catalogue.ceda.ac.uk/uuid/86d964d18ac242a58e13bb8d1c849b48

Archiving convention: YYYY/MM
Click here for a further explanation of archiving conventions .

netCDF File Structure using the st300 file for 27th March 2017 as an example - click here for an explanation of netCDF file structure.

Summary of Global attributes - Click on the name to view the value

char Conventions
char title
char description
char source
char institution
char history
char references
char comment
char rights
char license
char acknowledgement
char publisher
char creator_name
char creator_email
char creator_url
char project
char project_principal_investigator
char project_principal_investigator_contact
char platform_name
char platform_type
char platform_catalogue_url
char instrument_name
char instrument_location
char instrument_location_keywords
float instrument_altitude_above_mean_sea_level_m
float instrument_electromagnetic_frequency_Hz
float instrument_beam_one_way_half_power_half_width_degrees
float instrument_antenna_side_length_m
char instrument_manufacturer
char instrument_model
char instrument_serial_number
char instrument_catalogue_url
char calibration_sensitivity
char calibration_certification_date
char calibration_certification_url
char operational_software
char operational_software_version
short observation_year
byte observation_month
byte observation_day
char observation_start_time
char observation_end_time
char observation_altitude_mode
char observation_range_resolution_string
byte observation_range_resolution_number
short observation_bottom_range_gate_number
short observation_top_range_gate_number
char processing_software
char processing_software_version_string
byte processing_software_version_number
byte processing_software_sub_version_number
short processing_nominal_smoothing_period_minutes
char featureType
char data_product
byte data_product_level
char comment_processing_changes
char comment_instrument_location
char comment_instrument_technique
char comment_horizontal_wind
char comment_vertical_beam_variables
char comment_dB_units
char comment_tropopause
char comment_spectral_width
char comment_qc_flag
char comment_qc_details

Summary of Dimensions - Click on the name to view the corresponding coordinate variable

latitude
longitude
time
altitude

Summary of Variables - Click on a name to view the corresponding attributes

float latitude(latitude)
float longitude(longitude)
float time(time)
float altitude(altitude)
float eastward_wind(time, altitude)
float northward_wind(time, altitude)
float upward_wind(time, altitude)
float signal_power(time, altitude)
float aspect_sensitivity(time, altitude)
float corrected_spectral_width(time, altitude)
float tropopause_altitude(time)
byte tropopause_sharpness(time)
float noise_power(time)
byte qc_flag_horizontal_wind(time, altitude)
byte qc_flag_vertical_beam(time, altitude)
short qc_details_vertical_beam(time, altitude)
byte qc_flag_aspect_sensitivity(time, altitude)
byte qc_flag_corrected_spectral_width(time, altitude)
float horizontal_wind_compensation_factor(time, altitude)
byte number_of_cycles_in_smoothing_period(time)
float spectral_width(time, altitude)

List of Global Attribute values

char Conventions = "CF-1.6"

char title = "
  Altitude profiles of Cardinal data products derived from observations
  made by the 46.5 MHz NERC MST Radar at Aberystwyth.
"

char description = "
  This file contains altitude profiles of a variety of atmospheric data
  products related to winds, waves, turbulence, temperature gradients,
  and humidity.
"

char source = "
  The 46.5 MHz Natural Environment Research Council (NERC)
  Mesosphere-Stratosphere-Troposphere (MST) Radar at Capel Dewi (near
  Aberystwyth, UK).
"

char institution = "
  The Natural Environment Research Council (NERC)
  Mesosphere-Stratosphere-Troposphere (MST) Radar Facility at Aberystwyth,
  http://mst.nerc.ac.uk , is funded by NERC through the National Centre
  for Atmospheric Science (NCAS), https://www.ncas.ac.uk . It is managed
  by RAL Space at the Science and Technology Facilities Council's (STFC)
  Rutherford Appleton Laboratory (RAL), https://www.ralspace.stfc.ac.uk .
"

char history = "
  2017-09-11T13:31:42 - netcdf file created on computer jasmin-sci2.ceda.ac.uk .
  2017-09-11T13:53:36 - data passed visual quality control check and file delivered to CEDA for ingestion into archives.
"

char references = "
  The following publications mostly relate to wind data quality and
  representativeness. They are referred to in other global/variable
  attributes.
   Thomas, L., I. Astin, and R. M. Worthington (1997). A statistical
  study of underestimates of wind speeds by VHF radar. Ann. Geophys.,
  15, 805-812. DOI: 10.1007/s00585-997-0805-8.
   Hooper, D. A., and L. Thomas (1998). Complementary criteria for
  identifying regions of intense atmospheric turbulence using lower VHF
  radar. J. Atmos. Sol.-Terr. Phys., 60(1):49-61.
  DOI: 10.1016/S1364-6826(97)00054-0
   Hooper, D. A., and J. Arvelius (2000). Monitoring of the Arctic winter
  tropopause: A comparison of radiosonde, ozonesonde and MST radar
  observations. Proceedings of the Ninth International Workshop on
  Technical and Scientific Aspects of MST Radar, pages 385-388. Sci. Comm.
  on Sol.-Terr. Phys. Secr., Boulder, Colorado, USA.
   Hooper, D. A., J. Nash, T. Oakley, and M. Turp (2008). Validation of a
  new signal processing scheme for the MST radar at Aberystwyth. Ann.
  Geophys., 26(11), 3253-3268. DOI: 10.5194/angeo-26-3253-2008 .
   Parton, Graham, Anthony Dore, and Geraint Vaughan (2010). A
  climatology of mid-tropospheric mesoscale strong wind events as observed
  by the MST radar, Aberystwyth. Meteorol. Apps., 17, 340-354, 2010.
  DOI: 10.1002/met.203 .
   Hooper, David A., David M. Edwards, Gemma Holmes, Kevin Linklater,
  Tim Oakley, Colin Parrett, and Myles Turp (2013). The usefulness of
  model-comparison statistics for wind-profiling radar operators. In
  R. Latteck and W. Singer, editors, Proceedings of the Thirteenth
  International Workshop on Technical and Scientific Aspects of MST
  Radar, pages 141-150. Leibniz-Institute of Atmospheric Physics at the
  Rostock University, Kühlungsborn, Germany. Available from
  http://purl.org/net/epubs/work/11539671 .
   Lee, C. F., G. Vaughan, and D. A. Hooper (2014). Evaluation of wind
  profiles from the NERC MST radar, Aberystwyth, UK. Atmos. Meas. Tech.,
  7(9), 3113-3126. DOI: 10.5194/amt-7-3113-2014 .
"

char comment = "
  Information about the data are spread across a wide range of global
  attributes. See, in particular, those with names starting with
  'comment_'. If there are any comments specific to the data contained
  within the file, they will be given by a global_attribute
  'comment_specific'.
"

char rights = "
  Copyright 2017 Natural Environment Research Council
  (NERC), http://www.nerc.ac.uk . Refer also to the 'license' and
  'acknowledgement' global attributes.
"

char license = "
  Use of the data is covered by the (UK) Open Government Licence,
  http://www.nationalarchives.gov.uk/doc/open-government-licence .
  Refer also to the 'rights' and 'acknowledgement' global attributes.
"

char acknowledgement = "
  The National Centre for Atmospheric Science (NCAS),
  https://www.ncas.ac.uk , must be acknowledged as the data provider
  wherever the data from this file are used. Refer also to the
  'rights' and 'license' global attributes.
"

char publisher = "
  This data file is published by the Centre for Environmental Data Analysis
  (CEDA), http://catalogue.ceda.ac.uk/uuid/86d964d18ac242a58e13bb8d1c849b48 .
"

char creator_name = "David A. Hooper"

char creator_email = "david.hooper@ncas.ac.uk"

char creator_url = "http://mst.nerc.ac.uk"

char project = "
  National Centre for Atmospheric Science (NCAS) long term measurement
  programme at Capel Dewi.
"

char project_principal_investigator = "David A. Hooper"

char project_principal_investigator_contact = "david.hooper@ncas.ac.uk"

char platform_name = "capel-dewi"

char platform_type = "stationary_platform"

char platform_catalogue_url = "http://catalogue.ceda.ac.uk/uuid/8b723580e0e5426d888b273e42f76c1b"

char instrument_name = "nerc-mstrf-radar-mst"

char instrument_location = "52 25 28.2936 N, -4 0 19.6913 E"

char instrument_location_keywords = "Capel Dewi, Aberystwyth, Ceredigion, Wales, UK"

float instrument_altitude_above_mean_sea_level_m = 50.0

float instrument_electromagnetic_frequency_Hz = 46500000.0

float instrument_beam_one_way_half_power_half_width_degrees = 1.5

float instrument_antenna_side_length_m = 104.120003

char instrument_manufacturer = "Rutherford Appleton Laboratory"

char instrument_model = "not applicable"

char instrument_serial_number = "not applicable"

char instrument_catalogue_url = "http://catalogue.ceda.ac.uk/uuid/6fcd4e23841b4aa0af75b8bb783011ea"

char calibration_sensitivity = "
  Although the instrument is not calibrated, measures of the horizontal
  wind data quality are produced on a monthly basis from comparisons
  against model data - see Hooper et al. (2013) in the 'references'
  global attribute.
"

char calibration_certification_date = "not applicable"

char calibration_certification_url = "not applicable"

char operational_software = "mstdaq.c and tools_mst_radar_schedule.py"

char operational_software_version = "not applicable"

short observation_year = 2017

byte observation_month = 3

byte observation_day = 27

char observation_start_time = "2017-03-27T00:03:01"

char observation_end_time = "2017-03-27T23:56:02"

char observation_altitude_mode = "st"

char observation_range_resolution_string = "300"

byte observation_range_resolution_number = 2

short observation_bottom_range_gate_number = 18

short observation_top_range_gate_number = 147

char processing_software = "module_mst_processing_v4_0.py"

char processing_software_version_string = "v4.0"

byte processing_software_version_number = 4

byte processing_software_sub_version_number = 0

short processing_nominal_smoothing_period_minutes = 33

char featureType = "timeSeriesProfile"

char data_product = "cardinal"

byte data_product_level = 3

char comment_processing_changes = "
  The version 4 (v4) processing scheme is closely related to the
  version 3 (v3) scheme described by Hooper et al. (2008) in global
  attribute 'references'. The main difference is that the horizontal
  wind components in the v4 Cardinal files represent time averages over
  a nominal period given by the global attribute
  'processing_nominal_smoothing_period_minutes'. This reduces the
  random measurement error of the single cycle estimates in the v3
  Cartesian files. A second difference is the way in which values are
  calculated for the variable 'horizontal_wind_compensation_factor',
  which corrects for the effects of aspect sensitivity - see Thomas et al.
  (1997), Hooper et al. (2008), and Lee et al. (2014) in the global
  attribute 'references'. It was previously based on the (single cycle)
  ratio of signal powers observed at zenith angles of 4.2° and 6.0°. The
  v4 factor is based on the more-robust ratio of signal powers observed at
  zenith angles of 0.0° and 6.0°. Moreover, the signal powers are averaged
  over a nominal period given by the global attribute
  'processing_nominal_smoothing_period_minutes'. These improved estimates
  of horizontal wind are used to calculate values of the variable
  'corrected_spectral_width'.
"

char comment_instrument_location = "
  The values of the variables 'latitude' and 'longitude' have been
  taken from a Global Positioning System (GPS) receiver that was placed
  close to the centre of the MST radar's antenna array. The coordinates
  should be accurate to within +/- 5 m. The value of global attribute
  'instrument_altitude_above_mean_sea_level_m' has been taken from
  the Ordnance Survey Landranger 135 map for the area. It should be
  accurate to within +/- 5 m.
"

char comment_instrument_technique = "
  The NERC MST Radar is a 46.5 MHz pulsed Doppler radar that is used to
  study the atmosphere. The radar returns are primarily from 'clear-air'
  targets, i.e. structures in atmospheric refractive index that have scale
  sizes of half the radar's wavelength (approximately 3 m) along the beam
  pointing direction. The refractive index depends on humidity (within the
  lowest 10 km of the atmosphere), on air density (within the lowest few
  10s of km), and on free electron density (above 50 km). The radar return
  signals are parameterised by their power, Doppler shift, and spectral
  width. The signal power depends on the vertical gradient of (potential)
  refractive index and so relates to atmospheric structure. The Doppler
  shift depends on the radial component of the wind along the beam
  pointing direction. Observations must be made in the vertical direction
  and at an off-vertical angle in at least two orthogonal azimuths in
  order for the three dimensional wind vector to be derived. The spectral
  width depends on the intensity of any turbulence with the radar
  observation volume. However, the values must be corrected for the
  effects of beam-broadening, which depends on the horizontal wind speed
  and on the radar's beam width - see global attribute
  'comment_spectral_width'.
"

char comment_horizontal_wind = "
  The horizontal wind components, given by the variables 'eastward_wind'
  (v_E) and 'northward_wind' (v_N), have been smoothed in time by a
  nominal amount given by the global attribute
  'processing_nominal_smoothing_period_minutes'. Horizontal wind speeds
  (v_H) and directions can be derived using the following relationships:
   v_H = sqrt(v_E^2 + v_N^2)
   φ_met = 180/pi * atan2(-v_E,-v_N)
   φ_vect = 180/pi * atan2(v_E,v_N)
  where sqrt indicates the square root, ^2 indicates raised to the power
  of 2, φ_met is the meteorological convention wind direction (i.e. from
  which it is blowing), φ_vect is the vector convention wind direction
  (i.e. towards which it is blowing), 180/pi is the conversion factor
  between angles in radians and in degrees, and atan2(y,x) is the four
  quadrant arctangent function. Note that some computing environments,
  such as Microsoft Excel, use an atan2(x,y) function for which the order
  of the arguments must be reversed.
"

char comment_vertical_beam_variables = "
  The values of variables 'noise_power', 'signal_power', 'upward_wind', and
  'spectral_width' are taken from the first vertical beam dwell of each
  observation cycle. The values of the latter 3 are taken from the range
  gates that are closest in altitude to those at 6.0° off-vertical, which
  are used to derive the horizontal wind components. All three variables
  share the same quality control flag variable, 'qc_flag_vertical_beam',
  and quality details variable, 'qc_details_vertical_beam'.
"

char comment_dB_units = "
  Radar return signal powers have dimensions of watts. The values are
  uncalibrated and are stored in dB units, P_dB, rather than linear
  units, P_linear:
   P_dB = 10.0 * log10(P_linear)
  where log10 is the logarithm to base 10. The ratio of signal powers in
  linear units is equivalent to the difference in dB units.
"

char comment_tropopause = "
  The altitude and sharpness of the tropopause - stored by variables
  'tropopause_altitude' and 'tropopause_sharpness' - are derived from
  altitude profiles of the vertical beam signal power following Hooper and
  Arvelius (2000) - see global attribute 'references'. A tropopause
  sharpness value of 0 corresponds to a signal power gradient of less than
  2.0 dB/km, a value of 1 to less than 3.5 dB/km, a value of 2 to less than
  5.0 dB/km, and a value of 3 to greater than 5.0 dB/km. Note that these
  gradients are twice those used by Hooper and Arvelius (2000), which were
  based on signal amplitudes rather than powers. The tropopause details
  contained in the v3 Cartesian files are based on single cycle profiles
  of signal power. The values in the v4.0 Cardinal files are the medians
  of the v3 values taken over a nominal period given by global attribute
  'processing_nominal_smoothing_period_minutes'.
"

char comment_spectral_width = "
  The observed spectral width, σ_obs (m s-1), for MST radar
  observations made in the vertical pointing direction - i.e. given by
  variable 'spectral_width' - is determined both by the spread of
  velocities caused by turbulence, σ_turb (m s-1), and by the beam
  broadening component, σ_beam (m s-1):
   σ_obs^2 = σ_turb^2 + σ_beam^2
  where ^2 represents raised to the power of 2. The beam broadening
  component is determined by the radar's beam width, σ_0, and by the
  horizontal wind speed, v_H (m s-1):
   σ_beam = v_H * sin(σ_0)
  where
   sin^2(σ_0) = sin^2(θ_1whphw) / (4 * ln2)
  and θ_1whphw is the one-way half-power half-width of the beam,
  which is given by the global attribute
  'instrument_beam_one_way_half_power_half_width_degrees', and ln is the
  natural logarithm. The values of σ_turb are stored by the variable
  'corrected_spectral_width'. These are set to 0.0 where the beam
  broadening component exceeds the observed spectral width. The observed
  spectral widths have a finite minimum value and so care should be taken
  when interpreting the corrected values under conditions of small wind
  speeds. In order to convert from corrected spectral width to turbulence
  eddy dissipate rate, ε (W kg-1), it is necessary to know the
  Brunt-Vaisala frequency, ω_B (radians s-1), e.g. from radiosonde data:
   ε = C * σ_turb^2 * ω_B
  where C is a constant of proportionality, which is typically assumed to
  lie between 0.25 and 0.50. Refer to Hooper and Thomas (1998) in the
  'references' global attribute for more details.
"

char comment_qc_flag = "
  For many of the variables in the v4 Cardinal files, an indication of
  which values are considered to be reliable is given by an associated
  quality control flag variable. The name of the latter is given by
  the 'ancillary_variables' attribute of the main variable (and the
  'associated_variables' attribute of a quality control flag variable
  gives the names of the main variables associated with it). A quality
  control flag value of 1 implies that the associated values are
  reliable. A value of 0 is never used and any value higher than 1 implies
  that the associated values are either unreliable or should only
  be used with caution. Refer to the 'flag_values' and 'flag_meanings'
  attributes of each quality control flag variable for specific details.
  In some cases, additional quality control details are given by an
  associated quality control details flag - see global attribute
  'comment_qc_details'.
"

char comment_qc_details = "
  In most circumstances it is best to make use of the associated quality
  control flag variable in order to determine which variable values are
  are considered to be reliable. The 'qc_details_vertical_beam'
  variable contains all of the information, encoded bitwise, used to
  determine the overall reliability. The same technique is used in the
  radial and Cartesian data files and so not all of the following bits
  are used for the vertical beam signal here.
   bit 00: 1 if the signal is available
   bit 01: 1 if the peak smoothed power spectral density (PSD) is greater
   than a set threshold above the noise Power Spectral Density.
   bit 02: 1 if the signal belongs to a radial chain
   bit 03: 1 if the signal fits overall radial continuity
   bit 04: 1 if a secondary signal belongs to a radial chain
   bit 05: 1 if the signal has passed a uni-directional time continuity
   test
   bit 06: 1 if the signal has passed a bi-directional time continuity
   test
  If bit number n has a value of 1, the following condition will be met:
   qc_details_vertical_beam & 2^n == 2^n
  where & represents a bitwise AND function and ^ represents raised to the
  power of 2.
"


Variable attribute values

float latitude(latitude)
standard_name = "latitude"
long_name = "Instrument latitude"
units = "degrees_north"
comment = "
  Refer to global attribute 'comment_instrument_location' for more
  details about this variable.
"
float longitude(longitude)
standard_name = "longitude"
long_name = "Instrument longitude"
units = "degrees_east"
comment = "
  Refer to global attribute 'comment_instrument_location' for more
  details about this variable.
"
float time(time)
standard_name = "time"
long_name = "Start time of observation cycle"
units = "seconds since 2017-03-27 00:00:00 +00:00"
axis = "T"
comment = "
  The values refer to the start of observation cycles. The data
  acquisition computer synchronises its clock using Network Time
  Protocol (NTP).
"
float altitude(altitude)
standard_name = "altitude"
long_name = "Altitude above mean sea level"
units = "m"
axis = "Z"
positive = "up"
comment = "The values refer to the centres of the range gates."
float eastward_wind(time, altitude)
standard_name = "eastward_wind"
long_name = "Eastward wind component"
units = "m s-1"
coordinates = "latitude longitude"
ancillary_variables = "qc_flag_horizontal_wind"
comment = "
  Refer to global attribute 'comment_horizontal_wind' for more details
  about this variable.
"
float northward_wind(time, altitude)
standard_name = "northward_wind"
long_name = "Northward wind component"
units = "m s-1"
coordinates = "latitude longitude"
ancillary_variables = "qc_flag_horizontal_wind"
comment = "
  Refer to global attribute 'comment_horizontal_wind' for more details
  about this variable.
"
float upward_wind(time, altitude)
standard_name = "upward_wind"
long_name = "Upward wind component"
units = "m s-1"
coordinates = "latitude longitude"
ancillary_variables = "qc_flag_vertical_beam"
v3_cartesian_file_variable_name = "vertical_beam_radial_velocity"
comment = "
  The values have an accuracy of the order of 0.1 m s-1, which means
  that they cannot be used to identify small-magnitude synoptic-scale
  vertical velocities. However, they are well-suited for identifying
  mountain/convectively-generated waves and convection. Refer to global
  attribute 'comment_vertical_beam_variables' for more details about
  this variable.
"
float signal_power(time, altitude)
long_name = "Vertical beam signal power"
units = "dB"
coordinates = "latitude longitude"
ancillary_variables = "qc_flag_vertical_beam"
v3_cartesian_file_variable_name = "vertical_beam_signal_power"
comment = "
  Refer to global attributes 'comment_vertical_beam_variables' and
  'comment_dB_units' for more details about this variable.
"
float aspect_sensitivity(time, altitude)
long_name = "Ratio of vertical to 6.0° off-vertical signal powers"
units = "dB"
coordinates = "latitude longitude"
ancillary_variables = "qc_flag_aspect_sensitivity"
comment = "
  The values are based on the signal powers averaged across all
  vertical beam dwells within each observation cycles and those
  averaged across all 6.0° off-vertical dwells. Vertical beam values are
  taken from the range gates that are closest in altitude to those at
  6.0° off-vertical. Refer to Hooper and Thomas (1998) in the
  'references' global attribute for more details. Refer also to
  global attribute 'comment_dB_units'.
"
float corrected_spectral_width(time, altitude)
long_name = "Beam broadening corrected vertical beam spectral width"
units = "m s-1"
coordinates = "latitude longitude"
ancillary_variables = "qc_flag_corrected_spectral_width"
v3_cartesian_file_variable_name = "beam_broadening_corrected_spectral_width"
comment = "
  Refer to global attributes 'comment_vertical_beam_variables' and
  'comment_spectral_width' for more details about this variable.
"
float tropopause_altitude(time)
standard_name = "tropopause_altitude"
long_name = "Radar-derived tropopause altitude"
units = "m"
coordinates = "latitude longitude"
comment = "
  Refer to global attribute 'comment_tropopause' for more details about
  this variable.
"
byte tropopause_sharpness(time)
long_name = "Radar-derived tropopause sharpness index"
units = "1"
coordinates = "latitude longitude"
flag_values = [0, 1, 2, 3]
flag_meanings = "indefinite lower_intermediate upper_intermediate definite"
v3_cartesian_file_variable_name = "tropopause_sharpness_factor"
comment = "
  Refer to global attribute 'comment_tropopause' for more details about
  this variable.
"
float noise_power(time)
long_name = "Vertical beam median spectral noise power over altitude"
units = "dB"
coordinates = "latitude longitude"
v3_cartesian_file_variable_name = "vertical_beam_median_noise_power"
comment = "
  The noise power spectral density is derived for each spectrum using
  a statistical technique - refer to Hooper et al. (2008) in the
  'references' global attribute. The noise power represents this value
  integrated across the width of the spectrum. The value of this variable
  represents the median value over the altitude profile for the first
  vertical beam dwell of each observation cycle. It undergoes a diurnal
  variation of a few dB as a result of variations in sky noise. Refer
  also to global attributes 'comment_vertical_beam_variables' and
  'comment_dB_units'.
"
byte qc_flag_horizontal_wind(time, altitude)
long_name = "Quality control/reliability flag for horizontal wind variables"
units = "1"
coordinates = "latitude longitude"
flag_values = [1, 2]
flag_meanings = "reliable not_reliable"
associated_variables = "eastward_wind northward_wind"
comment = "
  Refer to global attribute 'comment_qc_flag' for more details about
  this variable.
"
byte qc_flag_vertical_beam(time, altitude)
long_name = "Quality control/reliability flag for vertical beam variables"
units = "1"
coordinates = "latitude longitude"
flag_values = [1, 2]
flag_meanings = "reliable not_reliable"
associated_variables = "signal_power upward_wind spectral_width"
comment = "
  Refer to global attribute 'comment_qc_flag' for more details about
  this variable.
"
short qc_details_vertical_beam(time, altitude)
long_name = "Quality control/reliability details for vertical beam variables"
units = "1"
coordinates = "latitude longitude"
comment = "
  Refer to global attribute 'comment_qc_details' for more details about
  this variable.
"
byte qc_flag_aspect_sensitivity(time, altitude)
long_name = "Quality control/reliability flag for aspect sensitivity values"
units = "1"
coordinates = "latitude longitude"
flag_values = [1, 2]
flag_meanings = "reliable unreliable"
associated_variables = "aspect_sensitivity"
comment = "
  Refer to global attribute 'comment_qc_flag' for more details about
  this variable.
"
byte qc_flag_corrected_spectral_width(time, altitude)
long_name = "Quality control/reliability flag for corrected spectral width values"
units = "1"
coordinates = "latitude longitude"
flag_values = [1, 2, 3]
flag_meanings = "reliable overcorrected unreliable"
associated_variables = "corrected_spectral_width"
comment = "
  A flag value of 2 indicates that the beam broadening correction is
  larger than the observed spectral width and so the corrected value
  is set to zero. The associated corrected_spectral_width values are
  still worth plotting. Refer also to global attributes
  'comment_qc_flag' and 'comment_spectral_width'.
"
float horizontal_wind_compensation_factor(time, altitude)
long_name = "Horizontal wind aspect sensitivity compensation factor"
units = "1"
coordinates = "latitude longitude"
v3_cartesian_file_variable_name = "horizontal_wind_theta_s_compensation_factor"
comment = "
  The values of variable 'horizontal_wind_compensation_factor' are
  included for reference purposes only. They have already been applied
  to the values of variables 'eastward_wind' and 'northward_wind'. Refer
  to global attribute 'comment_processing_changes' for more details.
"
byte number_of_cycles_in_smoothing_period(time)
long_name = "
  The number of cycles available in the smoothing period used to derive
  the horizontal wind components.
"
units = "1"
coordinates = "latitude longitude"
comment = "
  Refer to global attribute 'comment_horizontal_wind' for more details
  about this variable.
"
float spectral_width(time, altitude)
long_name = "Vertical beam observed spectral width"
units = "m s-1"
coordinates = "latitude longitude"
ancillary_variables = "qc_flag_vertical_beam"
v3_cartesian_file_variable_name = "vertical_beam_spectral_width"
comment = "
  This variable has been included for reference purposes only. Use
  should instead be made of the 'corrected_spectral_width' variable
  for turbulence information. Refer also to global attributes
  'comment_vertical_beam_variables' and 'comment_spectral_width' for
  more details about this variable.
"