Observation operator for weather-radar refractivity

1
Observation operator for weather-radar
refractivity Olivier Caumont1, Lucas Besson2,
Laurent Goulet3, Sophie Bastin2, Jacques Parent
du Châtelet2,4, Laurent Menut5, Frédéric
Fabry6 1 CNRM-GAME (Météo-France, CNRS) 2
LATMOS 3 DIRSE (Météo-France) 4 Observing
Systems Department (Météo-France) 5 LMD 6
McGill University IODA-MED meeting 16 May 2014
2
IODA-MED deliverables
No update since last years meeting
Talk by Clotilde Augros
3
What is refractivity?

• Refractivity N (n-1) x 106, where n index of
  refraction of air.
• Refractivity may be expressed as (Smith and
  Weintraub 1953)

P pressure (hPa) e partial pressure of water
vapour (hPa) T temperature (K)

• Refractivity mainly depends on moisture when
  temperature is high (at constant pressure)
• 1 N unit 1 relative humidity at 20C
• At constant pressure
• High N moist and/or cold
• Low N dry and/or warm

(Fabry et al. 1997)
4
Principle of refractivity measurement by weather
radar

• Measurement by radar based on radar pulses
  propagation time through the atmosphere, which
  depends on refractivity.
• Phase change between radar and target or between
  2 targets depends on refractivity averaged over
  radar ray path (Fabry et al. 1997), i.e. less
  than a few hundred metres above ground.
• In practice, measurement of time phase change.
  Need for initial values, usually interpolated
  from automatic weather stations (AWSs) in
  homogeneous situation.
• Technique initially for klystron (
  stable-frequency) transmitters. Adaptation for
  magnetron ( drifting-frequency) transmitters
  (Parent du Châtelet et al. 2012).

5
Summary of endeavour related to radar refractivity

• Work done so far
• Formulation for magnetron transmitters (Parent du
  Châtelet et al. 2012)
• Link between refractivity and atmospheric
  phenomena (Besson et al. 2012)
• Technical proposals for improved-quality
  refractivity retrievals (Besson and Parent du
  Châtelet 2013)
• Definition of quality index for target selection
• Investigation of the use of faster antenna
  rotation speeds, additional elevations and
  dual-polarization returns
• Observation operator for refractivity (Caumont et
  al. 2013)
• Sensitivity study to formulation of observation
  operator
• Long-term comparisons of radar observations vs.
  Arome
• Comparison of radar refractivity with automatic
  weather stations and numerical simulations during
  HyMeX SOP1 (Besson et al., in prep. for HyMeX
  special issue)
• Use of refractivity retrievals produced in real
  time during HyMeX SOP1
• Cross-validation with independent observations
  and models
• First attempt to relate real refractivity data
  with Mediterranean meteorological processes

6
Available observations

• 3 operational radars
• Nîmes, Bollène, Opoul
• 7 automatic weather stations (AWS)
• Nîmes-Garons, Nîmes-Courbessac, Tarascon (Nîmes
  radar)
• Visan (Bollène radar)
• Perpignan, Leucate, Durban-Corbière (Opoul radar)

7
Available models

• WRF
• Initial boundary conditions nudging from NCEP
  global model
• 2 nested domains 54- and 9-km horizontal
  resolutions
• N at 2 m AGL from innermost domain
• AROME-WMED
• Initial boundary conditions Arpege global
  model
• Horizontal resolution 2.5 km
• 3-h forecasts from a 3DVar assimilation cycle
• N at 10 m AGL

D-1, 00 UTC
Date, Time
D-1, 12 UTC
D3, 18 UTC
WRF simulation
NCEP analysis
NCEP forecasts
8
Refractivity time series _at_ Nîmes-Courbessac
8 August 30 November 2012
High correlation coefficients between radar
refractivity and other data Radar vs AWS
0.89 Radar vs Arome-WMED analysis 0.90 Radar vs
Arome-WMED forecast 0.84 Radar vs WRF analysis
0.83 Radar vs WRF forecast 0.79
Similar results at other AWS locations
Large differences at times - between WRF and
other data on 18, 19, and 20 October - diurnal
cycle poorly simulated on 8, 9, and 10 September
(needs further investigation)
9
IOP6 (24 September 2012) Time series
1 Convection in the vicinity of Bollène -
precipitation - humidity increases while
temperature decreases - refractivity
increases 2 Convection in the vicinity of
Nîmes - precipitation - humidity alreeady
close to 100 - refractivity remains
constant 3 Convection in the vicinity of
Bollène - precipitation - humidity already
close to 100 - refractivity remains
constant 4 Front passage - humidity decreases
markedly - refractivity decreases markedly
1
4
3
2
4
4
2
4
10
IOP6 (24 September 2012) Front passage
Refractivity from Nîmes and Bollène radars
Front passage
11
IOP6 (24 September 2012) Radars vs. models
Good agreement between Nîmes radar and
models Less agreement between Bollène radar and
models - correct magnitude near the radar -
large discrepancy at far range Large
discrepancies probably caused by mountains
(Massif Central to the west and Alps to the east)
which have a double impact on radar
retrievals - lower-quality targets -
calibration of retrieval algorithm
12
On-going and future activities

• On-going work
• Investigate the relationship with near-ground
  turbulence (PhD thesis of R. Hallali _at_ LATMOS
  off HyMeX),
• Improve calibration
• Perspectives
• Further assessment of usefulness in process
  studies (cold pool, valley effects, breeze,
  low-level flow feeding HPEs, etc.)
• Model validation in AWS-sparse areas
• Data assimilation (coordinate with
  ZAMG/University of Vienna effort to assimilate 3D
  GPS-tomography refractivity data?)

13
References

• Besson, L., J. Parent du Châtelet, 2013
  Solutions for improving the radar refractivity
  measurement by taking operational constraints
  into account. J. Atmos. Oceanic Technol., 30,
  17301742. DOI 10.1175/JTECH-D-12-00167.1
• Besson, L., C. Boudjabi, O. Caumont, J. Parent du
  Châtelet, 2012 Links between weather phenomena
  and characteristics of refractivity measured by
  precipitation radar. Bound.-Lay. Meteor., 143(1),
  7795, DOI 10.1007/s10546-011-9656-7.
• Besson, L. et al. Comparison of refractivity
  measurement by radar with automatic weather
  stations, AROME-WMED and WRF forecasts
  simulations during the SOP1 of HyMeX campaign. In
  prep. for HyMeX special issue of QJRMS.
• Caumont, O., A. Foray, L. Besson, J. Parent du
  Châtelet, 2013 A radar refractivity change
  observation operator for convective-scale models
  Comparison of observations and simulations.
  Bound.-Lay. Meteorol., 148(2), 379397, DOI
  10.1007/s10546-013-9820-3.
• Parent du Châtelet, J., C. Boudjabi, L. Besson,
  O. Caumont, 2012 Errors caused by long-term
  drifts of magnetron frequencies for refractivity
  measurement with a radar Theoretical formulation
  and initial validation. J. Atmos. Oceanic
  Technol., 29(10), 14281434, DOI
  10.1175/JTECH-D-12-00070.1.

14
Thank you for your attention!