Climatology of Superrefraction Observed by GPS Radio Occultation

1
Climatology of Superrefraction Observed by GPS
Radio Occultation

• Dione (Dee) Lee Rossiter
• University of California, Berkeley
• COSMIC-UCAR

2
Outline

• Background
• GPS Radio Occultation
• The Problem Superrefraction
• Motivation
• GPS Radio Occultation
• Current Research
• Proposed Questions
• Procedure
• Generating Plots
• Finding Reoccurring Features
• Conclusions

3
Background Global Positioning System (GPS)
Satellites
Low-Earth Orbit (LEO) Satellites

• Equipped with a GPS receiver, a LEO can track GPS
  radio signals
• These GPS signals are required to pass through
  the atmosphere where they are refracted

GPS Satellite
LEO Orbit
Atmosphere
Radio Signal
LEO Satellite
4
Background Radio Occultation (RO)

• GPS signals are refracted by the Earths
  atmosphere as they travel to a receiver in LEO.
• Refractivity is a function of
• - electron density in the
  ionosphere
• - temperature, pressure, and water
  vapor in the stratosphere and
  troposphere

5
Background Radio Occultation (RO)
From Anthes et al., 2003

• Phase and amplitude
• ?Doppler shift
• positions and velocities
• ?Bending angles, a, as a function of impact
  parameter, a

?Profiles of refractivity vs. altitude
?Profiles of atmospheric properties vs.
altitude
6
Motivation GPS RO
Meteorology

• Provide global coverage in time and space
• Provide fundamentally unbiased atmospheric
  measurements
• The technique is mission independent
• Provide advancements in space weather research
• GPS RO will provide high quality soundings at a
  low cost!

Climate
Ionosphere
7
The Problem Superrefraction

• Caused by a sharp decrease in refractivity with
  an increase of altitude

• The radius of curvature, rc, becomes smaller than
  the radius of the atmosphere, ra

• Ray remains trapped in the atmosphere and results
  in a temporary extinction of the radio signal
  reaching the LEO

• Commonly occurs at the top of the Planetary
  Boundary Layer (PBL) at 2 km

8
BackgroundRadio Occultation (RO)
From Anthes et al., 2003

• Phase and amplitude
• ?Doppler shift
• positions and velocities
• ? Bending angles a as a function of a

?Profiles of refractivity vs. altitude
9
Abel Transform

• The Abel inversion integrates from the top of the
  atmosphere to the surface
• The Abel inversion works great up until the
  superrefraction layer

10
The Problem Superrefraction
True Retrieved
True Retrieved
True Retrieved
From Sokolovskiy, 2003
Below the superrefraction layer the function

• Results in a negative N bias
• Becomes multi-valued and therefore invalid

11
Motivation Current Research

• GPS RO will soon be utilized in an array of
  atmospheric models (weather, climate,
  ionospheric composition)
• Attaining the highest level of accuracy is
  essential!
• Problem Superrefraction causes errors!
• Understand ? Predict ? Remove Errors

12
Proposed Questions

• What are the reoccurring features of
  superrefractions?
• Annual?
• Seasonal?
• Where does it occur?
• Geographically?
• Topographically?
• Can we predict where or when it might occur or
  recognize when it is occurring in order to
  truncate the sounding or to remove the negative
  bias altogether?
• (Understand ? Predict ? Remove Errors)

13
Generating Plots

• Divide the globe by 15 latitudinally

14
Generating Plots

• Divide the globe by 15 latitudinally

• Take latitude band and divide up by 19 lon.
  bins

• Combine Challenging Mini-Satellite Payload for
  Geophysical Research and Application (CHAMP)
  occultations within bin

15
Generating Plots

• Divide the globe by 15 latitudinally

• Take latitude band and divide up by 19 lon
  bins

• Combine Challenging Mini-Satellite Payload for
  Geophysical Research and Application (CHAMP)
  occultations within bin

• Create refractivity vs. altitude stats for
  each lat lon grid against European Center for
  Medium Range Weather Forecast (ECMWF) model

• Color code the different mean bias
  interpolate across entire latitude band

16
June-Aug/75º to 90º lat
-90
17
June-Aug/0º to 15º lat
2003
18
Dec-Feb/-75º to -60º lat
2003
19
Dec-Feb/-30º to -15º lat
2003
20
Finding Reoccurring Features
Dec-Feb/-30º to -15º lat
2001
2002
2003
Altitude (km)
Altitude (km)
Altitude (km)
Longitude
Longitude
Longitude
21
Scatter Plot

• Measures individual occultations bias at 0.5 km
  that did not agree within 4 of ECMWF model
• Color codes the different bias

22
Scatter Plots

• Midlatitude mostly effected
• Land vs. ocean
• High occurrence in some areas

Dec-Feb 2003

• Greater neg. bias in some areas
• Less positive bias

23
Soden Bretherton (1994)

• Compare ECMWF with Special Sensor
  Microwave/Imager (SSM/I)
• Conclude the model had problems in predicting the
  dry subtropical ridges off the west coast of
  continents where marine stratocumulus clouds
  often occur
• This is where dry air sinks from above and
  creates a sharp vertical gradient in water vapor
  near the top of the PBL
• conditions needed for superrefraction to occur!!!

24
Finding Trends
Latitude
CHAMP-ECMWF June-Aug 2003
(10)-2
Longitude
25
Finding Trends
Latitude
(10)-2
CHAMP-ECMWF Dec-Jan 2003
Longitude
26
Conclusions

• There are seasonal and geographic features
  associated with superrefraction
• Geographically, superrefraction occurs off the
  west coast of continents
• We can possibly use the superrefraction
  climatology found in this research to predict
  where and when to truncate soundings or correct
  or negative bias
• Future Research Diurnal cycles in the PBL

27
Acknowledgements

• Research Mentors
• Bill Kuo
• Bill Schreiner
• Chris Rocken
• Writing Mentor
• Chris Halvorson
• Doug Hunt
• Karl Hudnut
• Kim Prinzi-Kimbro
• All of the COSMIC project office
• All of the SOARS staff and protégés

28
Questions?