2nd Low-Latitude Ionospheric Sensor Network Workshop São José dos Campos 7-10 November 2011 Charles S. Carrano, Cesar E. Valladares, Keith M. Groves Boston College, Chestnut Hill, MA, 02467 Deducing Ionospheric Turbulence Parameters from High-Rate GPS Observations during the COPEX Campaign
21
Embed
Charles S. Carrano, Cesar E. Valladares, Keith M. Groves Boston College, Chestnut Hill, MA, 02467
Deducing Ionospheric Turbulence Parameters from High-Rate GPS Observations during the COPEX Campaign. Charles S. Carrano, Cesar E. Valladares, Keith M. Groves Boston College, Chestnut Hill, MA, 02467. - PowerPoint PPT Presentation
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Propagation of radio wave through scattering layer
produces a diffraction pattern on the ground
4
Our interest is to infer characteristics of scattering layer from the time series of intensity
fluctuations
We assume the plasma turbulence can be represented by a
homogenous scattering layer of thickness L
Ionospheric Turbulence Model (Rino, 1979)
5
( 1/2)2 20
( ) sN
Cqq q
121
( 1)/2 02 2
0
( )1000( ) sec2 2 2 [( 1) / 2]
pp
pe k
K qR r GC L
q p
wavelength
re
classical electron radius
spatial separation distance
Propagation (zenith) angle
p phase spectral index (p=2)
Step 1: apply coordinate transformation to account for irregularity anisotropy
Step 2: integrate through the scattering layer along the line of sight
Step 3: apply Fourier Transform in the transverse plane
• Power law model for 3D spectrum of electron density fluctuations, N:
CkL vertically integrated turbulent intensity
q, q0
wavenumber, outer scale wavenumber
K, modified Bessel function, Gamma function
G geometry enhancement factor
A,B,C propagation and B-field geometry factors
• Correlation function of phase fluctuations, R
(), after passage through layer:
The Model Intensity Spectrum (Booker and MajidiAhi, 1981)
6
1/2sec2zF
2 2( , ) 2 ( ) ( ) ( )f k R R kF R kF
( , ) exp (0, ) ( , ) exp (0, )g k f k f k f k
24
0
1 ( )2
S I k dk
• The S4 index is calculated by integrating spectrum over all wavenumbers:
0( ) 2 ( , )cos( )I k g k k d
Define Fresnel parameter:
and two functions that depend on F and the phase correlation function R
():
• The intensity spectrum at the reception plane is given by:
• A solution of the 4th moment equation governing intensity fluctuations following propagation of a plane wave through a thin phase-changing screen.
Space-to-Time Translation (Rino, 1979)
7
1/22 2
2 / 4sx sx sy sy
eff
CV BV V AVv
AC B
tan( )cos( )
tan( )cos( )
sx px pz
sy D py pz
V V V
V V V V
where A,B,C are propagation and magnetic field geometry factors
Vpx
, Vpy
, Vpz
are the geomagnetic north, east and down coordinates of the IPP velocity,
and VD
is the irregularity zonal drift velocity.
• Effective scan velocity (Rino, 1979):
• The scan velocity (Vsx
,Vsy
) of the ray path through the irregularities is given by:
2effkf v
( ) ( ) / effI f I k v
• Spatial wavenumbers translate to temporal frequencies in a model-dependent fashion:
and
Iterative Parameter Estimation (IPE)
8
max
min
22
max min
2 log ( ; , , ) log ( )f
k D mf
I f C L p V I f dff f
• Define a metric to quantify difference between model and measured intensity spectra:
• The independent variables in the ionospheric turbulence model are CkL, p, and V
D
• Solve for 3 unknown parameters iteratively using the Downhill Simplex Method
Integration performed over frequency range fmin
and fmax
to exclude receiver noise
All others are calculable from geometry and B field direction except the outer scale, anisotropy ratio, and layer height--we assume L0
= 10 km, a : b
= 50, and Hp
=350 km.
Two Examples of IPE Analysis
10
Weak Scatter Example Strong Scatter Example
High-Rate GPS Receivers Operating During COPEX
3
• Three sites located on nearly the same magnetic field
line
• Alta Floresta at dip equator
• Boa Vista and Campo Grande at magnetically
conjugate points, close to crests of equatorial
anomaly
• Three high-rate (10 Hz)
Ashtech Z-CGRS receivers operated by AFRL /
INPE
• Data collected Oct-Dec 2002 under solar max
conditions.
• We present results for the evening of 1-2 Nov 2002
Directly Measured Parameters
11
b)
Vertical TEC
Scintillation Index Decorrelation Time
TECU
S4m
m
(sec)
Measured and Modeled Scintillation Statistics
12
Measured Scintillation Index
Mod
el S
cint
illat
ion
Inde
x
Measured Decorrelation Time (sec)
Mod
el D
ecor
rela
tion
Tim
e (s
ec)
IPE results accurately reproduce …
S4 (depth of fading)
decorrelation time, (rate of fading)
This is makes sense, since S4
and can
be inferred from the intensity spectrum
Previous modeling efforts have tried to reproduce either S4
, or both S4
and . IPE reproduces the entire intensity spectrum
Parameters Inferred by IPE Analysis
13
a)
b)
TECU
p VD
(m/s)
Log(CkL)
Vertical TEC Turbulent Intensity
Phase Spectral Index Drift Velocity
Phase Scintillation Parameters inferred from IPE Analysis
14
a)a) 1
2 2 11
2 ( / 2)sec1000 (2 ) [( 1) / 2]
pp
e k effppT r G C L vp
11
1 02 2 2 221000 1
2
( )sec
4 ( )
ppp
e k p
qr G C L
Phase Spectral Strength Phase Variance
2(rad2)T (dB)
• Spectrum of phase fluctuations:
• Variance of phase fluctuations:
/22 20
( ) ,pTP f
f f
Local Time Variation at the Three COPEX Stations
15
b)
Turbulent Intensity
Phase Spectral Index Phase Spectral Strength
Comparison with Spaced-Receiver Drift Measurements
16
b)
Boa Vista
Campo GrandeAlta Floresta
VHF east link (channels 3-4)
VHF west link (channels 1-2)
GPS spaced-receivers
IPE analysis
VHF spaced-receiver drift provided by Robert Livingston (SCION
Associates)
GPS spaced-receiver drift provided by Marcio Muella et al. [2008]
Next Steps – Chains of Receivers from LISN Network
17
a)
b)
• Apply analysis to observations from meridional chain of GPS
receivers:
Can provide latitudinal and local time morphology of
ionospheric turbulence responsible for scintillations at low
latitudes.
• Apply analysis to observations from longitudinal chain of GPS
receivers:
Can provide evolution of turbulence within individual
bubbles (i.e. in a coordinate system that follows the
bubbles).
• These GPS receivers are already operating—let’s extract the
most information from them as possible!
Latitude (deg)
Long
itude
(deg
)
Conclusions
18
• Reference: Carrano, C. S., C. E. Valladares, K. M. Groves, Latitudinal and Local Time Variation of Ionospheric Turbulence Parameters during the
Conjugate Point Equatorial Experiment (COPEX) in Brazil, submitted to International Journal of Geophysics, 2011.
• We developed the IPE technique to characterize the turbulent ionospheric medium that produced GPS scintillations during the 2002 COPEX experiment in
Brazil.
• Our analysis of data collected on 1-2 Nov 2002 provides the latitudinal and local time variation of turbulent intensity, phase spectral index, and zonal drift.
The strength of turbulence tends to be largest where the TEC is largest (qualitatively)
The phase spectral index increases with local time from 2.5 to 4.5, we conjecture this is due erosion (decay) of small scale irregularities in the
turbulence.
Our estimates of the zonal irregularity drift are consistent with those provided by the spaced GPS receiver technique (but our technique requires
only 1 receiver)
Our drift estimates are similar at the three stations, unlike those provided by the VHF spaced-receiver technique which are larger at Campo Grande
than at Boa Vista.
• Next steps: Apply to LISN GPS receivers arranged in meridional/longitudinal chains
Extra Slides
Determination of Model Parameters CkL, p, and VD
20
5.0x1034
1.0x1035
5.0x1035
1.0x1036
2.5
3.0
3.5
4.0
75 m/s
150 m/s
300 m/s
400 m/s
2.3x1035
3.6 212 m/s
Varying CkL
Varying p Varying VD
High-Rate GPS Receivers Operating During COPEX
21
Station Geographic GeomagneticLat. Lon. Dip Angle Inclination Declination
Boa Vista 2.8°N 60.7°W 22.1°N 22.1° -13.0°Alta Floresta 9.9°S 56.1°W 3.38°S -3.2° -14.6°Campo Grande 20.5°S 54.7°W 22.3°S -21.8° -13.8°