seminar_20150424

HF PROPAGATION Results : Metal

Oxide Space Cloud (MOSC)

Experiment

Dev Joshi

Research Assistant

Department of Physics, Boston College(BC)

Institute For Scientific Research(ISR), BC

ISR SEMINAR

1

Ionospheric Modification

Experiments

HF Tx HF Rx ALTAIR Incoherent

Scatter RADAR

Metal Vapor

Release

Rocket

Ionosphere

Two types of Ionospheric modification experiments :

• Ionization Depletion

• Ionization Enhancement

2

HF PROPAGATION Results : Metal

Oxide Space Cloud Experiment

Outline : • Introduction

• Ionosphere Modification

Experiment

• HF propagation in ionosphere

• Observations

• Modeling Results

• Conclusions

3

Ionospheric Modification

Experiments

Ionospheric Modification Experiments: Various materials have been injected

into the high atmosphere creating perturbations to the ambient medium.

• To detect plasma drift velocities and electric fields.

• To create artificial comets.

• To exploit the ionosphere and magnetosphere as a plasma

laboratory without walls.

• To increase or decrease the plasma density in the

ionosphere to trigger large scale phenomenon.

• Payload for each rocket included

‒ Two canisters of samarium

(5 kg yield)

‒ Dual Frequency RF Beacon

(NRL CERTO)

• Ground diagnostics from 5 sites included:

‒ ,GPS/VHF

Scintillation Rxs, All-Sky Cameras,

Optical Spectrograph, Ionosondes,

Beacon Rx,

Incoherent Scatter Radar

HF Tx/RX 4

Ionization Depletion Chemistry

• H2O, H2, SF6

• An example with water and hydrogen molecules

Ionization Enhancement

• Barium, Strontium, Xenon, Lithium, or Cesium: Photoionization

• Lanthanides: Photoionization and Chemi- ionization

Ionization Depletion and

Enhancement

These panels show pre-event conditions (a) and the resultant

perturbations at 14 minutes (b), 40 minutes (c), and 107 minutes (d)

after the Shuttle thruster firing in Space Lab 2 mission.

An image of enhanced plasma in MOSC experiment 5

RF propagation : Appleton formula

𝑛2 = µ − 𝑖 χ 2 = 1 − 𝑋

1 −𝑖𝑍 − 𝑌𝑇2

2 1 −𝑋 −𝑖𝑍 ±

𝑌𝑇4

4 1 −𝑋 −𝑖𝑍 2+𝑌𝐿21 2

The Appleton formula:

𝑋 =𝑁𝑒2

є0𝑚ω2 , 𝑌𝐿 = 𝑒𝐵𝐿/𝑚ω , YT = e𝐵𝑇/mω 𝑎𝑛𝑑 𝑍 = υ/ω

where,

Two characteristic modes Possible:

+ Ordinary mode

- Extraordinary mode

A magnetoplasma is birefringent and

anisotropic. Reflection Condition:

The positive sign : 𝑋 = 1

And the negative sign gives

𝑋 = 1 − 𝑌, 𝑋 = 1 + 𝑌

Ordinary Wave

Extra-ordinary Wave

6

RF propagation : Appleton formula

α

α′

Higher Electron Density

Lower Electron Density Higher Refractive Index

Lower Refractive Index

α′ > α

Refraction of radio waves by the ionosphere due to the changes in

the electron density

When collisions are negligible :

𝑛2 = µ2 = 1 − 2𝑋(1 − 𝑋)

2(1 − 𝑋) − 𝑌𝑇2 ± 𝑌𝑇

2+ 4 1 − 𝑋 2𝑌𝐿2 1

2

When the magnetic field is negligible :

𝑛2 = µ − 𝑖 χ 2 = 1 −𝑋

1 − 𝑖𝑍

When both collisions and magnetic field effects are negligible :

𝑛2 = µ2 = 1 − X = 1 −𝑓𝑁

𝑓

2

• Cold Plasma

• No collisions

• No magnetic field

𝑓𝑁 = ; N is in electrons per cubic meter 9 𝑁

; 𝑓𝑁 = plasma frequency

7

= 𝜔

𝑝

2π =

𝑛𝑜𝑒2

ϵ0𝑚

Determination of Electron Densities

Ionosphere

T’ R’ T R

A

B

A’

B’

C

Earth

Oblique Vertical

Real Path

Virtual Path

T = Transmitter

R = Receiver

ℎ′ = 𝑐 𝑑ℎ

𝑢

ℎ𝑟

0 = 𝜇′𝑑ℎ

ℎ𝑟

0 Virtual Height :

8

Determination of Electron Densities

9

Incoherent Scatter Radar

Measures direct scatter from

electrons providing full height

density profile (f >> fpeak)

Ionosonde

Measures electron density profile up to the

peak density height only (f ≤ fpeak)

10

ALTAIR Radar

• Dual Frequency:

150 MHz/422 MHz

• Max Bandwidth:

7 MHz/18 MHz

• 46 m dish

• Peak Power

VHF: 6.0 MW

UHF: 6.4 MW

• Incoherent Scatter: Direct scatter from

electrons in the ionosphere (10-4 m2;

equivalent to a ~dime!)

Advanced Research Project Agency (ARPA)

Long-range Tracking and Identification

Radar (ALTAIR)

Transmitter/Receiver Geometry

Rongelap

Wotho

ALTAIR

Likiep

MOSC Release Location & Likiep-Wotho Mid-Point

11

• Rongelap-Wotho link geometry is predominantly N-S and great-circle path is far

from release region

• Likiep-Wotho path is nearly E-W and release point lies nearly on mid-point of

the link—should be ideal for observing SmO+ layer

N

E

12 12

Kwajalein Atoll

13 13

Roi Namur

14 14

THE RELEASE : MOVIE

15

CLOUD

Two Releases :

01 May and 09 May 2013

• Ionosphere during first release was disturbed, rising rapidly and Spread F formed within

minutes after release

• Ionosphere during second release is canonical quiescent

16

Rongelap TX Likiep TX

Mission 41102 09 May 2013 Pre-Release Sweep Wotho Receiver

• Sweeps from 2-30 MHz were completed every five (5) minutes - Plots show data from only

2-14 MHz since no signatures were observed at higher frequencies

• Slightly higher peak frequencies on Wotho Likiep path relative to Rongelap-Wotho links

probably due to longer path length, lower elevation angle propagation . 17

F -region

Ground Wave

F –region second hop


Mission 41102 09 May 2013 1st Post-Release Sweep Wotho Receiver

• On the Wotho geometry the layer extends up to 10 MHz peak frequency

• There is also a prominent secondary F region echo; the time delays will allow us to calculate

the range offsets

• The discrete nature of the echo suggests a localized perturbation that extends up to the F-

region peak

MOSC layer MOSC layer

F-layer Secondary Echo

18


Mission 41102 09 May 2013 2nd Post-Release Sweep Wotho Receiver

• The layer has now decreased to about 6 MHz peak density

• The F-region perturbation extends only mid-way up the density profile

• The delay is much greater on the Rongelap link geometry compared to the Wotho

geometry providing valuable information on the nature of the perturbation




19


Mission 41100 01 May 2013 3rd Post-Release Sweep Wotho Receiver

• No clear signatures on Rongelap link

• A bottomside F-region perturbation remains on the Wotho-Likiep link, as well as

a possible MOSC layer signature around 4 MHz

F region perturbation

MOSC layer?

20

21

Second Release: 09 May 2013

22

Ionospheric Models

Goal : Model the Background Ionosphere and the Cloud

Background Ionosphere :

International Reference Ionosphere (IRI)

• Standard model of the empirical ionospheric

climatology

• Developed by joint working group of Committee on

Space Research (COSPAR) and International Union

of Radio Science (URSI)

• ‘Input variables’ : R12 , iono_layer_parms =

[foF2,hmF2,foF1,hmF1,foE,hmE]

Parametrized Ionospheric Model (PIM)

• Global model of theoretical ionospheric

climatology

• Combination of four physics-based numerical

ionospheric models

• ‘Input Variables’: F10.7 index, SSN, Kp index

Plots courtesy of William McNeil

23 23

IRI and ALTAIR Profiles

Approximately 30 seconds after release the MOSC

cloud has a peak density of about 106 e-/cc, slightly less

than the background ionosphere (Ne = 106 corresponds

to a plasma frequency of 9 MHz).

The layer is about 30 km in diameter by this time.

BEFORE

30 SEC AFTER

Time Period for

Modeling Cloud

MOSC Launch 2: May 9, 2013

Modeling the Cloud

24

25

The averaged and

symmetrized cloud

profile is used to

model the cloud in

MATLAB with

latitude/longitude

increment at

0.0141degree and

height increment at

1.5510 km . The

central pixel

corresponds to

7.4369 MHz


Modeling the Cloud

Ray Tracing

Haselgrove Equations:

• Fermat’s principle - δ 𝑚𝑑𝑠 = 0 , m is ray refractive index and ds the element of path length

• Variational method – ray path equations 𝑑𝑥𝑖𝑑𝑡

= 𝜕𝐺

𝜕𝑢𝑖

, 𝑑𝑢𝑖

𝑑𝑡 = -

𝜕𝐺

𝜕𝑥𝑖 ; 𝐺 = 𝑢 μ

• Numerical Integration of ray path equations : Runge Kutta method

Coleman Method :

• Point to point ray tracing

• Variational principle is discretized and resulting discretized equations solved

PHaRLAP:

• HF radio wave ray tracing toolbox developed by DSTO, Australia

• 2D raytracing is an implementation of the 2-D equations developed by Coleman

• 3-D engine is based upon the equations of Haselgrove

26

Ray Tracing : Example

27

• HF signals received well off the great

circle path to the receiver

• The artificial cloud bends the HF

energy through large angles

• Excellent agreement between model

and observations

Rongelap

Wotho

MOSC

Release

Point

3D Ray Trace Analysis

Rongelap To Wotho

28

• Great circle path to the receiver passes

through the MOSC volume

• Multiple paths between ionosphere, cloud

and receiver expected

Likiep

Wotho

MOSC Release Point

Likiep Wotho

MOSC

3D Ray Trace Analysis

Likiep To Wotho

29


Mission 41100 01 May 2013 Pre-Release Sweep Wotho Receiver

30

F -region

E - layer

Ground Wave

F-region – Second hop


Mission 41100 01 May 2013 1st Post-Release Sweep Wotho Receiver

• Note that the Likiep signature is only evident in high end of frequency range, showing up near f = 8 MHz

(~07:42 UT); one might conclude this is results from the temporal evolution of the cloud, yet the

Rongelap link shows the signature beginning at less than 4 MHz at least 40 seconds earlier.

• One possibility is that the lower frequency components on the direct Likiep-Wotho link were actually

blocked, refracted or ducted by the presence of the MOSC cloud.



31


Mission 41100 01 May 2013 2nd Post-Release Sweep Wotho Receiver

• Peak density has decreased significantly over 5 minute time scale, primarily due to

expansion of the cloud


32


Mission 41100 01 May 2013 3rd Post-Release Sweep Wotho Receiver

• The Rongelap path shows a low density residual signature of the cloud

• The Likiep path shows an emerging perturbation in the F-region

• By 07:56 and beyond MOSC has disappeared and Spread F dominates

MOSC layer F region perturbation

33

34

First Release: 01 May 2013


Modeling the Ionosphere

The ‘ALTAIR’ is modelled by applying the same relative differences as in PIM in lat-lon

plane at an altitude. The technique is applied at all altitudes. As is seen, modeled ‘ALTAIR’

fails to capture the shape of the high frequency tilt in PIM. 35



36

The change in height of the peak density across longitudes in PIM isn’t seen

in the modeled ‘ALTAIR’ ionosphere.



37

An averaged shape function is chosen to apply to the change in the height of the

peak density across longitudes.



38

The applied shape function produced delay in good agreement except 9 MHz.



39

• The anchor profile is taken to be the

average of three ALTAIR radar profiles.

• This gives delay to a good agreement

with the observations.



The ‘modeled’ bottomside is in good agreement while the top side is above the observations.

The removal of tilt due to the shape function shows further disagreement as it raises the upside. 40

Optimization : Nalder Mead Down Hill Simplex Method ( Amoeba)*

• Nelder, J. and R. Mead ,1965

• Direct Method : No derivatives, only function values

• Idea : Move from high function (hot) areas to low function

(cold) areas by reflections, expansions and contractions

• “Amoeba Crawls Downhill with no assumption about function”

• Built-in function in MATLAB : fminsearch

Modeling : Optimization Method

41 *Minimization Technique suggested by Dr. Charles Carrano, ISR

42

Optimization : Model Ionosphere

PIM doesn’t have enough degrees of freedom to fit the ALTAIR radar profile while

optimization of foF2 and hoF2 in IRI closely matches the observed ALTAIR radar

profile.

Scale Vector = [ a b c …. e]

43

Optimization : Delay results

The IRI model, scaled and unscaled, doesn’t match the observed delay.

Optimization : Delay results

44

• Optimization of the scaling vector matches only the upside

• Frequency specific optimization of the scale vector exactly reproduces the observed delay.

Conclusions

• Ray tracing confirms sounder observations to high degree of fidelity

• The change in ambient natural propagation environment due to small

artificial modification can be successfully modeled

• Effects from arbitrary artificial plasma environments can be predicted

with accuracy

• Optimization technique represents a new method of assimilating

oblique ionosonde data to generate the background ionosphere

(numerous applications for HF systems)

• Future Work : Modeling of natural disturbances in the low latitude

propagation environment to understand the effects of Traveling

Ionospheric Disturbances (TIDs) and Spread F on perpendicular and

quasi-parallel (to B) paths.

46

Thank You.

47

seminar_20150424

Documents