Characterization of GPS Scintillations in the Polar Ionosphere

Characterization of GPS Scintillations in the Polar Ionosphere

Yaqi Jin

Faculty of Mathematics and Natural Sciences

UNIVERSITETET I OSLO

May 2016

© Yaqi Jin, 2016

Series of dissertations submitted to the Faculty of Mathematics and Natural Sciences, University of Oslo No. 1759

ISSN 1501-7710

All rights reserved. No part of this publication may be reproduced or transmitted, in any form or by any means, without permission.

Cover: Hanne Baadsgaard Utigard.Print production: Reprosentralen, University of Oslo.

III

Characterization of GPS Scintillations in the Polar Ionosphere

Yaqi Jin

Thesis submitted for the degree of

Philosophiae Doctor

Department of Physics

University of Oslo

May 2016

IV

V

Abstract The climatology map of the GPS phase scintillation at high latitudes identifies two regions of

high scintillation occurrences: around magnetic noon and magnetic midnight. The scintillation

occurrence rate is higher around noon, while the scintillation level is strongest around

magnetic midnight.

Paper 1 focuses on the dayside scintillation region. In order to resolve the role of the cusp

auroral processes in the production of irregularities, we put the GPS phase scintillation in the

context of the observed auroral morphology. Results show that the occurrence rate of the GPS

phase scintillation is highest inside the auroral cusp, regardless of the scintillation strength

and the interplanetary magnetic field (IMF). On average, the scintillation occurrence rate in

the cusp region is about 5 times as high as in the region immediately poleward of it. The

scintillation occurrence rate is higher when the IMF Bz is negative. When partitioning the

scintillation data by the IMF By, the distribution of the scintillation occurrence rate around

magnetic noon is similar to that of the poleward moving auroral form (PMAF): there is a shift

in the occurrence rate towards prenoon (postnoon) when the IMF By is positive (negative).

This indicates that the irregularities which give rise to scintillations follow the IMF By

controlled east-west motion of the cusp auroral forms and the flow of solar EUV ionized

plasma into the polar cap. Furthermore, the scintillation occurrence rate is higher when IMF

By is positive which can be explained as follows: during positive IMF By, the cusp is shifted

toward the postnoon sector where it may access to the higher density plasma. This suggests

that the combination of the auroral activities (e.g., PMAF) and the intake of the high density

solar EUV ionized plasma are crucial for the production of scintillations.

In Papers 2 and 3, we directly compare the relative GPS phase scintillation levels associated

with three phenomena: regions of enhanced plasma irregularities called auroral arcs, polar cap

patches, and auroral blobs which frequently occur in the polar ionosphere. We define two

types of auroral blobs: blob type 1 (BT-1) which is formed when islands of high-density F

region plasma (polar cap patches) enter the nightside auroral oval, and blob type 2 (BT-2)

which are generated locally in the auroral oval by intense particle precipitation alone. In a

case study (Paper 1) based on observations from Ny-Ålesund on January 13, 2013, we

detected several polar cap patches exiting the polar cap into the auroral oval (then termed BT-

1 blobs). The BT-1 blobs were associated with the strongest phase scintillation, followed by

VI

patches and BT-2 blobs (produced by pure auroral arcs). In the statistical study (Paper 3), we

show that BT-1 blobs are associated with significantly higher scintillation level than the

corresponding polar cap patches in general; however, there is no clear relationship between

the scintillation level inside the polar cap patch and the resulting BT-1 blob. For BT-2 blobs

we find that they are associated with much weaker scintillations than BT-1 blobs. Compared

to polar cap patches and BT-2 blobs, the significantly higher scintillation level for BT-1 blobs

implies that the auroral dynamics plays an important role in the structuring of BT-1 blobs.

Since BT-1 blobs are formed after patches merged into the auroral region, it will be important

to enable predictions of patches exiting the polar cap in space weather predictions of GPS

scintillations.

VII

Acknowledgments I owe many thanks to my main supervisor Jøran Moen for his invaluable guidance and

patience. Without him this PhD thesis would not have been accomplished. I also want to

thank my co-supervisor Wojciech Miloch. I was greatly impressed by his carefully reading of

every version of my paper drafts. Special thanks to Lasse Clausen, who has supported me to

run the EISCAT campaign in the winter of 2015 and to attend several international

conferences. These travels really made my PhD life more colorful. The technical support from

Bjørn Lybekk is greatly acknowledged. Bjørn is an incredibly helpful person and he knows

everything in the plasma and space physics group. His knowhow and support make him to be

the most important person in this group. Thanks to Espen Trondsen and Njål Gulbrandsen for

introducing me to the all-sky imager world.

I feel very lucky for having taken the course FYS9610 (Magnetospheric processes) which was

taught by Per Even Sandholt. I began to understand the Solar Wind-Magnetosphere-

Ionosphere coupling from a large scale picture. This is especially important to me since my

study focuses on a small field of view around Svalbard. Thanks to Xiaoyan Zhou who guided

me into the fantastic tour in the shock-aurora world.

In my PhD years, I was involved in the teaching of a Bachelor level course (FYS3610)

together with Lasse Clausen. There I have reread several textbooks and old scientific papers

in order to answer the questions raised by the students. I really learned a lot from this course.

Thanks to all the students during the three years.

I spent one semester staying at the University Centre in Svalbard (UNIS). Thanks to the

administration staff and teachers. It is their hard work that made me feel at home even if it is

the Northmost place on Earth I have ever been to.

Thank Jøran Moen, Wojciech Miloch, Lasse Clausen, Andres Spicher and Bjørn Lybekk for

proof-reading this thesis.

Finally I would like to thank my family, friends, classmates and colleagues.

Blindern, Oslo, May, 2016

Yaqi Jin

VIII

IX

Contents Abstract

Acknowledgments

1 Background ........................................................................................................................ 1

1.1 Introduction ................................................................................................................. 1

1.2 Space Weather ............................................................................................................. 3

1.3 Dungey Cycle and Ionospheric Convection ................................................................ 6

1.4 Dayside Aurora and Cusp Aurora ............................................................................... 7

1.5 Large Scale Plasma Structures at High Latitudes: Polar Cap Patches and Blobs ..... 10

1.5.1 Polar Cap Patches ............................................................................................... 11

1.5.2 Boundary Blobs, Sub-auroral Blobs, and Auroral Blobs ................................... 12

1.6 Instabilities................................................................................................................. 14

1.6.1 Kelvin-Helmholtz Instability .............................................................................. 15

1.6.2 Gradient Drift Instability .................................................................................... 16

1.6.3 Current Convective Instability ........................................................................... 19

1.7 Ionospheric Scintillation ............................................................................................ 21

1.7.1 Scintillation Theory ............................................................................................ 22

1.7.2 Observations of Ionospheric Scintillations ........................................................ 24

1.8 Previous GPS Scintillation Studies in the Auroral/Polar Cap Region ....................... 24

2 Instrumentation and Methodology ................................................................................... 31

2.1 All-Sky Imager .......................................................................................................... 31

2.2 GPS Ionospheric Scintillation and TEC Monitor (GISTM) ...................................... 35

2.2.1 The GPS Phase Scintillation Index .................................................................... 36

2.2.2 The GPS Amplitude Scintillation Index............................................................. 36

3 Summary of Papers .......................................................................................................... 39

3.1 Paper Abstracts .......................................................................................................... 39

3.2 Summary .................................................................................................................... 40

3.3 Future Work ............................................................................................................... 45

4 Bibliography ..................................................................................................................... 47

X

Paper 1 - Jin, Y., J. I. Moen, and W. J. Miloch (2015), On the collocation of the cusp aurora and the GPS phase scintillation: A statistical study, J. Geophys. Res. Space Physics, 120, doi:10.1002/2015JA021449.

Paper 2 - Jin, Y., J. I. Moen, and W. J. Miloch (2014), GPS scintillation effects associated with polar cap patches and substorm auroral activity: Direct comparison, J. Space Weather Space Clim., 4, A23, doi:10.1051/swsc/2014019.

Paper 3 - Jin, Y., J. I. Moen, W. J. Miloch, L. B. N. Clausen, and K. Oksavik (2016), Statistical study of the GNSS phase scintillation associated with two types of auroral blobs, J. Geophys. Res. Space Physics, 121, doi:10.1002/2016JA022613.

Publications not included in the thesis:

Paper 4 - Bjoland, L. M., Chen, X., Jin, Y., Reimer, A. S., Skjæveland, Å., Wessel, M. R., Burchill, J. K., Clausen, L. B. N., Haaland, S. E. and McWilliams, K. A. (2015), Interplanetary magnetic field and solar cycle dependence of Northern Hemisphere F region joule heating. J. Geophys. Res. Space Physics, 120: 1478–1487. doi: 10.1002/2014JA020586.

Paper 5 - Jin, Y., X. Zhou, J. Moen, and M. Hairston (2016), The Auroral-Ionosphere TEC

Response to an Interplanetary Shock. Geophys. Res. Lett., 42, doi: 10.1002/2016GL067766.

XI

List of Figures Figure 1.1: Schematic illustration of the work presented in this thesis. ................................... 2 Figure 1.2: An image of the Sun captured by NASA's SDO (Solar Dynamics Observatory) at 304 Å. ......................................................................................................................................... 3 Figure 1.3: Schematic illustration of the Dungey cycle. ........................................................... 6 Figure 1.4: Magnetic noon values of solar elevation angle (SEL, black lines) and twilight (blue shading) during the local winter solstice, for the Antarctic (left) and Arctic (right) regions. ....................................................................................................................................... 8 Figure 1.5: A schematic illustration of dayside auroral configurations for four different IMF orientations. ................................................................................................................................ 8 Figure 1.6: Selected pseudo-color images of 630.0 nm aurora during the period 0948.54 UT to 1019.40 UT on December 24, 1995. ...................................................................................... 9 Figure 1.7: Snapshots of the proton aurora on 18 March 2002 showing the continuous presence of the proton aurora spot. ......................................................................................... 10 Figure 1.8: Electron density contours measured on November 11, 1981, by the Chatanika incoherent scatter radar. .......................................................................................................... 13 Figure 1.9: Computer simulation of convective reconfiguration of a circular polar cap patch (colored region in a) as it convects from the polar cap through the auroral zone. ................. 14 Figure 1.10: The linear growth rates of the Kelvin-Helmholtz instability versus wave number k. ............................................................................................................................................... 15 Figure 1.12: Illustration of the F-region gradient drift instability. ......................................... 17 Figure 1.13: The simulation of the gradient drift instability. .................................................. 18 Figure 1.14: The conceptual picture of the current convective instability. ............................. 20 Figure 1.15: The irregularity spectrum and the corresponding amplitude and phase scintillation spectra for the weak scatter. ................................................................................ 22 Figure 1.16: Global morphology of the ionospheric scintillation activity (fades) during the solar maximum and solar minimum [Basu and Groves, 2001]. .............................................. 24 Figure 1.17: Detailed comparison of the amplitude scintillation (S4) and the GPS TEC for the period from 23 UT February 3 to 05 UT February 4, 1984. ................................................... 25 Figure 1.18: The vertical TEC, phase (σϕ) and amplitude scintillation indices (S4) from the GPS satellite PRN31 recorded by the GISTM in Ny-Ålesund, Svalbard between 21:00 and 22:00 UT, on 30 October, 2003 [Mitchell et al., 2005]. .......................................................... 26 Figure 1.19: (left) Scintillation maps showing the occurrence percentage of σϕ > 0.25 rad and (right) the occurrence percentage of S4 > 0.25. ................................................................ 28 Figure 1.20: The GPS scintillation climatology observed between October and December 2003 over Ny-Ålesund obtained by binning the data for conditions of IMF Bz > 0 (left column) and Bz < 0 (right column). ......................................................................................... 29 Figure 2.1: The coverage of four all-sky imagers when projected onto the geographic coordinate. ................................................................................................................................ 32 Figure 2.2: The NYA4 all-sky imager (left) and its control box (right). .................................. 33

XII

Figure 2.3: Illustration of the normal operational sequence for the NYA4 all-sky imager (Courtesy of Bjørn Lybekk). ..................................................................................................... 33 Figure 2.4: (left) Schematic drawing of a pin-hole camera. The focal length is f, θ is the zenith angle of the incident ray and r is the distance from the center of the image. (right) different projection method of the lens. .................................................................................... 34 Figure 2.5: The hardware of the GISTM system. .................................................................... 35 Figure 3.1: MLAT-MLT maps show the GPS phase scintillation occurrence rate for observations from year 2010 to 2013 in Ny-Ålesund binned by σϕ from (a) (0.1, 0.25) rad, (b) (0.25, 0.5) rad, and (c) 0.5 rad. ............................................................................................ 41 Figure 3.2: The distribution of the scintillation occurrence rate at (a) sub-aurora, (b) cusp, and (c) polar cap regions. ........................................................................................................ 42 Figure 3.3: Event overview. ..................................................................................................... 43 Figure 3.4: Summary of three papers in this thesis. ................................................................ 44

1

1 Background

1.1 Introduction When the interplanetary magnetic field (IMF) is directed southward, the solar wind-

magnetosphere-ionosphere coupling gives rise to a composite of several characteristic

phenomena. Figure 1.1 illustrates the ionospheric phenomena resulting from this coupling

which is of particular relevance to this thesis. One of the phenomena is the large scale twin

cell convection pattern as illustrated by black contours. The convection can carry the high

density dayside solar EUV (extreme ultraviolet) ionized plasma (shaded by purple in Figure

1.1) across the polar cap to the nightside. Polar cap patches, islands of high density F region

plasma, are created when the solar EUV ionized plasma enters the polar cap in the cusp

inflow region where they can be structured by the cusp auroral dynamics (red band in Figure

1.1). Svalbard is an ideal place to observe the cusp inflow region as it is situated right

underneath the cusp aurora around local noon and it is heavily instrumented. Polar cap

patches are termed auroral blobs when they enter the nighime auroral oval. This is a highly

dynamic region and the polar cap patches may be degraded into smaller structures as

illustrated in Figure 1.1. The auroral blobs are then transported towards the dayside by the

sunward return flow from the dawn and duskside and then they complete a full Dungey

convection cycle as presented in section 1.3.

The ionospheric scintillation is caused by electron density irregularities in the ionosphere. The

irregularities are often developed in the body of the large scale (>100 km) plasma structures

such as polar cap patches by plasma instability processes. The auroral oval is a region with

energy and particle deposition which serve as the instability drivers. In the high-latitude

ionosphere, scintillations may occur in the entire auroral oval and in the polar cap, bounded

by the auroral oval. These scintillations have for decades been attributed to polar cap patches

and auroral precipitation.

This PhD project aims to identify the most severe scintillation region in the polar ionosphere.

We study the relative contribution of polar cap patches and auroral precipitation to

scintillations. In the polar ionosphere, the plasma processes and auroral dynamics are different

between the dayside and nightside. Therefore, we study the associated scintillation separately.

2

Within the frame of the discussion above, I will briefly describe the system knowledge that

the current research project is based on: section 1.2 puts the GNSS scintillation studies into

the context of space weather; section 1.3 describes the Dungey cycle which is the magnetic

reconnection driven ionospheric convection; section 1.4 presents the relevant knowledge of

the dayside aurora and cusp aurora; section 1.5 reviews the literature on the large scale plasma

structures at high latitudes; section 1.6 describes the relevant instability modes in the high-

latitude F region ionosphere; section 1.7 presents the theory and observations of the

ionospheric scintillation. Finally, in section 1.8, we review the scintillation studies in the

auroral/polar cap region.

Figure 1.1: Schematic illustration of the work presented in this thesis. The purple color

indicates the solar EUV ionized plasma (the darker color indicates higher density). The cusp

aurora is shown as a red band around local noon. The nightside aurora is shown in green.

Polar cap patches and auroral blobs are shown as purple islands. The black contours

illustrate the classical twin cell convection pattern for IMF Bz < 0 and By = 0. The map of

Svalbard and North Scandinavia is shown in blue to indicate observations on the dayside and

nightside, respectively.

3

1.2 Space Weather According to the US National Space Weather Programme, space weather refers to “conditions

on the Sun and in the space environment that can influence the performance and reliability of

space-borne and ground-based technological systems, and can endanger human life or health”.

This definition implies the broadness of space weather phenomena which includes all the

fields in Space Physics, e.g., radiation hazards to spacecraft and human, disruptions of GNSS

(Global Navigation Satellite System) services and of communication systems.

Figure 1.2: An image of the Sun captured by NASA's SDO (Solar Dynamics Observatory) at

304 Å. An X1.8-class solar flare on the Sun was observed as a bright spot in the right of the

solar disk on December 20, 2015. The image shows a subset of EUV (extreme ultraviolet)

light that highlights the extremely hot solar material, which is typically colorized in red. X-

class denotes the most intense flares. The image is taken from

http://sdo.gsfc.nasa.gov/data/aiahmi/, courtesy of NASA/SDO and the AIA, EVE, and HMI

science teams. The 31-day running average of the daily sunspot number is superimposed on

4

the full-Sun image. The sunspot number data are from

http://omniweb.gsfc.nasa.gov/form/dx1.html.

The Sun is the ultimate source of space weather. As a result, the solar activity dominates the

“weather” and “climate” changes in the solar system. The most known solar activity is the

sunspot cycle which shows an 11-year cycle. At solar minimum the sunspot number and solar

activity are low, while at solar maximum the sunspot number and solar activity are high. The

solar cycle variation is shown by the white curve in Figure 1.2 as the 31-day running average

of the daily sunspot number. Apart from the solar cycle variation, the solar energy output also

exhibits weather-like variation such as fast and slow solar wind streams, co-rotating

interaction regions, solar flares, coronal mass ejections (CME). The solar wind (plasma)

streams from the Sun at supersonic speed (200 km/s – 800 km/s). Due to the frozen-in

theorem the plasma carries the Sun’s magnetic field forming the IMF. Figure 1.2 shows a

solar image which was recorded by the SDO (Solar Dynamics Observatory). The image

shows the burst of an intense solar flare, which is a powerful burst of radiation. Harmful

radiation can disturb the ionosphere. The GPS and communication systems are degraded

when the radio signals pass through the disturbed ionosphere. In addition to the burst of solar

irradiance in solar flare, the CME associated with it carries massive amounts of magnetized

plasma. The CME can affect the Earth’s space environment considerably as it magnetically

couples to the Earth's magnetosphere.

For practical usage, Jacobsen and Schafer [2012] reported the real-time kinematic (RTK)

positioning service in Norway (57° - 72° N) during a strong geomagnetic storm which was

produced by a CME. During the storm, the positioning service was affected all over Norway

and the correction data from the network were unusable for several hours due to limited

visible satellites. Therefore, the users could not get promised positioning accuracy of

centimeter as in the normal RTK service. Large positioning errors were observed up to 3

meters from Southern Norway at the Hønefoss monitor station (60.14° N, 10.25° E). The

positioning service in Northern Norway is expected to be worse and is disturbed more

frequently since it is just underneath the auroral oval during night.

That geomagnetic storm presented in Jacobsen and Schafer [2012] reached a level of G3

according to the NOAA space weather scale for geomagnetic storms. The NOAA space

weather scales are introduced in a way to communicate to the general public the current and

future space weather conditions and their possible effects on humans and systems. The scale

5

for geomagnetic storms is shown in Table 1.1. The occurrence of each storm level is shown in

the last column. On average, 200 G3-level storms occur during one solar cycle (11 years),

thus, the disturbance of the RTK service presented by Jacobsen and Schafer [2012] is not a

rare event and one should be prepared to tackle it.

Table 1.1: Goemagnetic storms in the NOAA Space Weather Scales [Poppe, 2000].

Scale Description Effect Physical measure

Average Frequency

(1 cycle = 11 years)

G 5 Extreme

Power systems: Widespread voltage control problems and protective system problems can occur, some grid systems may experience complete collapse or blackouts. Transformers may experience damage. Spacecraft operations: May experience extensive surface charging, problems with orientation, uplink/downlink and tracking satellites. Other systems: Pipeline currents can reach hundreds of amps, HF (high frequency) radio propagation may be impossible in many areas for one to two days, satellite navigation may be degraded for days, low-frequency radio navigation can be out for hours, and aurora has been seen as low as Florida and southern Texas (typically 40° geomagnetic lat.).

Kp = 9 4 per cycle (4 days per

cycle)

G 4 Severe

Power systems: Possible widespread voltage control problems and some protective systems will mistakenly trip out key assets from the grid. Spacecraft operations: May experience surface charging and tracking problems, corrections may be needed for orientation problems. Other systems: Induced pipeline currents affect preventive measures, HF radio propagation sporadic, satellite navigation degraded for hours, low-frequency radio navigation disrupted, and aurora has been seen as low as Alabama and northern California (typically 45° geomagnetic lat.).

Kp = 8, including

a 9-

100 per cycle (60 days per

cycle)

G 3 Strong

Power systems: Voltage corrections may be required, false alarms triggered on some protection devices. Spacecraft operations: Surface charging may occur on satellite components, drag may increase on low-Earth-orbit satellites, and corrections may be needed for orientation problems. Other systems: Intermittent satellite navigation and low-frequency radio navigation problems may occur, HF radio may be intermittent, and aurora has been seen as low as Illinois and Oregon (typically 50° geomagnetic lat.).


cycle)

G 2 Moderate

Power systems: High-latitude power systems may experience voltage alarms, long-duration storms may cause transformer damage. Spacecraft operations: Corrective actions to orientation may be required by ground control; possible changes in drag affect orbit predictions. Other systems: HF radio propagation can fade at higher latitudes, and aurora has been seen as low as


cycle)

6

New York and Idaho (typically 55° geomagnetic lat.). G 1 Minor Power systems: Weak power grid fluctuations can

occur. Spacecraft operations: Minor impact on satellite operations possible. Other systems: Migratory animals are affected at this and higher levels; aurora is commonly visible at high latitudes (northern Michigan and Maine).

Kp = 5 1700 per cycle (900 days per cycle)

1.3 Dungey Cycle and Ionospheric Convection

Figure 1.3: Schematic illustration of the Dungey cycle. (a) The circulation of magnetic field

lines [Kivelson and Russell, 2001]. (b) Sketch showing the high-latitude ionospheric plasma

flow due to unbalanced dayside reconnection (b1) and unbalanced tail reconnection (b2),

respectively. (This is from Figure 3 of Cowley and Lockwood [1992]).

Dungey [1961] proposed the circulation of the Earth’s magnetic field and plasma caused by

magnetic reconnection between the IMF and the terrestrial magnetic field. The succession of

the Dungey cycle after magnetic reconnection is shown by numbered field lines in Figure

1.3a. The Earth’s magnetic field line 1 connects with the southward IMF 1’ during dayside

reconnection and becomes open. Then the newly opened field moves tailward with the solar

wind. After some time, field lines 6 and 6’ meet and reconnect at an X-line in the tail, after

which the field line returns to the dayside. The inset in the bottom of Figure 1.3a shows the

7

foot-points of these field-lines in the high-latitude ionosphere and the corresponding plasma

flows, with an anti-sunward flow in the polar cap and a sunward return flow at lower

latitudes.

Cowley and Lockwood [1992] established a two-component flow model of the ionospheric

convection. The two time dependent components are the dayside magnetopause reconnection

and tail reconnection. This concept is also referred to as Expanding and Contracting Polar

Cap (ECPC) paradigm which successfully describes pulsed flow dynamics as well as the

dynamic size of the auroral oval/polar cap. The ECPC process is schematically illustrated in

Figure 1.3b. The circle represents the OCB. The polar cap area increases and the circle

expands as the dayside reconnection adds open flux to the system. The plasma flow crosses

the OCB only in the dashed line portion of the circle which maps to the dayside neutral line.

Elsewhere the boundary moves with the plasma flow. This is termed as unbalanced dayside

reconnection which is manifested as an expansion of the auroral oval. Panel b2 illustrates the

flow driven by the unbalanced nightside reconnection. The OCB contracts as the nightside

reconnection closes open flux. The flow is strongest near the nighside merging gap. In general

the total flow is a sum of the two components shown in Figure 1.3b.

1.4 Dayside Aurora and Cusp Aurora This section illustrates the basics and the dynamics of the dayside aurora and cusp aurora.

This builds a background for interpreting the results in Paper 1.

8

Figure 1.4: Magnetic noon values of solar elevation angle (SEL, black lines) and twilight

(blue shading) during the local winter solstice, for the Antarctic (left) and Arctic (right)

regions. AACGM magnetic latitudes (epoch 2010) are overlaid as red lines. Magnetic poles

are shown as red dots; geographic poles are shown as golden dots. A 1° “typical” optical

cusp (for IMF Bz ~ 0), from 75.4°-76.4° MLat (magnetic latitude) is shown as a red band

[Johnsen and Lorentzen, 2012]. Observatories well located for optical cusp measurements

(SEL lower than -12° and within viewing distance of cusp auroral altitudes above 10°

elevation angle) are indicated as numbered green dots. For comparison, other existing and

historical observatories are shown in dark red. (This figure is taken from Holmes [2014]).

Since ground-based optical measurements require darkness, there are limited ground-based

observatories which allow for cusp auroral measurements. Figure 1.4 is taken from Holmes

[2014] which displays the nominal cusp aurora marked as the red band. The cusp aurora can

be observed from ground where the red band overlaps with the dark blue area (representing

the winter darkness). Observatories used for daytime optical measurements are marked by the

green dots. Green dots number 6 and 8 mark Longyearbyen and Ny-Ålesund, respectively.

Figure 1.5: A schematic illustration of dayside auroral configurations for four different IMF

orientations. Seven different forms are marked by numbers. The auroral motion is indicated

by black arrows. (This is from Figure 1 of Sandholt et al. [1998]).

9

The dayside aurora has been extensively studied by Sandholt et al. by using ground-based

optical measurements [see e.g., Sandholt et al., 2002]. As shown in Figure 1.5, Sandholt et al.

[1998] sub-divided the dayside aurora into seven categories as a function of the orientation of

the IMF, and the cusp auroral configurations are divided into type 1 cusp aurora for

magnetopause reconnection and type 2 cusp aurora for lobe reconnection [Sandholt et al.,

1996a, 1996b; Øieroset et al., 1997]. Type 1 occurs at lower latitudes (~70°-75° MLAT),

while Type 2 occurs at higher latitudes (~75°-79° MLAT). Type 1 aurora is often associated

with poleward moving auroral forms (PMAFs) whose east-west motion is regulated by the

IMF By component. The cusp aurora around magnetic noon is due to low energy

magnetosheath particles and is generally dominated by the red-line (630.0 nm) emission over

the green-line (557.7 nm).

Figure 1.6: Selected pseudo-color images of 630.0 nm aurora during the period 0948.54 UT

to 1019.40 UT on December 24, 1995. The auroral emission is mapped onto a geographical

map by assuming an emission altitude of 250 km. The images have been cut at an elevation

angle of 15°. (This is from Figure 2 of Moen et al. [1999]).

Moen et al. [1999] reported a continuous observation of the cusp auroral dynamics in

response to an IMF By polarity change. Figure 1.6 displays the auroral forms during the

10

period when the IMF Bz was negative, and the IMF By showed a smooth transition from

negative to positive at ~09:52 UT (IMF shifted to the ground-based measurements). Figure

1.6 shows a clear transition of the region of the auroral activity before and after the transition

in the IMF By: being prenoon when By < 0 before 09:52 UT and postnoon when By > 0 after

09:52 UT.

One continuous observation of the cusp aurora for northward IMF Bz was reported by Frey et

al. [2003] by using observations from the IMAGE spacecraft. Figure 1.7 clearly shows the

dawn-dusk shift of the proton spot in association with the IMF By variation.

Figure 1.7: Snapshots of the proton aurora on 18 March 2002 showing the continuous

presence of the proton aurora spot. The Y and Z components of IMF are shown in the upper

left inset, with positive By > 0 pointing to the left and positive Bz pointing up. The dayside

proton aurora spot is seen uninterruptedly over ~4 h. The spot appears on the dayside at ~80°

latitude. Its location in magnetic local time (MLT) is correlated with the IMF By, being in the

pre-noon (post-noon) sector for negative (positive) By. (This is from Figure 2 of Frey et al.

[2003]).

1.5 Large Scale Plasma Structures at High Latitudes: Polar Cap Patches and Blobs Four different types of large scale plasma structures have been classified at high latitudes: 1)

polar cap patches, 2) boundary (and sub-auroral) blobs, 3) Sun-aligned arcs, and 4) auroral

11

blobs. All four types of structures have been studied and they were found to be spatially

collocated with kilometer scale irregularities.

1.5.1 Polar Cap Patches

Polar cap patches are localized enhancements in the F layer plasma density (usually enhanced

by a factor of 2-10 above the background) that 1) have horizontal dimensions ranging from a

few hundred kilometers to 1000 km, 2) are observed in the polar cap, and 3) represent the

largest scale structure that is often associated with kilometer-scale irregularities [Tsunoda,

1988]. Hill [1963] reported the first definitive observations of polar cap patches by using

high-latitude ionograms. The characteristics of patches were extensively studied by the US

Air Force Geophysics Laboratory (AFGL) by using 630.0 nm all-sky imaging photometer,

ionosonde, scintillation receivers and incoherent scatter radars in Greenland and Alaska

[Buchau et al., 1983, 1985; Weber et al., 1984, 1985, 1986; de la Beaujardière et al., 1985].

They found that polar cap patches occurred predominately during southward IMF. Polar cap

patches are identified optically by 630.0 nm airglow intensities several times above the

background [Weber et al., 1986]. Polar cap patches drift anti-sunward over the polar cap with

speeds on the order of 300–1000 m/s. Their UT (Universal Time) dependence and high

densities suggest that the patch material is originated from the solar EUV ionization on the

dayside. High density patches above 300 km altitude can survive for hours and are transported

anti-sunward over long distances (~3000 km) before they exit the polar cap and move into the

nightside auroral zone where they turn into blobs.

The airglow emission line of polar cap patches comes from atomic oxygen at the wavelength

of 630.0 nm. When the enhanced F region plasma is drifting in the polar cap, the dominant

ion (O+ ion) can react with the ambient neutral O2 and result in O2+ ion in the following

process:

ion can recombine with the electron directly and form the excited oxygen atom O*:

When the excited oxygen atom relaxes from 1D to the ground state, one photon is emitted

which corresponds to the wavelength of 630.0 nm [see e.g., Kivelson and Russell, 2001;

12

Hosokawa et al., 2011]. The energy level of 1D is metastable with lifetime of 110 seconds.

The airglow emission intensity is determined by the F region ion density (O+) and the

recombination rate as well. Hence the intensity is strongly affected by the neutral O2 density

and temperature as well [Sojka et al., 1997].

The 630.0 nm airglow is due to the recombination of O+ while the red (line) aurora at 630.0

nm is due electron bombardment of atomic oxygen:

(where has less energy than ), followed by

The red aurora is often associated with much higher intensity (~kR) than the faint airglow

patches (~100 R).

1.5.2 Boundary Blobs, Sub-auroral Blobs, and Auroral Blobs

The studies of boundary blobs and auroral blobs were conducted by the using scanning mode

of incoherent scatter radars [Vickrey et al., 1980; Kelley et al., 1982; Rino et al., 1983; Weber

et al., 1985]. An example of the observed boundary blob and auroral blob is shown in Figure

1.8. The auroral blobs are produced by particle precipitation at lower F and E regions. The

boundary blobs are located near the equatorward part of the auroral E region and they are

originated from the transportation of solar EUV produced plasma (polar cap patches) from the

central polar cap into the midnight auroral sector, and sunward along the equatorward portion

of the auroral zone. The sub-auroral blobs resemble boundary blobs but are found in the

ionospheric trough region. The sub-auroral blobs are consistent with the convective nature of

boundary blobs. During the substorm activity, the region of enhanced westward electric field

expands equatorward with the auroral oval which results in the equatorward transport of the

boundary blob. When the auroral oval moves poleward, the boundary blob remains at the

lowest latitude, which is called a sub-auroral blob.

The similarity between boundary blobs and patches in sizes (the cross-sectional areas

transverse to the magnetic field) strongly suggests that boundary blobs are simply

reconfigured patches [Tsunoda, 1988]. Modelling works have been conducted to verify this

13

hypothesis [Robinson et al., 1985; Anderson et al., 1996]. Robinson et al. [1985] used a two

cell convection model and a circular patch in the central polar cap (Figure 1.9a) to show the

reconfiguration of a polar cap patch. The patch changes in shape as it enters the exit region

around midnight sector. Then the patch changes into latitudinally confined and longitudinally

elongated structures which are supported by the experimental observations. Note that in this

simulation, the patch is continuously pulled out of the polar cap. This may be unrealistic since

the tail reconnection which allows patches to cross the OCB is often pulsed. In the real world,

polar cap patches may be sliced into substructures when they exit the polar cap [Moen et al.,

2015].

Figure 1.8: Electron density contours measured on November 11, 1981, by the Chatanika

incoherent scatter radar. The contours are plotted as a function of altitude and geomagnetic

north distance from the radar (in 100 km cadence). Originally from Rino et al. [1983] and

adapted by Schunk and Nagy [2000].

14

It should be noted that there is no unanimous definition of an auroral blob: Tsunoda [1988]

differentiates between boundary blobs and auroral blobs; however, other scientists [e.g., Basu

et al. 1990] called the large scale structures polar cap patches when found in the polar cap and

auroral blobs when found in the auroral zone. In this thesis, we follow the terminology of

Basu et al. [1990], but we will differentiate between two types of auroral blobs: blob type 1

(BT-1) is connected to the high-density polar cap patch material that has entered the auroral

oval; and blob type-2 (BT-2) relates to the plasma density enhancement/structure that has

been generated locally by particle precipitation.

Figure 1.9: Computer simulation of convective reconfiguration of a circular polar cap patch

(colored region in a) as it convects from the polar cap through the auroral zone. (a) The

initial conditions and the assumed convection model. The contours are equipotential lines

representing the convection pattern as seen in the noncorotating frame. Adapted from Figure

8 of Robinson et al. [1985].

1.6 Instabilities The large scale plasma structures are associated with electron density irregularities which can

be produced by plasma instabilities. There is a number of instability modes in the high-

15

latitude ionosphere. However, in this section, I describe the Kelvin-Helmholtz Instability

(KHI) and Gradient Drift Instability (GDI) which are believed to be the two dominant

processes in the high-latitude F region ionosphere. In addition to the two modes, the current

convective instability (CCI) is also described since it is very relevant to the GDI process.

1.6.1 Kelvin-Helmholtz Instability

KHI plays an important role in understanding the auroral arc dynamics [Keskinen et al., 1988;

Lyons and Walterscheid, 1985]. Keskinen et al. [1988] studied the nonlinear evolution of the

KHI which included the ionospheric collisional effects with neutrals. Figure 1.10 shows the

linear growth rate of different parameters. L is the characteristic scale length of the velocity

shear. The upper curve shows the classic KHI growth rate (without density gradient) which

shows a maximum value of at kL = 0.44 when the linear growth rate

maximizes. The growth rate is reduced when the collision is considered (see lower curves).

Figure 1.10: The linear growth rates of the Kelvin-Helmholtz instability versus wave number

k. The "classic" (no density gradient) case and four collisional cases (ν = 0.0, 0.10, 0.34,

0.50) with a density jump of 3:1 are shown by different curves. (This is from Figure 3 of

Keskinen et al. [1988]).

16

Figure 1.11: Numerical simulations of the collisional Kelvin-Helmholtz instability for ν = 0.1

at 4 instants: 0.005t, 2.56t, 4.65t, and 6.62t, where t = 8.4L/Vo. (This is from Figure 6 of

Keskinen et al. [1988]).

Figure 1.11 displays the result of the simulated density contours by Keskinen et al. [1988].

The linear phase is similar to the non-collisional case (not shown); however, the Pedersen

conductivity acts to prevent curling of the density contours. Small scale structures are

developed in the last panel of Figure 1.11.

1.6.2 Gradient Drift Instability

The GDI is schematically described in Figure 1.12. The X axis is pointing to the top, y to the

left and the Z axis is out of the paper. The background electric field E0 is in the Y direction

and the magnetic field B is in the negative Z direction. The high density region is shaded in

grey. In this scenario, the plasma is moving in the negative X direction (to the bottom). On the

trailing edge (top side), the density gradient is in the negative X direction, and it is opposite

on the leading edge. We consider the trailing edge first. When the density profile is slightly

perturbed, the pattern associated with the ion perturbation will drift in the Pedersen direction

to the left (represented by the solid curve), leaving the highly magnetized electrons (dashed

17

curve) behind them. This results in a charge separation and alternating polarization electric

field Ep as shown on the top of Figure 1.12. The resulting Ep × B (shown by blue arrows) is in

a direction such that the initial perturbation is amplified by moving dense plasma to less dense

plasma and vice versa. Irregularities grow along the existing mean gradient. While on the

leading edge (bottom side), the perturbed ions move in the Pedersen direction to the left with

respect to the electrons as well. However, the polarization electric field Ep was set up which

results in Ep × B (shown by blue arrows) that damps the initial density perturbation.

Therefore, the leading edge is stabilized by the GDI.

Figure 1.12: Illustration of the F-region gradient drift instability. The solid line represents

the ion density contour, while the dashed line represents the electron density contour. The

scheme represents a coordinate system drifting relative to the neutrals [Tsunoda, 1988].

In the simplest one dimensional case, the linear growth rate of GDI is [e.g., Tsunoda, 1988]:

Where is the “slip” velocity, i.e., plasma drift relative to the neutral gas, and

is the gradient scale length.

18

The analytical and numerical simulation techniques have been used to study the evolution of

density irregularities due to the GDI [Keskinen and Ossakow, 1982, 1983; Sojka et al., 2000;

Gondarenko et al., 2003]. Gondarenko and Guzdar [2004] simulated the 3D nonlinear

evolution of the GDI in structuring plasma patches. Figure 1.13A shows the coordinate of the

simulation, with X pointing toward the Sun, and Y towards dusk. The yellow-turquoise box

represents the initial state of the polar cap patch in 400 km×50 km size. The convection flow

is constant anti-sunward in 1 km/s.

Figure 1.13: The simulation of the gradient drift instability. (A) Coordinate and geometry of

polar cap patches. (B) Density contour at (a) t = 0.44 hour, (b) t = 0.8 hour, and (c) t = 1.5

hours. (C) (a) 500 km segment of data observed by Kivanc and Heelis [1997] using DE 2

satellite and density lineouts of (b,c) simulations for an arbitrary cut in the Y direction. DE 2

satellite was moving in the noon-midnight direction. Adapted from Gondarenko and Guzdar

[2004].

Figures 1.13B displays the density evolution in the X-Y plane near the peak of the density

profile in the Z direction. Panel a) shows the presence of small scale irregularities on the

trailing edge of the patch during the early stage (0.44 hour). As time progresses, the

19

fluctuations evolve to a larger scale length at 0.8 hour which is the inverse cascade nature of

GDI. In panel c), the instability reaches the leading edge and the patch becomes fully

structured. Note that even though in the late phase of the nonlinear evolution, the patch is not

disintegrated. Figure 1.13C shows the comparison between the density measurements from

DE 2 satellite [Kivanc and Heelis, 1997] and the density lineout along the midnight-noon for

an arbitrary cut in Y at the same time (in Figure 1.13Bc). Panels a) - b) show similar fully

structured patch. In panel c), the leading edge is steeper and is unstructured. Even at the same

instant, the lineouts at different y shows different structures which indicates that there is no

precise way of a comparison between simulations and observations.

1.6.3 Current Convective Instability

20

Figure 1.14: The conceptual picture of the current convective instability. (a) The geometry is

stable when there is no field aligned current. (b) The geometry becomes unstable due to the

field aligned current.

As shown in Figure 1.12, on the leading edge of the polar cap patch, the density gradient ( n)

is anti-parallel to the E × B drift direction, and the geometry is stable for the GDI. However,

if a field aligned current is present, the stable geometry may become unstable. This kind of

instability is generally called the current convection instability (CCI) [Ossakow and

Chaturvedi, 1979; Chaturvedi and Ossakow, 1981].

The CCI is similar to the GDI in which the polarized electric field is set up in the direction

perpendicular to the density gradient. The conceptual picture of CCI is shown in Figure 1.14.

In the Y-Z plane, the density gradient is going into the paper. The ambient magnetic field (B)

is pointing downward (negative Z direction), and the horizontal electric field E0

(perpendicular to B) is to the left. When the plasma density contour is slightly perturbed, the

ions (solid curve in Figure 1.14a) Pedersen drift to the left relative to the electrons (dashed

curve in Figure 1.14a). This causes charge separation which induces the electric field

perturbation (E’p). However, in this geometry, the corresponding E’p × B will damp the

amplitude of the initial density perturbation. This indicates the stabilization of this geometry

in the GDI scenario.

Figure 1.14b shows that the geometry becomes unstable in the presence of a field aligned

current (j ) which is antiparallel to B. Similar to the previous analysis, we assume a density

perturbation along the wave vector k which has a finite parallel component (k ) to the ambient

magnetic field. The upward j implies that the relative drift between ions and electrons is anti-

parallel to B. When projected on k, the relative motion gives charge separation and polarized

electric field E’’p. E’’p causes the density perturbation to grow in amplitude. However, as

shown in Figure 1.14a, the Pedersen motion due to the horizontal electric field E0 acts to

stabilize the gradient. If the relative motion projected on k is dominated by j rather than E0 ,

then the geometry is unstable. Therefore the instability criterion is

, where and and are the ion neutral collision frequency and

ion gyrofrequency, respectively. The maximum growth rate for the CCI is:

21

With (assuming j = 8 μA/m2, and the density n = 1011 m-3) and L = 50 km,

this yields 3.5×10-3 second-1 (growth time = 4.76 minutes).

Similar to the GDI, the CCI also shows asymmetry in destabilization, i.e., reversing the

direction of the field aligned current (downward) results in stable plasma condition. Note that

refers to the field aligned current carried by the (ionospheric) thermal plasma instead of the

precipitating particles. The precipitating electron beam has a negligible effect on the growth

rate of the CCI [Chaturvedi and Ossakow, 1983]. To conclude, the field aligned current can

destabilize the plasma configuration that is stable to the GDI or to enhance the growth rate of

an already unstable situation. However, later studies [Huba, 1984; Keskinen and Ossakow,

1982; Satyanarayana and Ossakow, 1984; Satyanarayana et al., 1985; Huba and Chaturvedi,

1986] imply that the CCI rarely overcomes the stabilizing effects of an unfavorable GDI. For

detailed information about the limit of CCI, see the review paper by Tsunoda [1988] and

Chapter 10 of Kelley [2009].

1.7 Ionospheric Scintillation The ionospheric scintillation was one of the first known space weather phenomena. Hey et al.

[1946] observed short-period of irregular fluctuations in the radio wave intensity (64 MHz)

from the radio star Cygnus, and subsequent observations indicated that this phenomenon was

locally produced in the Earth’s ionosphere, which was later termed as ionospheric

scintillation. The ionospheric scintillations are often defined as rapid fluctuations in the

received amplitude and phase of radio waves that pass through the ionosphere [Yeh and Liu,

1982; Kintner et al., 2007]. In the case of GPS, scintillations may reduce the accuracy of the

pseudo-range and phase measurements and thus degrade the positioning service. At times the

amplitude scintillation may be so intense that the signal power drops below the receiving

threshold, the receiver loses lock to the signal and the positioning service is not possible

[Garner et al., 2011]. Strong phase scintillation events can also lead GPS receivers to losses of

phase lock, which result in cycle slips.

Scintillation studies are important due to two reasons: the practical usage and the scientific

interest. For pragmatic purposes, it is important to enable the predication of the degradation of

the positioning and communication systems. On the other hand, the scintillation study can be

used to investigate the behavior of electron density irregularities in the ionosphere.

22

1.7.1 Scintillation Theory

Scintillation theory relates the observed signal statistics to the statistics of ionospheric

electron density fluctuations [for reviews, interested readers can refer to Yeh and Liu, 1982;

Bhattacharyya et al., 1992; Yeh and Wernik, 1993; Basu and Basu, 1993]. As an illustration,

Figure 1.15 demonstrates the spatial irregularity spectrum with the temporal spectra of

amplitude and phase scintillation for weak scatter. One can see that slopes of the high-

frequency asymptotes of the ionospheric irregularity and scintillation spectra coincide.

Knowledge of the irregularity spectrum is important for understanding the physical

mechanisms of formation and evolution of irregularities [Wernik et al., 2003; Wernik et al.,

2007; Basu et al., 1988a, 1991].

The problem of scintillations is basically to solve the propagation of radio waves in a random

media. Several scintillation theories have been developed [see e.g., Lovelace, 1970; Yeh and

Liu, 1982 for reviews]. The phase screen model is the most popular scintillation theory

[Booker et al., 1950; Rino, 1979a, b; Carrano et al., 2012]. The phase screen model assumes

that the ionospheric irregularities are within a thin layer which is called a phase screen. For

weak scattering condition, only the phase is affected by the random fluctuations in the

refractive index when the radio wave propagates through the irregularity layer. As the wave

propagates to the ground, the perturbed wave front will set up an interference pattern which

results in amplitude fluctuations according to Huygens’s principle.

Figure 1.15: The irregularity spectrum and the corresponding amplitude and phase

scintillation spectra for the weak scatter. (This is from Figure 1 of Wernik et al. [2003]).

23

The spectra of the radio wave intensity and phase deviations are [Kintner et al., 2007]:

where q is the horizontal wave number of the phase fluctuations across the screen, is

the power spectrum of the irregularity density, and are the Fourier transform of

the intensity and phase autocorrelation function, and is the first Fresnel radius,

where λ is the wavelength of the incident signal, and d is the distance from the phase screen to

the receiver.

The term is known as the Fresnel filtering function which provides an upper limit

on the scale size of the irregularities. The upper limit of the scale size, or the Fresnel radius

F, occurs where the sin2 term goes to unity, or when the argument is equal to

rad. From , we get the first Fresnel radius of ~360 m for GPS L1 signals,

assuming an irregularity layer altitude of 350 km ( ) and a signal ray path

elevation of 90°.

Unlike the amplitude scintillations, the phase scintillations have a maximum at q = 0 due to

the cos2 term. The next local maxima are when the argument becomes nπ rad. Because the

one-dimensional phase spectrum at the phase screen typically has the form

where n is of order 2, the majority of the phase fluctuation power is found at small q.

Note that the above formulae which relate the spectra of the radio wave with the spectra of the

electron density fluctuation can be applied when considering one dimensional weak

scattering. For strong scattering, both amplitude and phase of the incident plane wave are

altered when crossing the phase screen. Also the phase screen may only be approximated as

one-dimensional in specific conditions such as near the geomagnetic equator for signals from

high-elevation satellites. Consideration of these more complex environments can be found in

the work of Rino and Fremouw [1977] and Rino [1979a, b].

24

1.7.2 Observations of Ionospheric Scintillations

Early studies made use of radio stars to study the amplitude fluctuations at HF (High

Frequency). With the advent of satellite beacons (e.g., Sputnik, ATS, WIDEBAND, and

HILAT), surveys were expanded globally and long-term measurements were also made [see

e.g., Basu et al., 1988b]. For studies using satellite beacons, see e.g., Aarons [1997], Aarons

[1982], Yeh and Liu [1982], Wernik et al. [2003]. These previous studies can be summarized

in Figure 1.16 [Basu and Groves, 2001]. The diagram shows the global scintillation activity

during solar maximum (left) and solar minimum (right). Scintillations are weaker during solar

minimum. The diagram highlights several ionospheric scintillation regions: the region around

the magnetic equator during post-sunset time, the nightside auroral oval and dayside cusp, and

the region within the polar cap at all local times.

Figure 1.16: Global morphology of the ionospheric scintillation activity (fades) during the solar maximum and solar minimum [Basu and Groves, 2001].

Later studies took advantage of the GNSS satellites, which have greatly advanced the

understanding of scintillations. In the next section, I constrain the review of the literature to

GPS scintillations at high latitudes since this is the subject of this thesis.

1.8 Previous GPS Scintillation Studies in the Auroral/Polar Cap Region The ionospheric scintillation is strongly dependent on the frequency of trans-ionospheric

signals, being lower scintillation strength when the frequency goes higher [see e.g.,

Bhattacharyya et al., 1990]. The GPS scintillation was usually conducted by using L1

25

(1.57542 GHz), and therefore, the amplitude scintillation is less pronounced than those with

radio stars and polar beacon satellites, especially in the polar region.

Figure 1.17: Detailed comparison of the amplitude scintillation (S4) and the GPS TEC for the

period from 23 UT February 3 to 05 UT February 4, 1984. The GPS data were observed in

Thule, Greenland (86° CGLat). (This is from Figure 4 of Weber et al. [1986]).

Weber et al. [1986] was the first to study the GPS scintillation effect of the F region polar cap

patches. They directly observed polar cap patches drift in the anti-sunward direction over a

distance of ~3000 km from the central polar cap to the poleward edge of the auroral oval. The

GPS TEC (total electron content) increased by 10-15 TECU for the polar cap patches over the

background value of 5 TECU. The amplitude scintillation measurements indicated the

presence of ionospheric irregularities throughout the entire patch. The detailed comparison of

TEC with S4 is shown in Figure 1.17. Although the amplitude scintillation is very low (S4 ≤

0.2), the variations which indicate the internal structures of patches are clearly seen. Note that

some of the TEC variations (marked by “1” – “10”) are likely substructures of patches. For

“patches” 3, 7, and 8, the amplitude scintillation increases are presented on the trailing edge,

while the others are more symmetric. Assuming that the plasma is moving in the anti-sunward

direction faster than the neutral atmosphere (Vi > Vn), the trailing edge of the patch is

unstable to the GDI [Linson and Workman, 1970]. The trailing edge may then be expected to

26

be characterized by more intense irregularities. Figure 1.17 clearly shows, however, that

irregularities also exist on the leading edge (e.g., “patch” 2) and internal structures inside the

patches was shown by Buchau et al. [1985] using digital ionosonde measurements.

Later on, the specialized GPS Ionospheric Scintillation and TEC monitor (GISTM) was

invented and deployed in the Arctic regions [Van Dierendonck et al., 1993; Mitchell et al.,

2005; Alfonsi et al., 2008; Jayachandran et al., 2009; Yin et al., 2009; Li et al., 2010]. More

case and statistical studies have been activated which include both amplitude and phase

scintillations. Mitchell et al. [2005] reported a case during the Halloween storm on 30 October

2003, where both amplitude (S4 up to 0.12) and phase scintillations (σϕ up to 0.22 rad) were

collocated with strong TEC gradients at the edge of a localized plasma enhancement which

was likely a polar cap patch. The observation is summarized in Figure 1.18. Mitchell et al.

[2005] proposed that the GDI is the mechanism for the generation of the irregularities which

cause scintillations at the L band frequency. However, it seems the scintillations occurred in

the leading edge of the density enhancement since it is accompanying with an increase of the

TEC variation. This is contradictory to the prediction of GDI which only occurs on the

trailing edge of patches (see section 1.6.2).

Figure 1.18: The vertical TEC, phase (σϕ) and amplitude scintillation indices (S4) from the GPS satellite PRN31 recorded by the GISTM in Ny-Ålesund, Svalbard between 21:00 and 22:00 UT, on 30 October, 2003 [Mitchell et al., 2005].

De Franceschi et al. [2008] further investigated the dynamics of the ionospheric plasma

during the Halloween storms by a new release of the multi-instrument data analysis system

(MIDAS). They were able to compare the large scale ionospheric structures (e.g., polar cap

27

patches) with the scintillation data observed by a chain of GPS scintillation receivers in

Northern Europe. They also found that the amplitude and phase scintillation maxima were

collocated with the TEC gradient near the edges of plasma patches.

De Franceschi et al. [2008] noticed that the scintillation maxima were on the leading edge of

the patches (the same as Mitchell et al. [2005]), which seems to contradict with the GDI. It

was suggested to be a likely effect of neutral winds, since the unstable condition of GDI

requires the ion velocity relative to the neutral is parallel to the density gradient [Tsunoda,

1988; Basu et al., 1990; Kivanc and Heelis, 1997; Coley and Heelis, 1998]. Another

explanation is the nonlinear evolution of the GDI. As shown in section 1.6.2, the GDI

originating at the trailing edge of patches spreads over the whole patch in the later phase of

the nonlinear evolution [Gondarenko and Guzdar, 2004]. Eventually the irregularities level at

the trailing and leading edges become comparable. However, this also results in small scale

irregularities throughout the patches, which was not seen in De Franceschi et al. [2008].

In addition to case studies of GPS scintillations which focused more on the physical

mechanisms of irregularities, several works have been conducted on the climatology of

scintillations [see e.g., Spogli et al., 2009; Alfonsi et al., 2011; Li et al., 2010; Prikryl et al.,

2011]. The high-latitude ionospheric scintillation depends on many variables that include

local time, season, magnetic latitude (MLat), magnetic local time (MLT), solar activity, and

geomagnetic activity. Next I review literatures which studied the GPS scintillation statistics

over North Europe which is particularly relevant for this PhD project.

Spogli et al. [2009] constructed “scintillation maps” as a function of MLT and MLat. Four

GPS scintillation receivers which cover MLat between 40° and 84° during the period October,

November, and December in 2003 were used to characterize the scintillation maps and were

compared with the modelled auroral oval. Figure 1.19 shows the distribution of the

occurrence rate of phase (σϕ) and amplitude (S4) scintillations. The percentage of occurrence

is evaluated for each bin of 3 h MLT×2˚ MLat.

Similar to the previous findings, two prominent scintillation regions are clear, one around

magnetic noon, the other around magnetic midnight. Phase scintillations are also seen inside

the polar cap (poleward of the auroral oval) which may be attributed to the polar cap patches

and/or sun-aligned arcs. More strikingly, the phase scintillation occurrence around midnight

favors the pre-midnight sector, which is similar to the MLT distribution of polar cap patches

28

exiting the polar cap [Moen et al., 2007]. When dividing the data into disturbed and quiet time

by the Kp index, the scintillation occurrence is generally larger and shifted to lower MLat

during disturbed days. This was explained by the expanded auroral zone during disturbed

condition [Spogli et al., 2009].

Figure 1.19: (left) Scintillation maps showing the occurrence percentage of σϕ > 0.25 rad

and (right) the occurrence percentage of S4 > 0.25. Note that the two color scales are

different. The grey (red) curves present the Feldstein oval for IQ = 3 (6). (This is Figure 4

from Spogli et al. [2009]).

The amplitude scintillations are much less frequently observed than the phase scintillations.

The amplitude scintillation only occurs in confined regions along the auroral oval around

noon and midnight and mainly came from the disturbed time. Note the amplitude scintillation

occurrence rate is below 0.3%.

Moen et al. [2013] used observations from Ny-Ålesund during the same period as Spogli et al.

[2009] to investigate the IMF dependence of scintillation climatology. Figure 1.20 is

reproduced from Moen et al. [2013]. The black and red curves in Figure 1.20 are auroral ovals

from the Feldstein model for IQ = 3 (quiet) and IQ = 6 (highly disturbed) conditions,

respectively [Holzworth and Meng, 1975]. The morphological picture reveals, as reported by

Spogli et al. [2009], that even under extremely disturbed conditions (such as during the

Halloween storms) moderate/strong amplitude scintillation (Figure 1.20b) occurs in the polar

cap at a very low occurrence rate, except in a few isolated sectors [Moen et al., 2013]. This is

independent of the IMF orientation (Note the different color scale and mesh grid in Figures

1.19 and 1.20). This indicates that, on average, the fragmentation of the tongue of ionization

(TOI) around noon does not result in amplitude scintillations exceeding the threshold of S4 >

29

0.25 [Moen et al., 2013]. The only sites where amplitude scintillation may become significant

are near the boundaries of the auroral oval, and particularly in the pre-midnight MLT sector

[Moen et al., 2013]. Concerning phase scintillations (Figure 1.20a), under southward IMF

conditions, magnetopause reconnection leads to soft particle precipitation in the cusp region

and is associated with larger occurrence of phase scintillation than under northward IMF

conditions [Moen et al., 2013]. It is interesting to note how the cusp footprint is reproduced in

the phase scintillation occurrence; the spread and the equatorward shift of the cusp are evident

in the phase scintillation occurrence when Bz is negative, while the cusp signature occurs at

higher latitudes when Bz is northward (Type 2 cusp) [Moen et al., 2013]. Since TOI/polar cap

patches are not expected for northward IMF, the presence of cusp scintillations for northward

IMF indicates that particle precipitation plays a central role in irregularity formation [Moen et

al., 2013].

Figure 1.20: The GPS scintillation climatology observed between October and December

2003 over Ny-Ålesund obtained by binning the data for conditions of IMF Bz > 0 (left

column) and Bz < 0 (right column). Black and red curves present the Feldstein model of the

auroral oval for IQ = 3 and IQ = 6, respectively. (This is Figure 8 from Moen et al. [2013]).

Li et al. [2010] statistically investigated seasonal variation as well as IMF and MLT

dependence of the scintillation occurrence rate by using data from two conjugate stations in

30

Ny-Ålesund (78.9˚N, 11.9˚E; 75.8˚N corrected geomagnetic latitude (CGMLat)) and

Larsemann Hills (69.4˚S, 76.4˚E; 74.6˚S CGMLat). They found that the occurrence level of

the GPS phase scintillation is significantly higher than that for the amplitude scintillation. The

occurrence rate of the GPS amplitude scintillation rarely exceeded 1%. The phase scintillation

activity in the Northern hemisphere mainly takes place in October, November and December,

whilst in the Southern hemisphere, it mainly occurs during May and June. This suggests that

the ionospheric scintillation at high latitudes favors the local winter months when there are

little or no sunlight [Li et al., 2010].

31

2 Instrumentation and Methodology In this chapter, I give an overview of two key instruments, the all-sky imager and the GPS

scintillation receiver, which we used as the experimental basis for this project.

2.1 All-Sky Imager The all-sky imager (also known as all-sky camera, all-sky imaging photometer) is a camera

system which is used to monitor the whole sky. By using the fish-eye lens, the all-sky imager

can have a field of view of 180°.

Table 2.1: Locations of the UiO all-sky imagers.

Name Location Latitude Longitude Magnetic

Latitude

Magnetic

Longitude Operation

Time

NYA2

Ny-Ålesund 78.92°N 11.93°E 76.24°N 110.19°E

1997-2012

NYA4 2010-2014

NYA6 2014-Now

LYR1 Longyearbyen 78.15°N 16.04°E 75.24°N 111.21°E 1996-2013

LYR5 2013-Now

AND2 Andøya 69.15°N 16.03°E 66.33°N 99.88°E 2002-2015

SKN4 Skibotn 69.43°N 20.38°E 66.28°N 103.41°E 2014-Now

The University of Oslo (UiO) has several all-sky imagers that are operated in Northern

Norway and Svalbard. The location of each station is shown in Table 2.1. In Table 2.1, the

first three characters in the name indicate the location of the all-sky imager, while the number

at the end indicates the number of the imager. For example, NYA4 indicates the all-sky

imager #4 is located in Ny-Ålesund, while SKN4 indicates the same all-sky imager is

operated in Skibotn. This is because the all-sky imager #4 was moved from Ny-Ålesund to

Skibotn in October 2014 and has been operated there since. Figure 2.1 displays the station

32

locations and the field of view of the corresponding all-sky imagers. Note that when talking

about the geographic coverage, the field of view is projected to a certain altitude. Usually, the

630.0 nm images are projected onto an altitude of 250 km, while the 557.7 nm images are

projected onto the altitude of 120 km [e.g., Sandholt et al., 2002]. The all-sky images are

often cut at the zenith angle of 75° because of the distortion at low elevation angles.

As the emission intensity at 630.0 nm of polar cap patches is low (50 R – 800 R), highly

sensitive cameras are needed to track the airglow patches during polar night. NYA4 has a

resolution of ~10 R and is able to track the airglow emission of polar cap patches. We have

mainly used data from NYA4 in this project.

Figure 2.1: The coverage of four all-sky imagers when projected onto the geographic

coordinate. The red circles indicate the coverage of each all-sky imager at 630.0 nm at 250

km altitude when cut at a zenith angle of 75°, while the green circles show the coverage of

each all-sky imager at 557.7 nm at 120 km altitude.

NYA4 is also called Hyper All Sky Imager 1 (HASI 1). The hardware of NYA4 is shown in

Figure 2.2. NYA4 was delivered from Keo Scientific Ltd. in 2010. The front optical system of

33

NYA4 is a Mamiya RB67 37mm/F4.5 medium-format achromatic fish-eye lens1. The final

lens in front of the CCD (charged coupled device) is an ultra-fast f/0.95 Canon 50 mm

compound Rangefinder objective. An EMCCD (electron multiplying charged coupled device)

is placed at the end of the camera system which is deep-cooled (down to -70°C) using a

thermoelectric cooler.

Figure 2.2: The NYA4 all-sky imager (left) and its control box (right).

Figure 2.3: Illustration of the normal operational sequence for the NYA4 all-sky imager

(Courtesy of Bjørn Lybekk). The all-sky imager captures 2 images at 630.0 nm (red) and 4

images at 557.7 nm (green). The integration and idle time are shown by numbers in seconds. 1 An achromatic Lens transmits light without separating it into constituent colors.

34

There are six filters in the filter wheel which are shown in Table 2.2. However, only two

filters are used in the normal campaign when NYA4 takes two images at 630.0 nm and four

images at 557.7 nm in one minute as shown in Figure 2.3. The exposure time is 2 seconds at

630.0 nm and 1 second at 557.7 nm, respectively. The exposure time and EMCCD gain can

be changed from the controlling software.

Table 2.2: Six filters in the NYA4 all-sky imager system.

Filter 1 Filter 2 Filter 3 Filter 4 Filter 5 Filter 6

630.0 nm 557.7 nm 427.8 nm 777.4 nm Open Blocked

Figure 2.4: (left) Schematic drawing of a pin-hole camera. The focal length is f, θ is the zenith angle of the incident ray and r is the distance from the center of the image. (right) different projection method of the lens.

Table 2.3: Different projections used in fish-eye lens design [KEO Consultants, 1993].

Gnomical (pin-hole)

Equi-distant (linear)

Orthographic

Equi-solid angle (Equal area)

35

Stereographic

The optical lens projects the incident light from a zenith angle θ to the radius r as shown in

Figure 2.4. Table 2.3 shows the common projection methods. The relation between radius and

incident zenith angle is also shown in the right panel of Figure 2.4 [Gulbrandsen, 2006]. The

projection method of NYA4 is close to the equi-solid angle projection.

2.2 GPS Ionospheric Scintillation and TEC Monitor (GISTM)

Figure 2.5: The hardware of the GISTM system. 1), the NovAtel GSV 4004B GPS box; 2), the

NovAtel dual frequency antenna; 3), the antenna cable; 4), the serial cable; 5), the power

cable; 6), the personal computer [Carrano and Groves, 2009].

One of the most popular types of GPS scintillation receiver is the NovAtel GSV4004B [Van

Dierendonck et al., 1993; Shanmugam et al., 2012]. The hardware of the GISTM system is

shown in Figure 2.5. The receiver can provide TEC (total electron content) data up to 1 Hz

based on dual-frequency measurements (L1 at 1.57542 GHz and L2 at 1.22760 GHz). The

36

phase (σϕ) and amplitude (S4) scintillation indices at L1 frequency are computed by the GPS

box from the raw data in 50 Hz and recorded automatically to the computer. We also record

the raw carrier phase and intensity data at a 50 Hz sampling rate. The raw carrier phase data

have a noise bandwidth of 15 Hz [Van Dierendonck et al., 1993].

2.2.1 The GPS Phase Scintillation Index

The phase scintillation index is defined as the standard deviation of the carrier phase that has

been detrended by the high-pass sixth-order Butterworth filter with a cutoff frequency of 0.1

Hz (Note that the GISTM allows the user to specify a different cutoff frequency during

installation):

,

where < . > denotes the expected value, and ϕ is the detrended carrier phase. The standard

deviation is computed over periods of 1 second, 3 seconds, 10 seconds, 30 seconds and 60

seconds. Thus, for every minute, 5 values (1second, 3 seconds, 10 seconds, 30 seconds and 60

seconds ) are stored in the ISMRB data log along with the time tag (in week number and

time of week). The σϕ of 60 seconds is often used as the GPS phase scintillation indices.

2.2.2 The GPS Amplitude Scintillation Index

The amplitude scintillation index (S4) is defined as the standard deviation of the received

signal power, based on a 50-Hz sampling rate, normalized to the average signal power over

one minute periods. The total S4, including ambient noise effects, is:

Where < . > denotes the expected value over 60 seconds, and P stands for the received signal

power. The correction part of the S4 value that is due to ambient noise can be written as:

is the signal-to-noise density. The correction can be subtracted from the total S4 value.

37

If the argument under the square root becomes negative, S4 is set to zero.

The amplitude scintillation was found to be very rare at high latitudes [Spogli et al., 2009; Li

et al., 2010; Prikryl et al., 2010; Moen et al., 2013], therefore we focus on the GPS phase

scintillation in our studies.

38

39

3 Summary of Papers

3.1 Paper Abstracts Paper 1: Jin, Y., J. I. Moen, and W. J. Miloch (2015), On the collocation of the cusp aurora

and the GPS phase scintillation: A statistical study, J. Geophys. Res. Space Physics, 120,

doi:10.1002/2015JA021449.

The climatology map of the GPS phase scintillation identifies two regions of high scintillation

occurrences: around magnetic noon and around magnetic midnight. The scintillation

occurrence rate is higher around noon, while the scintillation level is stronger around

magnetic midnight. This paper focuses on the dayside scintillation region. In order to resolve

the role of the cusp auroral processes in the production of irregularities, we put the GPS phase

scintillation in the context of the observed auroral morphology. Results show that the

occurrence rate of the GPS phase scintillation is highest inside the auroral cusp, regardless of

the scintillation strength and the interplanetary magnetic field (IMF). On average, the

scintillation occurrence rate in the cusp region is about 5 times as high as in the region

immediately poleward of it. The scintillation occurrence rate is higher when the IMF Bz is

negative. When partitioning the scintillation data by the IMF By, the distribution of the

scintillation occurrence rate around magnetic noon is similar to that of the poleward moving

auroral form (PMAF): there is a higher occurrence rate at earlier (later) magnetic local time

when the IMF By is positive (negative). This indicates that the irregularities which give rise to

scintillations follow the IMF By-controlled east-west motion of the aurora and plasma.

Furthermore, the scintillation occurrence rate is higher when IMF By is positive when the

cusp is shifted toward the post noon sector where it may get easier access to the higher

density plasma. This suggests that the combined auroral activities (e.g., PMAF) and the

density of the intake solar EUV ionized plasma are crucial for the production of scintillations.

Paper 2: Jin, Y., J. I. Moen, and W. J. Miloch (2014), GPS scintillation effects associated

with polar cap patches and substorm auroral activity: Direct comparison, J. Space Weather

Space Clim., 4, A23, doi:10.1051/swsc/2014019.

We directly compare the relative GPS scintillation levels associated with regions of enhanced

plasma irregularities called auroral arcs, polar cap patches, and auroral blobs that frequently

40

occur in the polar ionosphere. On January 13, 2013 from Ny-Ålesund, several polar cap

patches were observed to exit the polar cap into the auroral oval, and were then termed auroral

blobs. This gave us an unprecedented opportunity to compare the relative scintillation levels

associated with these three phenomena. The blobs were associated with the strongest phase

scintillation (σϕ), followed by patches and arcs, with σϕ up to 0.6, 0.5, and 0.1 rad,

respectively. Our observations indicate that most patches in the nightside polar cap have

produced significant scintillations, but not all of them. Since the blobs are formed after

patches merged into auroral regions, in space weather predictions of GPS scintillations, it will

be important to enable predictions of patches exiting the polar cap.

Paper 3: Jin, Y., J. I. Moen, W. J. Miloch, L. B. N. Clausen, and K. Oksavik (2016),

Statistical study of the GNSS phase scintillation associated with two types of auroral blobs, J.

Geophys. Res. Space Physics, 121, doi:10.1002/2016JA022613.

This study surveys space weather effects on GNSS (Global Navigation Satellite System)

signals in the night time auroral and polar cap ionosphere using scintillation receivers, all-sky

imagers, and the EISCAT Svalbard radar. We differentiate between two types of auroral

blobs: blob type-1 (BT-1) which are formed when islands of high density F region plasma

(polar cap patches) enter the nightside auroral oval, and blob type-2 (BT-2) which are

generated locally in the auroral oval by intense particle precipitation. For BT-1 blobs we have

studied 41.4 hours of data between November of 2010 and February of 2014. We find that

BT-1 blobs have significantly higher scintillation levels than their corresponding polar cap

patch; however, there is no clear relationship between the scintillation levels of the

preexisting polar cap patch and the resulting BT-1 blob. For BT-2 blobs we found that they

are associated with much weaker scintillations than BT-1 blobs, based on 20 hours of data.

Compared to patches and BT-2 blobs, the significantly higher scintillation level for BT-1

blobs implies that auroral dynamics plays an important role in structuring of BT-1 blobs.

3.2 Summary The intense scintillations occur at the dayside around the cusp and at nightside around

midnight at high latitudes which is consistent with the inflow and exit of polar cap patches.

The scintillation can also be found in the polar cap at all local time which is also tentatively

attributed to the contribution of polar cap patches. Therefore, we are motivated by the

41

coincidence of high scintillation regions and the production and exiting of polar cap patches.

Our study focuses on the strongest scintillation regions around noon (where the dayside high

density plasma is brought into the polar cap), and around magnetic midnight (where polar cap

patches are pulled out of the polar cap).

Figure 3.1: MLAT-MLT maps show the GPS phase scintillation occurrence rate for

observations from year 2010 to 2013 in Ny-Ålesund binned by σϕ from (a) (0.1, 0.25) rad, (b)

(0.25, 0.5) rad, and (c) ≥0.5 rad. The MLAT 70° and 80° and MLT 00, 06, 12, and 18 are

marked in each panel. Note that the occurrence rate color bar is different in each panel. (d

and e) The standard deviation (std) and mean value of σϕ, respectively. In each panel, the

solid red circles denote the auroral oval calculated with the Feldstein model for IQ = 3 and

the blue circle shows the MLAT of Ny-Ålesund station.

As a reference, we reproduced the climatology map of the GPS phase scintillation at high

latitudes in Figure 3.1 by using the data from 2010 to 2013. Figure 3.1 shows similar feature

as the previous climatology work: the scintillation occurrence rate being higher around

magnetic noon and magnetic midnight. The scintillations can also be observed in the polar

42

cap. In addition to the previous finding, we compare different scintillation levels (weak,

intermediate and strong) and find that weak scintillations occur more frequently around noon,

while strong scintillations are more prone to occur around midnight.

Figure 3.2: The distribution of the scintillation occurrence rate at (a) sub-aurora, (b) cusp,

and (c) polar cap regions. The red bars show the total scintillation occurrence rates (σϕ ≥ 0.1

rad).

The climatology map indicates that the statistical scintillation region on the dayside is offset

poleward of the modelled auroral oval (red circles in Figure 3.1). However, these studies

compared the statistical scintillation distribution with the modelled auroral oval and were

unable to compare the scintillation region with the actual auroral region. It is still an

unresolved problem whether the GPS scintillation is high in the cusp or poleward of it, i.e.,

the poleward offset of the GPS scintillation region could be due to newly formed polar cap

patches or due to the unrealistic auroral oval model. Paper 1 intends to study the scintillation

condition around the dayside cusp aurora, and the key result is shown in Figure 3.2. The

statistical result shows that the GPS phase scintillation is collocated with the cusp auroral

region. The scintillation occurrence rate in the region immediately poleward of the cusp

aurora is much lower. Thus, the Feldstein model is unable to represent the cusp location.

When studying the scintillation with respect to the modelled auroral oval, care should be

taken.

Given that strong scintillations are more frequently observed around midnight, we turn to the

nightside to study the GPS phase scintillation condition in the context of polar cap patches,

substorm, and auroral blobs. To make our discussion more convenient, we define two types of

auroral blobs: blob type 1 (BT-1) is connected to the high-density polar cap patch material

43

that has entered the auroral oval; and blob type-2 (BT-2) relates to the plasma density

enhancement/structures that are generated locally by particle precipitation.

In Paper 2, we present a case study on the GPS phase scintillation at night by using the all-sky

imager and GPS scintillation receiver in Ny-Ålesund. The result indicates that the BT-1 blobs

are associated with the strongest phase scintillation, followed by patches and BT-2 blobs.

Figure 3.3 presents the overview of the event. The BT-2 blobs (produced by local auroral

precipitation) from ~16:30 UT to 17:40 UT did not produce significant GPS phase

scintillations. 5 out of 6 observed polar cap patches (inverted integral signs in yellow) were

associated with significant GPS phase scintillations. When polar cap patch D crossed the

open-closed field line boundary (OCB) and formed the BT-1 blobs, enhanced scintillations

were observed.

Figure 3.3: Event overview. (a) The keogram scanned along the North-South cross section as

a function of scan angle from 15° North to 15° South from the 630.0 nm channel of the all-sky

imager in Ny-Ålesund, January 13, 2013, 15–23 UT. Letters A–F mark six polar cap patches

as they moved from north to south across the all-sky imager. The intensity is color coded

according to the log (counts) scale on the right. (b) Phase scintillation indices recorded by

the GPS scintillation monitor located in Ny-Ålesund, measurements from different satellites

44

are color coded on the right. (c) Mean phase scintillations for all satellites averaged over 10

minutes.

Paper 3 is a comprehensive study of the relative scintillation level with polar cap patches and

BT-1, BT-2 blobs. We present two case studies (BT-1 and BT-2 events, respectively) which

are followed by a statistical study. Results confirm that the BT-1 blob is associated with

higher scintillations than the polar cap patch and aurora in general.

Figure 3.4 summarizes the content of the three papers where Paper 1 focuses on the

scintillation condition at the dayside around the cusp auroral region, Papers 2 and 3 study the

scintillation condition around midnight.

Figure 3.4: Summary of three papers in this thesis.

Paper 1 shows that the GPS phase scintillation is well collocated with the cusp aurora around

local noon, while Papers 2 and 3 shows that the strongest phase scintillations occur in the

auroral region when polar cap patches exit the polar cap: i.e., the three papers reveal that the

highest scintillation regions are inside the auroral oval both on the dayside and nightside. In

45

conclusion, in the European Arctic sector, the highest scintillation occurrence rates appear

inside the auroral region around magnetic noon and midnight. In order to predict the locations

of worst scintillation impacts for the GNSS users, the key is to keep track of the positions of

the cusp aurora around magnetic noon (polar cap inflow region) as well as the substorm

aurora around magnetic midnight (polar cap exit region).

3.3 Future Work The current knowledge (especially the works presented in this thesis) can be used for

prototyping a space weather forecast product. As the final space weather product should be

physics-based, the space weather forecast will be improved gradually as the physics is

understood better and better. In the three papers, it is shown that the most disturbed

scintillation regions are collocated with the auroral oval around magnetic noon and around

magnetic midnight. In addition, the three papers highlight the importance of the interactions

of high density polar cap patches and auroral dynamics. Therefore, in order to predict the

worst scintillation scenario, it is important to predict both the high density polar cap patches

and the locations of auroral zones (i.e., OCB). There are several instruments which can be

used for these predictions. The GPS TEC data are now dense enough to track the motion of

polar cap patches [e.g., Zhang et al., 2013]. However, the problem of time delay for weeks of

the global TEC map makes it impossible to forecast or even nowcast the motion of polar cap

patches. The regional space weather product such as regional TEC map is applicable for

nowcast. For the auroral oval position, several approaches can be used: 1) The magnetometer

chain provides an all-weather condition monitoring of the auroral electrojet which may be

used to monitor the auroral oval; 2) The superDARN network provides a full coverage in the

polar region, and it is possible to track the auroral boundary. Furthermore, the polar cap

patches can also be tracked by the HF radar. 3) The polar orbiting satellites provide auroral

precipitation data, e.g., the OVATION Prime model was developed using energetic particle

measurements from the Defense Meteorological Satellite Program (DMSP) satellites (for

more information, see Newell et al. [2014]).

46

47

4 Bibliography Aarons, J. (1982), Global morphology of ionospheric scintillations, Proc. IEEE, 70(4), 360–378.

Aarons, J. (1997), Global positioning system phase fluctuations at auroral latitudes, J. Geophys. Res., 102(A8), 17219–17231, doi:10.1029/97JA01118.

Alfonsi, L., et al. (2008), Probing the high latitude ionosphere from ground-based observations: The state of current knowledge and capabilities during IPY (2007–2009), J. Atmos. Sol. Terr. Phys., 70(18), 2293–2308, doi:10.1016/j.jastp.2008.06.013.

Alfonsi, L., L. Spogli, G. De Franceschi, V. Romano, M. Aquino, A. Dodson, and C. N. Mitchell (2011), Bipolar climatology of GPS ionospheric scintillation at solar minimum, Radio Sci., 46, RS0D05, doi:10.1029/2010RS004571.

Anderson, D. N., D. T. Decker, and C. E. Valladares (1996), Modeling boundary blobs using time varying convection, Geophys. Res. Lett., 23, 579–582, doi:10.1029/96GL00371.

Basu, S., S. Basu, E. MacKenzie, P. F. Fougere, W. R. Coley, N. C. Maynard, J. D. Winningham, M. Sugiura, W. B. Hanson, and W. R. Hoegy (1988a), Simultaneous density and electric field fluctuation spectra associated with velocity shears in the auroral oval, J. Geophys. Res., 93(A1), 115–136, doi:10.1029/JA093iA01p00115.

Basu, S., E. Mackenzie, and S. Basu (1988b), Ionospheric constraints on Vhf Uhf communications links during solar maximum and minimum periods, Radio Sci., 23(3), 363–378, doi:10.1029/RS023i003p00363.

Basu, S., S. Basu, E. Mackenzie, W. R. Coley, J. R. Sharber, and W. R. Hoegy (1990), Plasma structuring by the gradient drift instability at high-latitudes and comparison with velocity shear driven processes, J. Geophys. Res., 95(A6), 77997818, doi:10.1029/JA095iA06p07799.

Basu, S., S. Basu, E. Costa, C. Bryant, C. E. Valladares, and R. C. Livingston (1991), Interplanetary magnetic field control of drifts and anisotropy of high-latitude irregularities, Radio Sci., 26(4), 1079–1103, doi:10.1029/91RS00586.

Basu Sa. and Su. Basu (1993), Ionospheric structures and scintillation spectra, in Wave Propagation in Random Media (Scintillation), eds. V. I. Tatarski, A. Ishimaru, and V. U. Zavorotny, pp. 139-153, The International Society for Optical Engineering, Bellingham, WA, USA.

Basu, S. and Groves, K. M. (2001) Specification and Forecasting of Outages on Satellite Communication and Navigation Systems, in Space Weather (eds P. Song, H. J. Singer and G. L. Siscoe), American Geophysical Union, Washington, D. C.. doi: 10.1029/GM125p0423.

Bhattacharyya, A., K. C. Yeh, and S. J. Franke (1992), Deducting turbulence parameters from trans-ionospheric scintillation measurements, Space Sci. Rev., 61, 335-386.

Bhattacharyya, A., R. G. Rastogi, and K. C. Yeh (1990), Signal frequency dependence of ionospheric amplitude scintillations, Radio Sci., 25(4), 289–297, doi:10.1029/RS025i004p00289.

Booker, H. G., J. A. Ratcliffe, and D. H. Shinn (1950), Diffraction froman irregular screen with applications to ionospheric problems, Phil.Trans. R. Soc. London, Ser. A, 856, 579-- 609.

Buchau, J., B. W. Reinisch, E. J. Weber, and J. G. Moore (1983), Structure and dynamics of the winter polar cap F region, Radio Sci., 18(6), 995–1010, doi:10.1029/RS018i006p00995.

48

Buchau, J., E. J. Weber, D. N. Anderson, H. C. Carlson, J. G. Moore, B. W. Reinisch, and R. C. Livingston (1985), Ionospheric structures in the polar-cap - their origin and relation to 250 Mhz scintillation, Radio Sci., 20(3), 325–338, doi:10.1029/RS020i003p00325.

Carrano, C.; Groves, K. (2009): Remote sensing the ionosphere, using GPS-SCINDA. IHY Africa/Scinda 2009 workshop, Zambia.

Carrano, C. S., K. M. Groves, and R. G. Caton (2012), The effect of phase scintillations on the accuracy of phase screen simulation using deterministic screens derived from GPS and ALTAIR measurements, Radio Sci., 47, RS0L25, doi:10.1029/2011RS004958.

Chaturvedi, P. K., and S. L. Ossakow (1981), The current convective instability as applied to the auroral ionosphere, J. Geophys. Res., 86(A6), 4811–4814, doi:10.1029/JA086iA06p04811.

Chaturvedi, P. K., and S. L. Ossakow (1983), Effect of an electron beam on the current-convective instability, J. Geophys. Res., 88(A5), 4119–4121, doi:10.1029/JA088iA05p04119.

Coley, W. R., and R. A. Heelis (1998), Structure and occurrence of polar ionization patches, J. Geophys. Res., 103(A2), 2201–2208, doi:10.1029/97JA03345.

Cowley, S. W. H., and M. Lockwood (1992), Excitation and decay of solar wind-driven flows in the magnetosphere-ionosphere system, Ann. Geophys., 10(1–2), 103–115.

De Franceschi, G., L. Alfonsi, V. Rornano, M. Aquino, A. Dodson, C. N. Mitchell, P. Spencer, and A. W. Wernik (2008), Dynamics of high-latitude patches and associated small-scale irregularities during the October and November 2003 storms, J. Atmos. Sol. Terr. Phys., 70(6), 879–888.

de la Beaujardière, O., V. B. Wickwar, G. Caudal, J. M. Holt, J. D. Craven, L. A. Frank, L. H. Brace, D. S. Evans, J. D. Winningham, and R. A. Heelis (1985), Universal time dependence of nighttime F region densities at high latitudes, J. Geophys. Res., 90(A5), 4319–4332, doi:10.1029/JA090iA05p04319.

Dungey, J. W. (1961), Interplanetary magnetic field and the auroral zones, Phys. Rev. Lett., 6(2), 47–48, doi:10.1103/PhysRevLett.6.47.

Frey, H. U., T. D. Phan, S. A. Fuselier, and S. B. Mende (2003), Continuous magnetic reconnection at earth's magnetopause, Nature, 426, 533.

Garner, T. W., R. B. Harris, J. A. York, C. S. Herbster, C. F. Minter III, and D. L. Hampton (2011), An auroral scintillation observation using precise, collocated GPS receivers, Radio Sci., 46, RS1018, doi:10.1029/2010RS004412.

Gondarenko, N. A., P. N. Guzdar, J. J. Sojka, and M. David (2003), Structuring of high latitude plasma patches with variable drive, Geophys. Res. Lett., 30(4), 1165, doi:10.1029/2002GL016437.

Gondarenko, N. A., and P. N. Guzdar (2004), Plasma patch structuring by the nonlinear evolution of the gradient drift instability in the highlatitude ionosphere, J. Geophys. Res., 109, A09301, doi:10.1029/2004JA010504.

Gulbrandsen N. (2006), Polar cap patch occurrence rate over Svalbard and its dependence on IMF By, Master thesis, University of Oslo.

Hey, J. S., S. J. Parsons, and J. W. Phillips (1946), Fluctuations in cosmic radiation at radiofrequencies, Nature, 158(4007), 234–234.

Hill, J. R. (1963), Sudden enhancements of F layer ionization in polar regions, J. Atmos. Sci., 20, 492.

49

Holmes J.M. (2014), The Protonics project: distributed observations of auroral dayside Doppler-shifted hydrogen emissions, PhD thesis, University of Oslo.

Holzworth, R. H., and C. I. Meng (1975), Mathematical representation of auroral oval, Geophys. Res. Lett., 2(9), 377–380, doi:10.1029/GL002i009p00377.

Hosokawa, K., J. I. Moen, K. Shiokawa, and Y. Otsuka (2011), Decay of polar cap patch, J. Geophys. Res., 116, A05306, doi:10.1029/2010JA016297.

Huba, J. D. (1984), Long wavelength limit of the current convective instability, J. Geophys. Res., 89(A5), 2931–2935, doi:10.1029/JA089iA05p02931.

Huba, J. D., and P. K. Chaturvedi (1986), The effect of finite “blob” size on the current convective instability in the auroral ionosphere, J. Geophys. Res., 91(A6), 7125–7130, doi:10.1029/JA091iA06p07125.

Jacobsen, K. S., and S. Schäfer (2012), Observed effects of a geomagnetic storm on an RTK positioning network at high latitudes, J. Space Weather Space Clim., 2, A13, doi:10.1051/swsc/2012013.

Jayachandran, P. T., et al. (2009), The Canadian High Arctic Ionospheric Network (CHAIN), Radio Sci., 44, RS0A03, doi:10.1029/2008RS004046.

Johnsen, M. G., and D. A. Lorentzen (2012), The dayside open/closed field line boundary as seen from space- and ground-based instrumentation, J. Geophys. Res., 117, A03320, doi:10.1029/2011JA016983.

Kelley, M. C., J. F. Vickrey, C. W. Carlson, and R. Torbert (1982), On the origin and spatial extent of high-latitude F region irregularities, J. Geophys. Res., 87(Na6), 4469–4475, doi:10.1029/JA087iA06p04469.

Kelley, M. (2009), The Earth’s Ionosphere: Plasma Physics and Electrodynamics, 2nd ed., 556 pp., Elsevier.

KEO Consultants (1993): Telecentric lens systems for wide-field low-f# monochromatic imaging, Tech. rep., KEO Consultants.

Keskinen, M. J., and S. L. Ossakow (1982), Nonlinear evolution of plasma enhancements in the auroral ionosphere, 1, Long wavelength irregularities, J. Geophys. Res., 87( A1), 144–150, doi:10.1029/JA087iA01p00144.

Keskinen, M. J., and S. L. Ossakow (1983), Theories of high-latitude ionospheric irregularities: A review, Radio Sci., 18(6), 1077–1091, doi:10.1029/RS018i006p01077.

Keskinen, M. J., Mitchell, H. G., Fedder, J. A., Satyanarayana, P., Zalesak, S. T., and Huba, J. D.: Nonlinear evolution of the Kelvin-Helmholtz instability in the high-latitude ionosphere, J. Geophys. Res., 93, 137–152, 1988.

Kintner, P. M., B. M. Ledvina, and E. R. de Paula (2007), GPS and ionospheric scintillations, Space Weather, 5, S09003, doi:10.1029/2006SW000260.

Kivanc, O., and R. A. Heelis (1997), Structures in ionospheric number density and velocity associated with polar cap ionization patches, J. Geophys. Res., 102(A1), 307–318, doi:10.1029/96JA03141.

Kivelson, M. G., and T. Russell (2001). Introduction to space physics, Cambridge University Press.

Li, G. Z., B. Q. Ning, Z. P. Ren, and L. H. Hu (2010), Statistics of GPS ionospheric scintillation and irregularities over polar regions at solar minimum, GPS Solutions, 14(4), 331–341.

Linson, L. M., and J. B. Workman (1970), Formation of striations in ionospheric plasma clouds, J. Geophys. Res., 75, 3211–3219, doi:10.1029/JA075i016p03211.

50

Lovelace, R. V. E. (1970), Theory and analysis of interplanetary scin-tillations, Ph.D. thesis, Cornell Univ., Ithaca, N. Y.

Lyons, L. R., R. L. Walterscheid, Generation of omega bands by shear instability of the neutral winds, J. Geophys. Res., 9012, 321, 1985.

Mitchell, C. N., L. Alfonsi, G. De Franceschi, M. Lester, V. Romano, and A. W. Wernik (2005), GPS TEC and scintillation measurements from the polar ionosphere during the October 2003 storm, Geophys. Res. Lett., 32, L12S0, doi:10.1029/2004GL021644.

Moen, J., H. C. Carlson, and P. E. Sandholt (1999), Continuous observation of cusp auroral dynamics in response to an IMF B-Y polarity change, Geophys. Res. Lett., 26(9), 1243–1246, doi:10.1029/1999GL900224.

Moen, J., K. Oksavik, L. Alfonsi, Y. Daabakk, V. Romano, and L. Spogli (2013), Space weather challenges of the polar cap ionosphere, J Space Weather Space Clim., 3, A02, doi:10.1051/swsc/2013025.

Moen, J., K. Hosokawa, N. Gulbrandsen, and L. B. N. Clausen (2015), On the symmetry of ionospheric polar cap patch exits around magnetic midnight, J. Geophys. Res. Space Physics, 120, doi:10.1002/2014JA020914.

Newell, P. T., K. Liou, Y. Zhang, T. Sotirelis, L. J. Paxton, and E. J. Mitchell (2014), OVATION Prime-2013: Extension of auroral precipitation model to higher disturbance levels, Space Weather, 12, 368–379, doi:10.1002/2014SW001056.

Ossakow, S. L. and Chaturvedi, P. K. (1979), Current convective instability in the diffuse aurora. Geophys. Res. Lett., 6: 332–334. doi:10.1029/GL006i004p00332.

Øieroset, M., P. E. Sandholt, W. F. Denig, S. W. H. Cowley (1997), Northward interplanetary magnetic field cusp aurora and high-latitude magnetopause reconnection, J. Geophys. Res., 102, 11349.

Poppe, B.B. (2000), New scales help public, technicians understand space weather. Eos Trans. AGU, 81 (29), 322–328, DOI: 10.1029/00EO00247.

Prikryl, P., P. T. Jayachandran, S. C. Mushini, D. Pokhotelov, J. W. MacDougall, E. Donovan, E. Spanswick, and J. P. S. Maurice (2010), GPS TEC, scintillation and cycle slips observed at high latitudes during solar minimum, Ann. Geophys., 28(6), 1307–1316.

Prikryl, P., P. T. Jayachandran, S. C. Mushini, and R. Chadwick (2011), Climatology of GPS phase scintillation and HF radar backscatter for the high-latitude ionosphere under solar minimum conditions, Ann. Geophys., 29(2), 377–392.

Rino, C. L., E. J. Fremouw (1977), The angle dependence of singly scattered wave fields, J. Atmos. Terr. Phys., 39, 859.

Rino, C. L. (1979a), Power law phase screen model for ionospheric scintillation 1. weak scatter, Radio Sci., 14(6), 1135–1145, doi:10.1029/RS014i006p01135.

Rino, C. L. (1979b), A power law phase screen model for ionospheric scintillation: 2. Strong scatter, Radio Sci., 14(6), 1147–1155, doi:10.1029/RS014i006p01147.

Rino, C. L., R. C. Livingston, R. T. Tsunoda, R.M. Robinson, J. F. Vickrey, C. Senior,M. D. Cousins, J. Owen, and J. A. Klobuchar (1983), Recent studies of the structure and morphology of auroral-zone F region irregularities, Radio Sci., 18(6), 1167–1180, doi:10.1029/RS018i006p01167.

51

Robinson, R. M., R. T. Tsunoda, J. F. Vickrey, and L. Guerin (1985), Sources of F region ionization enhancements in the nighttime auroral zone, J. Geophys. Res., 90(A8), 7533–7546, doi:10.1029/JA090iA08p07533.

Sandholt, P. E., C. J. Farrugia, M. Øieroset, P. Stauning, S. W. H. Cowley (1996a), Auroral signature of lobe reconnection, Geophys. Res. Lett., 23, 1725.

Sandholt, P. E., C. J. Farrugia, P. Stauning, S. W. H. Cowley, T. Hansen (1996b), Cusp/cleft auroral forms and activities in relation to ionospheric convection: Responses to specific changes in solar wind and interplanetary magnetic field conditions, J. Geophys. Res., 101, 5003.

Sandholt, P. E., C. J. Farrugia, J. Moen, O. Noraberg, B. Lybekk, T. Sten, and T. Hansen (1998), A classification of dayside auroral forms and activities as a function of interplanetary magnetic field orientation, J. Geophys. Res., 103(A10), 23,325–23,345, doi:10.1029/98JA02156.

Sandholt, P. E., H. C. Carlson, and A. Egeland (2002), Dayside and Polar Cap Aurora, Astrophysics and Space Science Library, vol. 270, Kluwer, Acad., Dordrecht, Netherlands.

Satyanarayana, P., and S. L. Ossakow (1984), Velocity shear stabilization of the current convective instability, J. Geophys. Res., 89(A5), 3019–3022, doi:10.1029/JA089iA05p03019.

Satyanarayana, P., J. Chen, and M. J. Keskinen (1985), Effects of finite current channel width on the current convective instability, J. Geophys. Res., 90(A5), 4311–4317, doi:10.1029/JA090iA05p04311.

Schunk, R. W., and A. F. Nagy (2000), The terrestrial ionosphere at high latitudes in Ionospheres, Cambridge Univ. Press, New York.

Shanmugam, S., J. Jones, A. MacAulay, and A.V. Dierendonck (2012), Evolution to Modernized GNSS Ionospheric Scintillation and TEC Monitoring, in: Proceedings of IEEE/ION PLANS, Myrtle Beach, South Carolina, April 2012, pp. 265–273, DOI: 10.1109/PLANS.2012.6236891.

Sojka, J. J., R. W. Schunk, M. D. Bowline, D. J. Crain (1997), Ambiguity in identification of polar cap F-region patches, J. Atmos. Terr. Phys., 59, 101.

Sojka, J. J., L. Zhu, M. David, and R. W. Schunk (2000), Modeling the evolution of meso-scale ionospheric irregularities at high latitudes, Geophys. Res. Lett., 27(21), 3595–3598, doi:10.1029/2000GL003813.

Spogli, L., L. Alfonsi, G. De Franceschi, V. Romano, M. H. O. Aquino, and A. Dodson (2009), Climatology of GPS ionospheric scintillations over high and mid latitude European regions, Ann. Geophys., 27(9), 3429–3437.

Tsunoda, R. T. (1988), High-latitude F region irregularities: A review and synthesis, Rev. Geophys., 26(4), 719–760, doi:10.1029/RG026i004p00719.

Van Dierendonck, A. J., J. Klobuchar, and Q. Hua (1993), Ionospheric scintillation monitoring using commercial single frequency C/A code receivers, paper presented at the 6th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GPS 1993), Salt Lake City, Utah, 22–24 Sept.

Vickrey, J. F., Rino, C. L. and Potemra, T. A. (1980), Chatanika/Triad observations of unstable ionization enhancements in the auroral F-region. Geophys. Res. Lett., 7: 789–792. doi:10.1029/GL007i010p00789

Weber, E. J., J. Buchau, J. G. Moore, J. R. Sharber, R. C. Livingston, J. D. Winningham, and B. W. Reinisch (1984), F layer ionization patches in the polar cap, J. Geophys. Res., 89(A3), 1683–1694, doi:10.1029/JA089iA03p01683.

52

Weber, E. J., R. T. Tsunoda, J. Buchau, R. E. Sheehan, D. J. Strickland, W. Whiting, and J. G. Moore (1985), Coordinated measurements of auroral zone plasma enhancements, J. Geophys. Res., 90(A7), 6497–6513, doi:10.1029/JA090iA07p06497.

Weber, E. J., J. A. Klobuchar, J. Buchau, H. C. Carlson, R. C. Livingston, O. Delabeaujardiere, M. Mccready, J. G. Moore, and G. J. Bishop (1986), Polar cap F layer patches: Structure and dynamics, J. Geophys. Res., 91(A11), 12,121–12,129, doi:10.1029/JA091iA11p12121.

Wernik, A. W., Secan, J. A., and Fremouw, E. J. (2003): Ionospheric irregularities and scintillation, Adv. Space Res., 31(4), 971–981.

Wernik, A. W., L. Alfonsi, and M. Materassi (2007), Scintillation modeling using in situ data, Radio Sci., 42, RS1002, doi:10.1029/ 2006RS003512.

Yeh, K. C., and C.-H. Liu (1982), Radio wave scintillations in the ionosphere, Proc. IEEE, 704, 325–378.

Yeh, K. C., and A. W. Wernik (1993), On ionospheric scintillation, in Wave Propagation in Random Media (Scintillation), edited by V. I. Tatarski et al., pp. 34–49, Int. Soc. for Opt. Eng., Bellingham, Wash.

Yin, P., et al. (2009), Imaging of the Antarctic ionosphere: Experimental results, J. Atmos. Sol. Terr. Phys., 71, 1757–1765, doi:10.1016/j.jastp.2009.09.014.

Zhang, Q. H., et al. (2013), Direct observations of the evolution of polar cap ionization patches, Science, 339(6127), 1597–1600.

Documents

Characterization of GPS Scintillations in the Polar Ionosphere