Data Processing Method for Geomagnetic Data Observation of

MAGDAS/CPMN System

SITI NOOR AISYAH AHMAD1, MOHAMAD HUZAIMY JUSOH

1, MOHD KHAIRUL MOHD

SALLEH1, MHD FAIROS ASILLAM

2 and MAGDAS/CPMN Group

3

1Faculty of Electrical Engineering

Universiti Teknologi MARA

Shah Alam, Selangor

MALAYSIA

2National Space Agency of Malaysia (ANGKASA)

Malaysia Space Centre

Banting, Selangor

MALAYSIA

3 International Center for Space Weather Science and Education (ICSWSE)

Kyushu University

JAPAN

Email: snaisyah12@yahoo.com; huzaimy@salam.uitm.edu.my

Abstract: - Space weather study has increasingly attracts the attention of many scientists to explore the

interaction between solar activity and geomagnetic activity. During the previous Space Weather initiative

program, called as International Space Weather Initiative (ISWI) period (2010-2012), International Center for

Space Weather Science and Education (ICSWSE), Kyushu University, Japan in collaboration with National

Space Agency of Malaysia (ANGKASA) and local universities has installed a magnetometer at National

Observatory Langkawi. In this paper, we will briefly discuss the data processing methods involve in order to

analyze the geomagnetic data observed by magnetometer from Langkawi station (LKW). The explanation of

the processing methods is based on the 24-hour data extracted during quiet and disturbed day.

Key-Words: - Magnetometer, geomagnetic data, magnetic pulsation and data processing method

1 Introduction

Magnetic pulsations or called ultra-low frequency

(ULF) pulsations is electromagnetic waves

generated in the magnetosphere. Its frequency

range is between 1 mHz and 1Hz. The generation

of magnetic field is defendant on solar and

processes in the magnetosphere. Earth’s magnetic

field observations play important role in the

understanding of the Earth’s electromagnetic

environment. Many experiments done by previous

researchers found that the variations in magnetic

fields are caused by the dynamo action in the upper

atmosphere. Daily variation (24 hours period) of

geomagnetic field components was first observed by

G. Graham in London [1]. The variations are then

observed as magnetic pulsations on the ground and

recorded in the range of Ultra Low Frequency

(ULF) with periods of 0.2 - 600 sec [2].

2 MAGDAS Instrumentation and

Geomagnetic Field Variation

International Center for Space Weather

Science and Education, ICSWSE, Kyushu

University, Japan has introduced a real-time

Magnetic Data Acquisition System of Circum-

pan Pacific Magnetometer Network, i.e.

MAGDAS/CPMN for space weather study and

application, which was deployed for the

International Heliophysical Year (IHY; 2007-

2009) [3]. By using this system, ICSWSE

conducted real-time monitoring and modelling

Mathematical and Computational Methods in Science and Engineering

ISBN: 978-960-474-372-8 126

of (1) global 3-dimensional current system, (2)

plasma mass density, and (3) penetrating

process of polar electric fields into the

equatorial ionosphere, in order to understand

the Sun-Earth coupling system and the

electromagnetic and plasma environment

changes [4]. To date, MAGDAS/CPMN

consists of three (3) unique chains of magnetic

observatories; the most magnetometers were

densely installed at 210° magnetic meridian, on

African longitude-sector and the other one is on

the sector along the magnetic equator (with total

of 71 stations worldwide), as shown in Figure 1.

From the magnetometer, we can extract

the ambient magnetic field, expressed by H

(Geomagnetic Northward), D (Geomagnetic

Eastward) and Z (Vertical Downward)

components.

Figure 1 Map of magnetometers installed under

MAGDAS/CPMN

2.1 Magnetometer

MAGDAS-9 (MAG-9) unit which was installed

at National Observatory Langkawi (LKW

station) consists of 3-component ring-core

fluxgate type magnetic sensor (magnetometer)

with 7 meter cable, pre-amplifier (preamp),

GPS (Global Positioning System) antenna with

cable, data logger for data control and 70 meter

cable. The main components of the

magnetometer system are main unit, pre-

amplifier and sensor as illustrated in Figure 2.

Data logger acts as a main unit to control

the power supply to the unit and communication

process. Magnetic field digital data (H + δH, D

+ δD, Z +δZ) are obtained with the sampling

rate of 10 Hz, and then 1 second and 1 minute

averaged data are recorded and transferred from

the oversea stations to the ICSWSE, Japan in

real-time [5]. The ambient magnetic field

components are digitized by using the field-

cancelling coils for the dynamic range of ±

70,000nT/32bits. The magnetic variations (δH,

δD, δZ) data are further digitized by the A/D at

preamp by 24 bits and 10 Hz resolution and

sampling frequency respectively. The long-term

inclinations (I) of the sensor axes are measured

by built in digital tilt meter with 0.1 arc-sec

resolution at calibrated accuracy ± 0.25 degree

(± 900 sec. degree). The temperature (T), are

also measured at both sensor and preamp with

resolution 0.01°C. The system synchronizes the

time of acquisition of the A/D conversion and

the GPS clock transmitted a pulse of 1 PPS

from the GPS module. These data are logging in

the Compact Flash Memory Card of 2 GB.

Figure 2 Diagram of MAGDAS system

2.2 Geomagnetic Data

Geomagnetic data (extracted from

magnetometer) is used in this study to monitor

ambient magnetic activity. The

MAGDAS/CPMN magnetometer is a ring core-

type fluxgate magnetometer that measures the

three components of the geomagnetic field;

Mathematical and Computational Methods in Science and Engineering

ISBN: 978-960-474-372-8 127

Horizontal component (H), Declination

component (D), and the Vertical component (Z)

as shown in Figure 3.

Figure 3 Geomagnetic field components; [F]

Total intensity of the geomagnetic field,

Horizontal component (H), Declination

component (D) and Vertical component (Z)

The 1-sec resolution data from horizontal

component were extracted to examine the

geomagnetic pulsations, Pc3, Pc4 and Pc5 as

classified by International Association of Geomagnetism and Aeronomy (IAGA) as shown in

Table 1. The raw data from MAGDAS/CPMN

stations was first bandpass-filtered before we plotted

the dynamic power spectra density to identify the

occurrences of ultra-low frequency (ULF) at Pc3,

Pc4 and Pc5.

Table 1 IAGA classification of ULF waves in

1964

ULF

pulsations Period

(sec)

Frequency

(mHz)

Continuous

Pc1 0.2-5 200-500

Pc2 5-10 100-200

Pc3 10-45 22-100

Pc4 45-150 6.7-22

Pc5 150-600 1.7-6.7

Irregular

Pi1 1-40 25-200

Pi2 40-150 6.7-25

3 Data Processing

The flow of the data processing is shown in Figure

4. The observed geomagnetic which stored in data

cards should be processed to be convenient for end

user of data in research work. The data processing

method needs to be implemented to ensure the

quality of data and the processed data is useful for

scientific research. As current procedures, Matlab

programming language is used to process the raw

data covering the processes for data availability

screening, ambient noise check-up, plotting, band-

pass filtering, power spectra density and Fast

Fourier Transform (FFT) analysis.

Figure 4 Flowchart of process MAGDAS data

At low latitudes, the horizontal component

is the major part of the total field and the

vertical component is significantly affected by

the geological and geographic surroundings of

the station [6]. Due to this fact, for LKW station,

we only demonstrate the processing method of

horizontal component of the geomagnetic field.

Mathematical and Computational Methods in Science and Engineering

ISBN: 978-960-474-372-8 128

3.1 Raw Data

The raw data were analyzed based on the quiet and

disturbed days which are on 15 March 2010 and 12

March 2010 respectively. Figure 5 and Figure 6

show raw data of the horizontal magnetic field

components observed at LKW station. Both figures

show that H component recorded higher amplitude

at time 0000 to 1000 UT (Universal Time). This is

due to day time effect where Local Time (LT) for

LKW station is + 8 UT. The local H component

afterwards maintained at 4.12 x 104

nT during night

time from 1000 till 2300 UT. Other than that, one

can see clearly the H component recorded on

disturbed day (12 March 2010) is distracted as

compared to H variation recorded on quiet day (15

March 2010).

Figure 5 Raw data of H (nT) magnetic components

on 15 March 2010

Figure 6 Raw data of H (nT) magnetic components

measured on 12 March 2010

3.2 Band-pass Filter

Raw data from LKW station was band-pass filtered

to classify the geomagnetic pulsations; either Pc

(continuous pulsation) or Pi (irregular pulsation)

pulsations. Figure 7 a), b) and c) show the Pc3, Pc4

and Pc5 respectively, on quiet day. The Pc3, Pc4

and Pc5 on disturbed days are shown in Figure 8 a),

b) and c) respectively. The ULF pulsations observed

during disturbed day (Figure 8 b) and c)), mainly on

Pc 4 and Pc 5 ranges show higher fluctuation as

compared to other Pc during quiet day.

a)

b)

c)

Figure 7 ULF pulsations a) Pc3, b) Pc4 and c) Pc5

on 15 March 2010

a)

Mathematical and Computational Methods in Science and Engineering

ISBN: 978-960-474-372-8 129

b)

c)

Figure 8 ULF pulsations a) Pc3, b) Pc4 and c) Pc5

on 12 March 2010

3.3 Power Spectra Density

To further confirm the occurrences of geomagnetic

pulsations, Power Spectra Density (PSD) method

has been applied into the process of analysis. The

PSD for significant ULF range (0 – 100 mHz) was

plotted based on color spectrum which corresponds

to the algorithm of the power in nT2/Hz. The

calculation was implemented based on Hanning

window through the data and Fast Fourier

Transform (FFT) on the subset of the signal within

the window.

Figure 9 shows PSD plot for quiet day where

no clear Pc (all ranges) can be observed and only 2

Pi events detected at around time 1400 and 1900

UT. For PSD plot on disturbed day as shown in

Figure 10, Pc 4 and Pc 5 events can be observed,

which occurred at time around 0200 to 0700 UT.

Figure 9 Power Spectra Density on 15 March 2010

Figure 10 Power Spectra Density on 12 March 2010

4 Data Analysis and Discussion

In this work, we have analyzed horizontal

component of geomagnetic data at LKW station,

which located at low latitude region. The data were

divided into 2 categories; quiet and disturbed period,

available at Data Analysis Center for Geomagnetism

and Space Magnetism, Kyoto University, Japan

(WDC). The selection of the quietest days (Q-days)

are derived from the magnetic activity indices by

index ranges through 0 to 9 with 0 is being quietest

or most disturbed day and 9 being least of both.

Furthermore, the selection of the most disturbed

days (D-days) are derived from the magnetic

activity indices by index ranges through 1 to 5 with

1 is being quietest or most disturbed day and 5 being

least. To further compare with solar wind events, we

have plotted the solar wind speed and solar wind

input energy. Solar wind speed events and other

parameters (proton density [cm-3

], magnetic field in

x, y and z-direction [nT]) on March 2010 were

obtained from the Space Physics Data Facility

(SPDF) based at NASA’s Goddard Space Flight

Center. While solar wind input energy need to

calculate using Akasofu epsilon, ɛ [7] as equation

(1):

epsilon, ɛ = Vsw B2 F(θ) Io

2 (Watt or erg/s) (1)

Where Vsw is solar wind speed [km/s], B is total

magnetic field [nT], Io is Earth’s radius [km]

and F (θ) is a function of the angle, θ (By/Bz) The occurring of ULF pulsations can be

determined by referring to the solar wind parameters

(solar wind speed and solar wind input energy).

Figure 11 shows a solar wind speed and solar wind

input energy from 11-16 March 2010. Solar wind

speed and solar wind input energy reached a higher

peak level on disturbed day (12 March 2010) as

compared to the quiet day which occurred 3 days

later on 15 March 2010.

Mathematical and Computational Methods in Science and Engineering

ISBN: 978-960-474-372-8 130

Figure 11 Solar Wind Speed (top) and Solar Wind

Input Energy (bottom) from 11-16 March 2010

5 Conclusion

The data processing method of geomagnetic data

recorded by magnetometer from earth station

Langkawi (LKW) has been discussed based on the

24-hour data extracted during quiet and disturbed

day. The occurrence of ULF pulsations is influenced

by the solar wind parameters (solar wind speed and

solar wind input energy). The ability of the

MAGDAS/CPMN magnetometer to measure ULF

pulsations is important to understand the space

weather using geomagnetic field data. By applying

the aforementioned data processing methods, it is

possible to extract and investigate the possible

relationship of space weather and the activities on

the lithosphere. However, further analysis and

evaluation involving extension of observational data

with advanced statistical analysis method are needed

to ensure the relationship of space weather and

geomagnetic activity can be comprehensively

explained.

Acknowledgement: The authors are grateful to the

MAGDAS/CPMN Group by International Center

for Space Weather Science and Education

(ICSWSE), Kyushu University, Japan for providing

the geomagnetic data at Langkawi station and

National Space Agency for maintaining the

equipment. The authors also want to thank

OMNIWeb Data Explorer, Space Physics Data

Facility from NASA for providing the data of solar

wind parameters. This project is funded under the

Ministry of Higher Education (MOHE) Malaysia

grants (600-RMI/DANA 5/3/PSI (175/2013) and

600-RMI/ERGS 5/3 (81/2012).

References:

[1] G.A. Agbo, A.O. Chikwendu and T. N.

Obiekezie, “Variability of Daily Horizontal

Component of Geomagnetic Field Component

at Low and Middle Latitudes, Indian J.Sci.Res.,

vol: 1(2), pp: 1-8, 2010.

[2] Robert L.McPherron, “Magnetic Pulsations:

Their Sources and Relation to Solar Wind and

Geomagnetic Activity”, Surveys in Geophysics,

vol: 26, pp: 545-592, 2005.

[3] K.Yumoto and the MAGDAS/CPMN Group,

“MAGDAS/CPMN Project for Litho-Space

Weather during IHY/ISWI (2007-2012)”, Space

Environment Research Center, Kyushu

University. [4] K. Yumoto and the 210°MM Magnetic

Observation Group, “The Step 210° Magnetic

Meridian Network Project”, J. Geomag.

Geoelectr., vol: 48, pp: 1297-1309, 1996

[5] K.Yumoto, “Capacity Building of MAGDAS”, Sun

and Geosphere, vol: 6(2) pp: 81 – 83, 2011.

[6] R. G. Rastogi and K. N.Iyer, “Quiet Day

Variation of Geomagnetic H-Field at Low

Latitudes”, J. Geomag. Geoelectr., 28, 461-479,

1976.

[7] S.I. Akasofu, “Energy Coupling between the

Solar Wind and the Magnetosphere”, Space

Science Reviews, vol: 28, no:2, pp: 121-190,

1981.

Mathematical and Computational Methods in Science and Engineering

ISBN: 978-960-474-372-8 131