Calibration and evaluation of the secondary sensors for the Mini-EUSO space instrument

(1)

Calibration and evaluation of the secondary sensors for

the Mini-EUSO space instrument

Jonah Ekelund

Space Engineering, master's level 2018

Luleå University of Technology

Department of Computer Science, Electrical and Space Engineering

(2)

Master’s Thesis

Calibration and evaluation of the secondary sensors for the Mini-EUSO space instrument

Author:

Jonah Ekelund Supervisor:

Marco Casolino Examiner:

Anita Enmark

(3)

Acknowledgement

I would like to thank my supervisor Marco Casolino and everyone at RIKEN that helped me with my thesis work or everyday things in Japan. I also want to thank the Sweden Japan Foundation for the Scholarship that made it possible for me to stay in Japan for six months. My mother and father also have my gratitude for the support before, during and after my trip to Japan.

(4)

Abstract

The Mini-EUSO (Mini - Extreme Universe Space Observatory) is an instrument for observation of ultra-high energy cosmic rays (UHECR) from space. It is designed to observe Earth from the international space station (ISS) in the ultra-violet (UV), visible (VIS) and near-infrared (NIR) light ranges. The UV sensor is the main sensor, designed and built by the EUSO collaboration. The visible and near-infrared sensors are secondary sensors. These are two cameras, FMVU-13S2C-CS and CMLN-13S2M-CV, from Point Grey Research Inc. The near-infrared light camera has a phosphor coating on the sensor to convert from near-infrared light to visible light, which is detectable by the camera’s CCD.

This thesis deals with the calibration and evaluation of the secondary sensors. This is done by first evaluating the bias and dark current for both cameras. After which a calibration is done using the light measurement sphere, located at the National Insti- tute of Polar Research (NIPR) in Midori-cho, Tachikawa-shi, Japan. Due to the low sensitivity of the near-infrared light camera, an evaluation of its ability to see celestial objects are also performed.

It is found that the visible light camera has a high bias with values around 5 ADU (Analog-to-Digital unit), but almost non-existing dark current, with mean values below 1 ADU. The visible light camera has good sensitivity for all the colors: red, green and blue. However, it is most sensitive to green. Due to this, it is easy to saturate the pixels with too much light. Therefore, saturation intensity was also examined for the shutter times of the visible light camera. This is found to be between 900µWm⁻²sr⁻¹ and 1 · 10⁷µWm⁻²sr⁻¹, depending on color and shutter time.

The near-infrared light camera is the opposite; it has a low bias with values below 1 ADU and a high dark current. The values of the dark current for the near-infrared light camera are highly dependent on the temperature of the camera. Mean values are below 1 ADU for temperatures around 310K, but mean values of almost 2 ADU at temperatures around 338K. The sensitivity of the near-infrared light camera is very low, therefore, the only way to detect a difference between the light levels of the light measurement sphere was to use a high ADC amplification gain. With this it was found that there is a power-law behavior, values between 1.33 and 1.50, of the relationship between pixel values and light intensity. This is likely due to the phosphor coating used to convert to visible light. When trying to detect celestial objects, the faintest object detected was Venus with a magnitude of less than -4.

(5)

List of Figures

1.1 Spectrum of the Cosmic ray flux for energies 10⁸-10²¹eV . . . . 2

1.2 Spectrum of the Cosmic ray flux for energies 10¹⁷-10²¹eV . . . . 2

1.3 A representation of the processes in an EAS. . . . 4

1.4 A representation of a binary increase . . . . 5

1.5 The total amount of detected space objects in Earth orbit . . . . 12

1.6 The EUSO-Balloon instrument. . . . 15

1.7 Artist’s interpretation of the JEM-EUSO instrument . . . . 17

2.1 Mini-EUSO conceptual design . . . . 22

2.2 Representation of the basic working principle of a CCD. . . . . 26

2.3 Representation of a CMOS pixel. Credit: Walthman (2013) . . . . 27

2.4 The light measurement sphere . . . . 30

2.5 The cameras mounted on the mounting table. . . . 31

2.6 The reflective coting inside the sphere . . . . 31

2.7 The reflectance of the reflectance coating . . . . 31

2.8 The spectrum of the calibration device . . . . 31

2.9 Least square fit to Planck’s Law on the light spectrum with the highest intensity from Figure 2.8 . . . . 32

2.10 VIS camera: The dark current dependence on shutter time. . . . 36

2.11 VIS camera: The mean pixel value over the entire (1980x960) dark current image plotted against operation time. . . . 37

2.12 VIS camera: The mean value of the pixels in the dark current image containing dark current plotted against the operation time . . . . 37

2.13 The response curve for the visible light camera . . . . 39

2.14 Representation of the Bayer filter of the visible light camera . . . . 39

2.15 Spectrum of the flat-field foil . . . . 40

2.16 VIS Camera: An image that should be completely uniform . . . . 41

2.17 VIS Camera: A color coded version of Figure 2.16 . . . . 41

(8)

LIST OF FIGURES Jonah Ekelund 2.18 VIS Camera: Pixel value from the flat-field image, before compensation 42 2.19 VIS Camera: Pixel value from the flat-field image, after compensation . 42 2.20 VIS Camera: Pixel value from a close to ideal image, before compensa-

tion. Note difference in x-axis compared to Figure 2.19 and 2.18 . . . . 42 2.21 VIS Camera: Pixel value from a close to ideal image, after compensation.

Note difference in x-axis compared to Figure 2.19 and 2.18 . . . . 42 2.22 VIS Camera: Color-map describing how the pixel values relate to the

mean value of all the pixels in the image with the same color. . . . . . 43 2.23 VIS Camera: Pixel values dependence on radiance . . . . 44 2.24 VIS Camera: The mean pixel values dependence on the shutter time . . 45 2.25 VIS Camera: Time to saturation depending on intensity. . . . 46 2.26 NIR Camera: Image of alpha radiation hitting the ccd of the near-

infrared light camera . . . . 49 2.27 The Near-infrared light camera’s sensitivity range . . . . 50 2.28 NIR Camera: The sensitivity of the near-infrared light camera fitted to

the sum of three Gaussian functions. . . . 50 2.29 NIR Camera: The dark current dependence on shutter time with a linear

fit. . . . 51 2.30 NIR Camera: The temperature development with respect to operation

time . . . . 52 2.31 NIR Camera: The mean pixel value over the entire (1980x960) dark

current image plotted against temperature. . . . 52 2.32 NIR Camera: Data from the light measurement sphere with box integration 55 2.33 NIR Camera: Data from the light measurement sphere with assumed QE 55 2.34 NIR Camera: Same data as Figure 2.32 but fitted to a power-law . . . 57 2.35 NIR Camera: Same data as Figure 2.33 but fitted to a power-law . . . 57 2.36 NIR Camera: Picture of the moon with visual magnitude -11.9 . . . . . 58 2.37 NIR Camera: Observations of the moon at different magnitudes . . . . 59 2.38 NIR Camera: Photon count from observations of the moon . . . . 60 2.39 NIR Camera: Magnitudes relative the Brightness of Venus. . . . 60

(9)

List of Tables

1.1 Defining parameters of the JEM-EUSO, Mini-EUSO, EUSO-Balloon and

K-EUSO instruments. . . . . 17

2.1 Mini-EUSOmain characteristics. . . . 23

2.2 Visible light camera, technical information of the camera used in this project. . . . . 24

2.3 Near-infrared light camera, technical information of the camera used in this project. . . . 25

2.4 The ratios of different intensities . . . . 33

2.5 VIS camera: Settings used for the visible light camera during testing. . 34

2.6 Naming of the Bayer pattern . . . . 39

2.7 Naming of the Bayer pattern in different applications . . . . 39

2.8 VIS Camera: Shutter times without striping . . . . 40

2.9 Fit values for Figure 2.23 . . . . 44

2.10 NIR camera: Settings used for the near-infrared light camera during testing. . . . 48

2.11 Fit parameters for Figure 2.34 and 2.35. . . . . 56

2.12 Table of values from Figure 2.37 . . . . 59

2.13 The K_moon for the line fits in Figure 2.38 . . . . 60

2.14 The Magnitudes of the moon relative Venus. . . . 60

(10)

Chapter 1 Introduction

The purpose of Mini-EUSO space instrument is to observe ultra-high energy cosmic rays (UHECR) and test technology for the future JEM-EUSO instrument. Like JEM-EUSO, Mini-EUSO will observe Earth’s atmosphere from the international space station (ISS) giving it the possibility to observe both the northern and southern hemisphere.

The main part of Mini-EUSO is its Ultra-violet camera, for the purpose of air fluorescence observation. However, Mini-EUSO also has two secondary sensors, one near-infrared light camera and one visible light camera, for complementary data collection. The purpose of this thesis is the calibration and evaluation of these secondary cameras.

This first chapter will briefly review some relevant phenomena of interest to the Mini-EUSO instrument and give an overview of the EUSO collaboration. Chapter 2 will then give an overview of the Mini-EUSO instrument and in detail go through the work performed for this thesis.

1.1 Cosmic Rays

Soon after radiation was discovered in the 19th century, it was also discovered that ions were created at ground-level seemingly out of nowhere. The rate of creation was between 10 and 20 ions per cm³ per second. It was postulated that these ions were generated out of natural γ-rays from the Earth (Letessier-Selvon and Stanev 2011).

From this a number of expeditions to measure the ionization rate at different alti- tudes were launched, beginning with measurements from the top of the Eiffel Tower.

One of the most notable being that of Victor Hess in 1912. By flying a balloon he discovered that at the altitude of 5 km, the ionization actually increased by a factor of two compared to ground level (Letessier-Selvon and Stanev 2011). This meant that the

(11)

1.1. COSMIC RAYS Jonah Ekelund

Figure 1.1: Spectrum of the Cosmic ray flux for energies 10⁸-10²¹eV compiled by Hanlon (2008)

Figure 1.2: Spectrum of the Cosmic ray flux for energies 10¹⁷-10²¹eV compiled by Hanlon (2008)

cause for the ionization could not be from the ground, instead, it had to originate from above. This discovery gave him the Nobel Prize in physics the year 1936 (Westerhoff 2012).

Subsequently, at an altitude above 9 km, Werner Hohlh¨orster measured even higher ionization levels. In these early stages, cosmic rays were actually called H¨ohenstralung (high altitude radiation).

By measuring the cosmic rays at locations far away from each other, the conclusion was drawn that the cosmic rays formed ”showers”; A term first used by the Italian Rossi (Letessier-Selvon and Stanev 2011).

Much of the knowledge gained throughout these early years lead to the publication

”Cosmic ray theory” by Rossi and Greisen (1941).

1.1.1 Cosmic ray spectrum

The cosmic ray spectrum as it is known today is depicted in Figure 1.1. The early observations, described above, were of energies at the lower end of the spectrum. Here, the particle’s have low enough energy to be affected by the Earth’s magnetic field. This

(12)

Jonah Ekelund 1.1. COSMIC RAYS bends the particles path and often makes it hard to locate the origin of the cosmic rays (Letessier-Selvon and Stanev 2011).

In the figure, there are two features marked that are of extra interest. The first is easily seen in the spectrum at ∼10¹⁵eV. This is the knee. The second one is the ankle at ∼3 · 10¹⁸eV. This is better seen in Figure 1.2.

The cosmic ray spectrum can generally be said to follow a power law, E^α. However, the value for α depends upon which part of the spectrum is referred to. At energies lower than the knee α is about 2.7 and at energies between the knee and ankle α is about 0.3 higher. At energies above the ankle, α becomes similar to energies lower than the knee (Letessier-Selvon and Stanev 2011).

For energies above 10¹⁸eV the cosmic rays are called Ultra-High energy cosmic rays (UHECR) (Bertaina and Parizot 2014). This is also the part that is of interest to the EUSO project.

1.1.2 Extensive air shower

Extensive Air Showers, or EAS as they will be referred to from here onwards, can be caused by different kinds of particles entering the atmosphere. For example, these can be, hadrons, gamma-rays or electrons (Westerhoff 2012). Figure 1.3 is a representation of how one of these showers can progress. This figure can give a first overview of the different processes that take place in an EAS.

Beginning from the top, the first interaction between the cosmic ray and a molecule in the atmosphere initiates the EAS. Exactly where in the atmosphere this interaction takes place depends on both the energy of the cosmic ray and the compositions (Grieder 2010; Letessier-Selvon and Stanev 2011; Westerhoff 2012).

After this first interaction, through a number of subsequent interactions, many different particles are created, for example, pions, muons, electrons and hadrons (Letessier- Selvon and Stanev 2011). When these particles reach the ground they can then be detected by ground-based instruments, see section 1.1.3.

Before the shower reaches the ground, it also causes air fluorescence and Cherenkov light. Air fluorescence is isotropic, see Section 1.1.2 and is therefore of interest for the JEM-EUSO, see section 1.4 and 2.1. The Cherenkov light is mainly in the direction of the cosmic ray, see Section 1.1.2 (Letessier-Selvon and Stanev 2011). However, due to reflection and scattering in the air, this can also be seen by a spaceborne instrument like EUSO (Fuglesang 2017).

(13)

1.1. COSMIC RAYS Jonah Ekelund

Figure 1.3: A representation of the processes in an EAS. Credit: Grieder (2010)

(14)

Jonah Ekelund 1.1. COSMIC RAYS

Figure 1.4: A representation of a binary increase in number of particles N after step number d. In the Heitler model, d represents the number of radiation lengths, λr, that the particles have traveled.

Heitler’s model for electromagnetic cascades

Heitler’s model is a simplified model that only describes the electromagnetic part of the full EAS. However, it is a good starting place to get a basic idea of what happens. This model will then be expanded upon in Section 1.1.2 to account for hadron interactions.

The idea in this model is that the cascade is binary. In other words that in each step the number of particles are doubled. Thereby also assuming that all interactions for a certain step happen in a time interval before the next step, Figure 1.4.

If the first particle is a high energy electron, then the model assumes that in the first stage the electron splits its energy exactly in two. It does this through Bremsstrahlung, leaving the electron with half of the initial energy and a photon with the other half (Letessier-Selvon and Stanev 2011).

In the next stage, two different reactions take place. The first is the same as in the first stage, the electron dividing its energy in half. The second reaction is that of the photon. The photon also divides its energy in half. However, it does this by creating an electron-positron pair with equal energy. In the next step, the positron behaves like the electron, splitting its energy in half. This continues in each step with the electrons and positrons dividing its energy through Bremsstrahlung and the photons by creation an electron-positron pair (Letessier-Selvon and Stanev 2011).

(15)

1.1. COSMIC RAYS Jonah Ekelund Following this idea, the number of particles (N) after stage d is then given by

N (d) = 2^d (1.1)

These reactions continue until the energy of the electrons are below a critical energy E_c^γ. This happens when the rate of energy loss to Bremsstrahlung is the same as the rate of energy loss to ionization. What the critical energy is, depends on the medium, for air it is around E_c^γ = 80MeV (Letessier-Selvon and Stanev 2011).

The actual length of each step, d, is dependent on the medium and is therefore given by the radiation length, λ_r; which is the mean length that the radiation travel before it reacts.

From this simple model, a number of key properties of an electromagnetic cascade can be derived. The number of particles at the maximum of the cascade

Nmax = E₀ Ec^γ

(1.2) which is dependent on the initial energy E₀. The location of the maximum

X_max = X₀+ λ_rln E₀ Ec^γ

(1.3) which, beyond previously mentioned properties, depends on the location of the initial reaction X₀ and the radiation length λ_r of the medium.

How fast X_max changes with the energy is defined as D₁₀≡ dXmax

d log₁₀E₀ = 2.3λ_r (1.4)

This is the elongation rate. As can be seen, this is dependent on the radiation length.

And thereby on the medium, for air the elongation rate is 85 g/cm² (Letessier-Selvon and Stanev 2011)

Hadron cascade

When expanding Heitler’s model to account for hadrons the radiation length, λ_r, is exchanged for the hadron interaction length, λ_l. Which is the mean distance that a hadron travels through the medium before it interacts with the medium.

Beginning with the assumption that the cosmic ray is a single proton, in the first reaction of the hadron cascade neutral and charged pions are created. The ratio is 1:2, or N_π neutral and 2N_π charged, N_π is thereby one-third of the total number

(16)

Jonah Ekelund 1.1. COSMIC RAYS of pions produced in one step. The neutral pion immediately decays and initiates an electromagnetic cascade. The charged pions, however, continue to interact with hadrons and thereby increasing the size of the hadron cascade (Letessier-Selvon and Stanev 2011).

When the charged pions reach a critical energy, E_c^π, they decay into muons, which is detected at ground level. The equation that describes the muon count is

N_µ^p = E₀ E_c^π

β

(1.5) where β = ln (2Nπ)/ ln (3Nπ). This equation is extracted by assuming that all pions decay into muons when they reach critical energy. As can be seen from this equation, the exact muon count does not have a linear dependence on the initial energy. It has, through β, a dependency on the average number of pions created in each step (Letessier- Selvon and Stanev 2011). According to simulations done by Alvarez-Mu˜niz et al. (2002) the value of β is in the range of 0.9 to 0.95.

Also, as electron cascades are produced in each step of the hadron cascade, the size of the total cascade is dependent on how long it takes the pions to reach E_c^π.

The dependence of the shower maximum X_max^p on the initial energy of the cosmic ray E0 is very complex and hard to describe in a simple model. This is because the large number of particles created in each reaction. How elastic, or inelastic, the reactions are, also affect the position of the shower maximum (Letessier-Selvon and Stanev 2011).

Larger Atoms

To expand the model to nuclei consisting of more than one proton the superposition principle is used. In other words, the nucleus is assumed to be A nucleons, each with the energy E0/A. From this follows that the shower maximum, relative to proton shower maximum, is located

X_max^A = X_max^p − λ_rln (A) (1.6) after the initial interaction. The number of muons are

N_µ^A= N_µ^pA^1−β (1.7)

more than for a proton shower (Letessier-Selvon and Stanev 2011).

Air fluorescence

Air fluorescence, also known as atmospheric fluorescence, is caused by the nitrogen molecules in the air. These get excited when the charged particles of the EAS pass by.

(17)

1.1. COSMIC RAYS Jonah Ekelund This leads to ultra-violet light being emitted (300 to 420nm). While most of the other effects of an EAS are in the beam direction, air florescence radiates isotropically. This means that a detector designed for the detection of air fluorescence does not necessarily have to be in the path of the beam to detect the signal (Letessier-Selvon and Stanev 2011).

By sampling the UV light in both space and time, the trajectory of the EAS can be determined. By knowing this the longitudinal development of the EAS can be reconstructed. Which leads to the location of X_max and the primary energy E₀ (Grieder 2010).

While this method is more direct than other detection methods, see section 1.1.3, the height of the primary interaction cannot be seen. This is because the light from this is too faint and therefore does not reach the detector with a strong enough signal.

This makes it hard to get the primary mass and also limits the accuracy of the X_max (Grieder 2010).

Air cherenkov light

When a charged particle travels through a dielectric medium at speeds higher than the phase speed of light in that medium, then photons will be created traveling in the same direction as the particle. This is called Cherenkov radiation, or in this case when it is created in the atmosphere air Cherenkov light. As air Cherenkov light is directional it is best observed with a similar scheme as the particle detectors described in section 1.1.3.

Multiple detectors are arranged in an array formation. This is because the Cerenkov light originates from the charged particles in the EAS and thereby will be most intense near the shower core. For this kind of detection, the relevant data of interest is the photon count, arrival time and coordinates of the detection. This Cherenkov light contains information about the longitudinal development of the shower (Grieder 2010).

Because the air Cherenkov light is directional, it is not possible to measure it directly from a space observatory looking in the nadir direction. However, the Cherenkov light diffracts in the atmosphere which causes some light to radiate isotropically. This causes a slight increase in the light from the EAS. Also, the reflection from the ground gives a peak in the photon count after the arrival of the light directly from the EAS. If there are stratus clouds for the light to reflect off of, then this peak is closer to the main peak (Bertaina and Parizot 2014).

1.1.3 Current observational methods

As seen in the Figures 1.1 and 1.2, the amount of cosmic rays that hit the atmosphere at the higher energies are very small, 1 particle/km²/century. It is not realistic to have

(18)

Jonah Ekelund 1.1. COSMIC RAYS one small detector and expect to get enough detections for exploration of UHECR.

One possibility to get enough detections is to build large arrays of detectors. For example the Pierre Auger observatory, covering 3000km², and the Telescope Array Project, covering 700km²(Coutu 2016; Hanlon 2016).

Pierre Auger is a hybrid detector system. This means that it consists of a surface array and air fluorescence detectors. The surface array consists of 1660 water- Cherenkov detectors placed in a triangular grid with a shortest separation between detectors of 1500m (Collaboration 2015). These detectors detect muons and electrons by the Cherenkov light they create when passing through a tank filled with ultra clean water. The amount of Cherenkov light that the particles produce is related to the energy-to-mass ratio of the particles (Billoir 2014). The detectors use GPS to timecode the detections, which enables the possibility to see which detectors detected a specific shower first (Letessier-Selvon and Stanev 2011).

The air fluorescence part of the detector complement the ground array data. How- ever, as these are ground-based they have the limitations. Firstly, they can only operate during the night, effectively cutting the possible observation time in half. Secondly, they can only be used when the weather is clear, further limiting the available observation time (Westerhoff 2012). The total duty cycle of a ground-based air-fluorescent detector is only around 10% (Letessier-Selvon and Stanev 2011).

(19)

1.2. METEORS Jonah Ekelund

1.2 Meteors

Meteors are a phenomenon that humans have been able to see for a very long time.

However, it was first in the 17th century that the understanding of meteors, as they are known today, started to take form. Edmond Halley postulated in 1714 that they had an extraterrestrial origin. Before it had been believed that meteors were explosions of flammable vapors (Hughes 1982).

In 1719, Halley calculated the height and speed of a meteor by using elevation data taken from several different points. The values he calculated was 119km and 8.0 km/s.

This speed is below Earth’s escape velocity, 11.2 km/s. This was the first time someone tried to calculate the speed and height of a meteor (Hughes 1982).

In the 19th century, the connection between meteorites and meteors were made by E.F.F. Chladni. Previously meteorites had been believed to be created by thunder- storms.

In 1866 John Browning tried to get the spectra of meteors in a meteor shower.

He was successful in identifying four different kinds of spectra. One similar to a solar spectrum, except for the lack of violet. Another similar to the first, but with a more dominant yellow part. The third being almost only yellow, but with some traces in the red and green part. The forth having only a green part. While Browning did reference knowledge from chemistry on how the colors of burning chemicals are related to what chemical it is, he did not draw any conclusion on the composition of the meteors he observed. Brownings work did, however, lead to the birth of meteor chemistry (Hughes 1982).

The idea that meteors are debris from a larger object was mentioned in 1861. How- ever, at the time there was no data to support this and therefore it was dismissed as speculation. Just a couple of years later, in 1863, Hubert Anson Newton proved that the annual maximums of meteor showers were close to that of a sidereal year. He, therefore, made the conclusion that the showers were caused by the Earth passing through a shower of meteoroid particles (Hughes 1982).

During the 1860s many calculations on the orbits of meteors, or more accurately the objects causing the meteors, were made. With the orbits calculated, a resemblance with orbits from some comets was quickly discovered. This lead to meteors and comets being closely associated with each other. (Hughes 1982).

In the late 19th century and early 20th century, photography became a tool used in meteor observation. An example of this is Elkin that, in the year 1900, used two cameras separated by 3 km. Each camera was equipped with a rotating shutter; Which was a bicycle wheel covered with screens. The screens were placed such that gaps were formed between them letting light through. By rotating the wheel in front of the camera

(20)

Jonah Ekelund 1.3. SPACE DEBRIS with a fixed angular frequency, Elkin could calculate the angular velocity of the meteor.

If the meteor was observed by both of the cameras, he could also calculate the linear velocity (Hughes 1982).

1.2.1 Observation

Optical observation is used to observe the light emitted by the meteor due to the entry into Earth’s atmosphere. The time the meteor is visible is generally between 0.5 to 3 seconds. The meteors generally become bright enough to be observed between 75 to 120km and are visible until 30 to 100km. When observing meteors this way, they are often classified according to the magnitude scale. This scale is defined according to

M − M₀ = −2.5 log₁₀ I I₀

(1.8) where M₀ is the magnitude of a reference object, I₀ is the intensity, in W/m², of the reference object and I is the intensity, in W/m², of the object the magnitude, M , is calculated for. The brightest meteors, with a visual magnitude less than -4, are called fireballs.

The light emitted by the meteor comes from the kinetic energy of the meteor. While the exact phenomena vary depending on velocity and mass of the meteor, in general terms the light is caused by the friction between the meteoroid and the air in the atmosphere. This heats the air until it is hot enough to form a plasma around and behind the meteor.

Due to the heat, the surface material of the meteoroid is ablated causing the meteoroid to decrease in mass. If the meteoroid is small it will disappear, while larger meteoroids do not ablate completely and continue down to impact the ground (Bertaina, Cellino, et al. 2015).

1.3 Space debris

Since the launch of Sputnik-1 in October 1957, the total amount of debris left by human space activities in orbit around Earth has increased (Klinkrad 2006). Figure 1.5 shows the total amount of detected objects in orbit around Earth, the category with the highest numbers are fragmentation debris. This, together with ”Mission-related debris”, is what most often is referred to as space debris.

Fragmentation debris is as the name suggests, debris from the fragmentation of a spacecraft. Sometimes this is by accident, a malfunction leading to the spacecraft

(21)

1.3. SPACE DEBRIS Jonah Ekelund

Figure 1.5: The total amount of detected space objects in Earth orbit (black), and the amount from different sources (colored). Plot is from NASA Orbital Debris Program Office - Quarterly News (Anx-Meador 2017)

exploding or a collision between two spacecraft. However, some of the debris is inten- tional, often part of some anti-satellite weapons test (Klinkrad 2006). The collision between the French satellite Cerise (95-033D) and a fragment from an Ariane-1 upper stage on July 24, 1996, is probably the first unintentional collision in space (Klinkrad 2006). Other common space debris are slag and dust from solid rocket motors or paint flakes, not seen in Figure 1.5 as they are often too small to be observed with current ground instruments.

Solid fuels often have aluminum powder in it to stabilize the burning process. Most of this aluminum exits the motor with the exhaust in the form of aluminum oxide (Al₂O₃) which forms dust with diameters between 1 µm ≤ d ≤ 50 µm. At the end of the burning processes some of the aluminum form slag together with insulation material from the booster. These can form particles in the size range 0.1 mm ≤ d ≤ 30 mm (Klinkrad 2006).

Not surprisingly, the location of the space debris is correlated to where spacecraft are most commonly located. For example, there is a lot of debris in low earth orbit (LEO) and geostationary orbit (GEO). The debris in LEO is low enough to be significantly affected by drag from the atmosphere. This eventually leads to debris falling into the atmosphere and burning up. According to Klinkrad (2006), all the launch activities up to January 2002 had generated a total of 27,044 cataloged objects. Out of these 18,051 had burned up in the atmosphere by that date.

There are still a lot of debris being generated. Also, debris left in GEO, have orbits

(22)

Jonah Ekelund 1.3. SPACE DEBRIS that do not degrade. All this leads to the increase of space debris that can be seen in Figure 1.5.

1.3.1 Observational systems

There are mainly two ways that are currently used to study space debris, from ground and in-situ.

Ground-Based

Ground-based systems use either radar or optical telescopes. Radar has a better re- solving power than optical, therefore it is used for orbits closer to Earth. However, as it is an active system the signal strength decrease with 1/r⁴, r here being the distance to the target.

Signals for optical systems only decrease with 1/r²; however, they have the limita- tion that they can only be used during the night. They are further limited by meteo- rological conditions. Also, the sun has to illuminate the targets, which means that it can only observe space debris in certain sun-target-telescope geometric configurations.

To calculate how much signal that can be expected for a target of a certain size the equation

P = F· A · A_t· α_tcos (_t,in) · cos (_t,out)

r² (1.9)

can be used. Here, F is the radiance of the sun at 1 AU, A is the area of the collector, A_t is the area of the space debris (assuming a plane), α_t is the albedo of the debris,

_t,in and _t,out is the incident angel of incoming and outgoing photons respectively and r is the distance to the debris (Klinkrad 2006).

In-situ

There have been some missions where a surface has been exposed to the space envi- ronment for the expressed purpose of studying impact events. One such mission was the Long Duration Exposure Facility (LDEF), deployed in April 1984 and retrieved in January 1990. Also, many surfaces that have been to space, exposed to the space en- vironment and later brought down to Earth, have also been studied for impact events.

An example of this is one of the solar panels of the Hubble space telescope that was replaced with a new one in 1993 (Klinkrad 2006).

The advantage of studying these kinds of impact events is that smaller sized impactors can be seen. Also, because the impactors leave a chemical residue, this can be analyzed and give information of where the impactor came from. However, these kinds

(23)

1.4. EUSO PROGRAM Jonah Ekelund of studies only give an approximate flux of how much small debris is currently in orbit.

It does not give any exact information of where the debris is located (Klinkrad 2006).

There are some small-particle sensors that can tell when and where they were impacted to give a better resolution. However, these still only show debris that has impacted the spacecraft (Klinkrad 2006).

1.3.2 Removal

Most of the management of the space debris today entails finding the debris then logging it into a database. If there is a large risk of something hitting an important operating satellite, then a maneuver is performed to prevent this from happening. For example, the international space station generally performs such a maneuver once every year (Klinkrad 2006). There are not really any programs for actively removing space debris from orbit. However, there are some ideas for how this could be accomplished.

GUERRA et al. (2017) describes a system where they increase the atmospheric drag of the debris by use of a foam. Another example is Aslanov and Yudintsev (2013) who analyses an idea for the use of a space tug with a tether to lower the orbit of debris.

These two are examples of ideas on how to remove large debris from Earth’s orbits.

However, not may ideas exist for removing small debris. Debris that is still hazardous for spacecraft, but small enough that it is not feasible to go and pick them up with a spacecraft. One way to deal with this small debris is to use a laser to cause ablation on the surface of the debris. By causing this ablation on the correct side of the debris, it will decrease the orbital speed of the debris and thereby lower the orbit. If the orbit becomes low enough, the debris will fall into the atmosphere (Ebisuzaki et al. 2015).

Targeting the space debris requires exact location data on the debris. It is in this role that the Mini-EUSO, and later on the JEM-EUSO, could be used (Ebisuzaki et al.

2015).

1.4 EUSO Program

EUSO stands for Extreme Universe Space Observatory and is a collaboration consisting of ”93 institutes in 16 countries” (Fuglesang 2017). The aim of this collaboration is to study the higher end of the UHECR spectrum.

1.4.1 Previous missions

Within the EUSO project there have been a number of pathfinder projects to prototype the detectors and prove that the technology works. Missions that have been completed

(24)

Jonah Ekelund 1.4. EUSO PROGRAM

Figure 1.6: The EUSO-Balloon instrument. Credit: J. H. Adams et al.

(2015)

are EUSO-balloon, EUSO-TA and EUSO-SPB. Here follows a short description of these missions.

EUSO-Balloon

This was a balloon-borne instrument, see Figure 1.6, designed to test all the key technologies needed for the final instrument. The balloon was launched in August 2014 from Timmins Stratospheric Balloon Base in Canada. It reached an altitude of almost 40km (Scotti and Osteria 2016).

The instrument operated for approximately five hours. Due to the short operation time and that the balloon had a relatively small observation area, it would have been lucky if the instrument observed any actual EAS. Therefore, a laser mounted on a helicopter was used to shoot across the instruments field of view (FoV) and thereby simulate an EAS (Scotti and Osteria 2016).

EUSO-TA

This instrument was a small ground based instrument, with a FoV of 11^o that was installed at the Telescope Array (TA) site in USA. This was done to use some of TA’s facilities for calibration, namely the Central Laser Facility (CLF) and the Electron Light Source (ELS). It started its operation in March 2015 (Bisconti 2016).

(25)

1.4. EUSO PROGRAM Jonah Ekelund EUSO-SPB

This was a super pressure balloon (SPB) flight with an updated version of the instrument flown on EUSO-Balloon. The balloon was launched from Wanaka, New Zealand, on April 25, 2017, and had a planned flight time of 100 days. Unfortunately, the mission had to be terminated early, 12 days and 4 hours into the mission, due to a problem with the balloon. In spite of the short mission time, 30 hour’s worth of data were recorded and successfully downloaded to ground (Wiencke and Olinto 2017).

1.4.2 Current and future missions

These missions are Mini-EUSO, K-EUSO and JEM-EUSO.

Mini-EUSO

This instrument will be sent to the ISS and from there, map the Earth’s UV emissions.

This will also be the first time that Fresnel lenses are used in space (Scotti and Osteria 2017). For more information regarding this instrument look at section 2.1.

K-EUSO

K-EUSO, also known as KLYPVE-EUSO, is a cooperation between the JEM-EUSO mission and the Russian KLYPVE mission. The name KLYPVE comes from the Rus- sian word ”ultra-high energy cosmic rays”.

K-EUSO will use a Schmidt camera and a mirror with a 4m diameter. K-EUSO is planned to operate for at least 2 years attached to the Russian MRM-1 ISS module (Casolino et al. 2017).

JEM-EUSO

The JEM-EUSO, see Figure, 1.7 is the instrument that all the previous technology demonstration missions lead up to. A comparison between JEM-EUSO and some of the precursor missions can be seen in Table 1.1.

Like Mini-EUSO, JEM-EUSU will be located on ISS. See Table 1.1 for a comparison between JEM-EUSO and some of the precursor instruments.

(26)

Jonah Ekelund 1.4. EUSO PROGRAM

Figure 1.7: Artist’s interpretation of the JEM-EUSO instrument attached to ISS. Credit: J. Adams et al. (2014)

Table 1.1: Defining parameters of the JEM-EUSO, Mini-EUSO, EUSO-Balloon and K-EUSO instruments.

JEM-EUSO Mini-EUSO EUSO-Balloon K-EUSO

Lens Shape Circular Circular Square Circular

Lens Area 4.5 · 10⁴ cm² 490 cm² 1 · 10⁴ cm² 2.2 · 10⁴ cm² Resolution 560 m 5.4 · 10³m 175 m

FoV/Pixel 0.08^◦ 0.8^◦ 0.23^◦ 0.058^◦

1.4 · 10⁻³rad 0.01 rad 4 · 10⁻³rad 1 · 10⁻³rad

FoV/PDM 3.84^◦ ±19^◦ ±5.5^◦ ±0.27^◦

6.7 · 10⁻²rad ±0.3 rad 9.6 · 10⁻²rad ±4.7 · 10⁻³rad

26.7 km 2.6 · 10²km 3.8km

N^o PDMs 137 1 1 52

N^o Pixels 315,648 2,304 2,304 119,808

(27)

1.5. REFERENCES Jonah Ekelund

1.5 References

Adams, J. H. et al. (2015). “The EUSO-Balloon pathfinder”. In: Experimental Astron- omy 40.1, pp. 281–299. issn: 1572-9508. doi: 10.1007/s10686-015-9467-9. url:

https://doi.org/10.1007/s10686-015-9467-9.

Adams, J.H. et al. (2014). “Calibration for extensive air showers observed during the JEM-EUSO mission”. In: Advances in Space Research 53.10. Cosmic Ray Origins:

Viktor Hess Centennial Anniversary, pp. 1506–1514. issn: 0273-1177. doi: https:

//doi.org/10.1016/j.asr.2013.10.009.

Alvarez-Mu˜niz, Jaime et al. (2002). “Hybrid simulations of extensive air showers”. In:

Phys. Rev. D 66 (3), p. 033011. doi: 10.1103/PhysRevD.66.033011. url: https:

//link.aps.org/doi/10.1103/PhysRevD.66.033011.

Anx-Meador, Phillip (2017). Orbital Debris - Quarterly News. Tech. rep. NASA Orbital Debris Program Office.

Aslanov, Vladimir and Vadim Yudintsev (2013). “Dynamics of large space debris removal using tethered space tug”. In: Acta Astronautica 91, pp. 149–156. issn: 0094- 5765. doi: http : / / dx . doi . org / 10 . 1016 / j . actaastro . 2013 . 05 . 020. url:

http://www.sciencedirect.com/science/article/pii/S0094576513001811.

Bertaina, M., A. Cellino, et al. (2015). “JEM-EUSO: Meteor and nuclearite observations”. In: Experimental Astronomy 40.1, pp. 253–279. doi: 10.1007/s10686-014- 9375-4. url: http://dx.doi.org/10.1007/s10686-014-9375-4.

Bertaina, M. and E. Parizot (2014). “The JEM-EUSO mission: a space observatory to study the origin of Ultra-High Energy Cosmic Rays”. In: Nuclear Physics B - Proceedings Supplements 256. doi: http://dx.doi.org/10.1016/j.nuclphysbps.

2014.10.033.

Billoir, Pierre (2014). “The Cherenkov Surface Detector of the Pierre Auger Observa- tory”. In: Nuclear Instruments and Methods in Physics Research Section A: Acceler- ators, Spectrometers, Detectors and Associated Equipment 766. RICH2013 Proceed- ings of the Eighth International Workshop on Ring Imaging Cherenkov Detectors Shonan, Kanagawa, Japan, December 2-6, 2013, pp. 78–82. issn: 0168-9002. doi:

http://dx.doi.org/10.1016/j.nima.2014.05.013.

Bisconti, Francesca (2016). “EUSO-TA prototype telescope”. In: Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 824. Frontier Detectors for Frontier Physics: Proceedings of the 13th Pisa Meeting on Advanced Detectors, pp. 603–605. issn: 0168-9002. url:

http://www.sciencedirect.com/science/article/pii/S0168900215011948.

Casolino, M. et al. (2017). “KLYPVE-EUSO: Science and UHECR observational capa- bilities”. In: 35th International Cosmic Ray Conference - ICRC 2017.

(28)

Jonah Ekelund 1.5. REFERENCES Collaboration, The Pierre Auger (2015). “The Pierre Auger Cosmic Ray Observatory”.

In: Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 798, pp. 172–213. issn: 0168- 9002. doi: http://dx.doi.org/10.1016/j.nima.2015.06.058.

Coutu, St´ephane (2016). “Highlights from the Pierre Auger Observatory”. In: Nuclear and Particle Physics Proceedings 279. Proceedings of the 9th Cosmic Ray Interna- tional Seminar, pp. 39–46. issn: 2405-6014. doi: http://dx.doi.org/10.1016/

j.nuclphysbps.2016.10.007. url: http://www.sciencedirect.com/science/

article/pii/S2405601416301882.

Ebisuzaki, Toshikazu et al. (2015). “Demonstration designs for the remediation of space debris from the International Space Station”. In: Acta Astronautica 112, pp. 102–113. issn: 0094-5765. doi: http://dx.doi.org/10.1016/j.actaastro.

2015.03.004. url: http://www.sciencedirect.com/science/article/pii/

S0094576515000867.

Fuglesang, Christer (2017). “The EUSO program: Imaging of ultra-high energy cosmic rays by high-speed UV-video from space.” In: Nuclear Inst. and Methods in Physics Research, A. check if published or still ”in press”. issn: 0168-9002.

Grieder, P. K.F. (2010). Extensive Air Showers [Elektronisk resurs] : High Energy Phe- nomena and Astrophysical Aspects A Tutorial, Reference Manual and Data Book.

1. Berlin, Heidelberg: Springer Berlin Heidelberg. isbn: 9783540769415.

GUERRA, Gabriele et al. (2017). “Active Space Debris Removal System.” In: INCAS Bulletin 9.2, pp. 97–116. issn: 20668201.

Hanlon, William F. (2008). “THE ENERGY SPECTRUM OF ULTRA HIGH EN- ERGY COSMIC RAYS MEASURED BY THE HIGH RESOLUTION FLY’S EYE OBSERVATORY IN STEREOSCOPIC MODE”. PhD thesis. The University of Utah.

— (2016). “Recent Results from the Telescope Array Project”. In: Nuclear and Particle Physics Proceedings 279. Proceedings of the 9th Cosmic Ray International Seminar, pp. 15–22. issn: 2405-6014. doi: http://dx.doi.org/10.1016/j.nuclphysbps.

2016.10.004.

Hughes, David W. (1982). “The history of meteors and meteor showers”. In: Vistas in Astronomy 26, pp. 325–345. issn: 0083-6656. doi: http://dx.doi.org/10.

1016/0083-6656(82)90010-1. url: http://www.sciencedirect.com/science/

article/pii/0083665682900101.

Klinkrad, Heiner. (2006). Space Debris [electronic resource] : Models and Risk Analysis.

Berlin, Heidelberg: Praxis Publishing Ltd, Chichester, UK. isbn: 9783540376743.

(29)

1.5. REFERENCES Jonah Ekelund Letessier-Selvon, Antoine and Todor Stanev (2011). “Ultrahigh energy cosmic rays”.

In: Rev. Mod. Phys. 83 (3). doi: 10 . 1103 / RevModPhys . 83 . 907. url: https : //link.aps.org/doi/10.1103/RevModPhys.83.907.

Rossi, Bruno and Kenneth Greisen (1941). “Cosmic-Ray Theory”. In: Rev. Mod. Phys.

13 (4), pp. 240–309. doi: 10.1103/RevModPhys.13.240. url: https://link.aps.

org/doi/10.1103/RevModPhys.13.240.

Scotti, Valentina and Giuseppe Osteria (2016). “EUSO-Balloon: The first flight.” In:

Nuclear Inst. and Methods in Physics Research, A 824.Frontier Detectors for Frontier Physics: Proceedings of the 13th Pisa Meeting on Advanced Detectors, pp. 655–657.

issn: 0168-9002.

— (2017). “The Mini-EUSO telescope on the ISS”. In: Nuclear Instruments and Meth- ods in Physics Research Section A: Accelerators, Spectrometers, Detectors and As- sociated Equipment 845. Proceedings of the Vienna Conference on Instrumenta- tion 2016, pp. 408–409. issn: 0168-9002. url: http://www.sciencedirect.com/

science/article/pii/S016890021630585X.

Westerhoff, Stefan (2012). “The search for the sources of ultrahigh-energy cosmic rays.”

In: AIP Conference Proceedings 1505.

Wiencke, L. and A. Olinto (2017). “EUSO-SPB Mission and Science”. In: 35th Inter- national Cosmic Ray Conference - ICRC 2017.

(30)

Chapter 2 Calibration

2.1 Mini-EUSO Instrument

The Mini-EUSO instrument comprises a compact telescope with a large field of view (±19^◦), based on an optical system employing two Fresnel lenses for increased light collection. The UV light is focused onto an array of photo-multiplier tubes (PMTs) and the resulting signal is converted to digital, processed and stored via the electronics subsystems on-board.

The instrument is designed and built by the members of the JEM-EUSO collaboration. The telescope components such as photo-detectors, electronics and optics are all based upon the state-of-the-art technologies originally developed for the JEM-EUSO mission.

The main part of the Mini-EUSO instrument is made to take images in the UV (300 -450 nm) part of the electromagnetic spectrum. In addition to this main system the instrument also features both a near-infrared light (NIR) and visible light (VIS) camera to provide complementary information on the observation conditions. The instrument has a mechanical interface with the UV transparent window on the Zvezda module. A conceptual design and layout of the detector is shown in Figure 2.1.

2.1.1 Ultra-violet light sensor

The Ultra-violet light sensor can be divided into three main systems: the optics, the photon collection and the data processing.

(31)

2.1. MINI-EUSO INSTRUMENT Jonah Ekelund

Figure 2.1: Mini-EUSO conceptual design: Optical system with two Fresnel Lenses (25 cm diameter) focuses the UV light on to a focal surface consisting of a single photo detection module (PDM), containing 36 multi-anode MAPMTs (Multi-Anode PMTs), all in all 2304 pixels. Secondary detectors are the visible light and near-infrared light. Adapted from: Fuglesang (2017).

Optics

The optics consist of two Fresnel type lenses, each 25 cm in diameter. Fresnel lens design means that the construction of a large aperture lens with a short focal length is possible with a much smaller mass and volume budget, making the concept well-suited to space application.

A selective filter (BG3 bandpass filter) that only allows the UV wavelengths of interest through is placed in front of the lens.

Photon collection

The optical system focuses the incoming light onto the focal surface for effective collection. The photon collection is realized through an array of 6 × 6 multi-anode photomul- tiplier tubes (MAPMTs), each with 64 pixels, resulting in a readout of 2304 channels.

Signals are pre-amplified and converted to digital before being passed to the data processing unit.

(32)

Jonah Ekelund 2.1. MINI-EUSO INSTRUMENT

Table 2.1: Mini-EUSOmain characteristics.

Parameter Value

Dimensions 37 × 37 × 62 cm³

Weight 30 kg

Power consumption 60 W

Link Power from the ISS (28 V)

Location UV transparent window of Russian Service Module

Operational requirements Work only in ISS local night. Work only with ISS main axis parallel to velocity vector.

Main observable UV light 300 - 450 nm

Spot size 3.6 - 3.8 mm RMS

Lenses Two, double-sided 25 cm diameter Fresnel lenses lens area 490 cm², material PMMA

Spatial resolution 5.3 · 10³ m

Focal Surface 1 PDM composed of 36 MAPMT, each 64 channels (total 2304 pixels)

16.7×16.7×19.24 cm³ FoV/Pixel 0.8^◦, 0.01 rad

FoV/PDM ±19^◦, ±0.3 rad, 2.6 · 10² km

Temporal resolution 2.5 µs and above (averages in 10 ms, 100 ms and 1 s)

Pointing Nadir

Data processing

Multiple trigger levels are used to filter out noise and identify events of interest. Rele- vant data is then stored at regular time intervals depending upon the complexity of the trigger. Data transfer to Earth takes place physically via the delivery of a hard drive and there is no telecommunication with the ISS systems or directly to ground from the instrument.

ASIC (Application Specific Integrated Circuit) Single photon counting data are read from ASIC board (one each PMT, for a total of 64 channels for each Spaciroc 3 ASIC from the Omega group) every 2.5 µs = 1 GTU (Gate Time Unit). All data is taken in photon counting mode and passes through the PDM-DP (PDM-Data processing) system before being sent to the CPU.

Triggering is performed on 128 GTU frames at a time. If the data passes the trigger

Calibration and evaluation of the secondary sensors for the Mini-EUSO space instrument

Calibration and evaluation of the secondary sensors for

the Mini-EUSO space instrument

Master’s Thesis

Calibration and evaluation of the secondary sensors for the Mini-EUSO space instrument

Contents

List of Figures

List of Tables

Chapter 1 Introduction

Chapter 2 Calibration