Passive optical detection of submillimeter and millimeter size space debris in low Earth orbit

Understanding of the space debris environment and accuracy of its observation-validated models are essential for optimal design and safe operation of satellite systems. Existing ground-based optical telescopes and radars are not capable of observing debris smaller than several millimeters in size. A new experimental and instrumental approach – the space-based Local Orbital Debris Environment (LODE) detector – aims at in situ measuring of debris with sizes from 0.2–10 mm near the satellite orbit. The LODE concept relies on a passive optical photon-counting time-tagging imaging system detecting solar photons (in the visible spectral range) reflected by debris crossing the sensor field of view. In contrast, prior feasibility studies of space-based optical sensors considered frame detectors in the focal plane. The article describes the new experimental concept, discusses top-level system parameters and design tradeoffs, outlines an approach to identifying and extracting rare debris detection events from the background, and presents an example of performance characteristics of a LODE sensor with a 6-cm diameter aperture. The article concludes with a discussion of possible sensor applications on satellites. & 2014 IAA. Published by Elsevier Ltd. All rights reserved. 1. Submillimeter and millimeter debris in low-Earth orbit Artificial satellites, rocket bodies, and their fragments orbit the Earth in increasing numbers and present danger to operational spacecraft, especially in low-Earth orbit (LEO) and geostationary orbit (e.g., [1–7]). The United States and the former Soviet Union, now Russia, have been operating a combination of radar and optical means for cataloging and monitoring space objects since 1960s [3,7– 9]. Other countries in Asia and Europe, particularly the People's Republic of China and the European Space Agency (as well as national programs in France and Germany), strive to expand their capabilities in the space situational awareness. ll rights reserved. The knowledge of the deteriorating space debris environment and accuracy of its observation-validated models are essential for optimal design and safe operation of satellite systems. The U.S. Space Surveillance Network routinely detects, tracks, and catalogs more than 16,000 orbiting space objects larger than 5–10 cm in LEO, with 400,000 observations taken each day [10]. Much more numerous debris smaller than several centimeters also orbit the Earth. Estimates put the number of debris with sizes from 1 mm to 10 cm in tens of millions and smaller than 1 mm in trillions [3]. Properties of such objects are characterized statistically rather than individually. A collision with a large,45 cm, object would likely cause a loss of a spacecraft and its catastrophic breakup [3], creating numerous fragments. Just one anti-satellite weapon test by the People's Republic of China in January 2007 [11] contributed more than 150 thousand debris one centimeter and larger in LEO [12]. Large orbiting objects are individually observed and cataloged and accidental M. Gruntman / Acta Astronautica 105 (2014) 156–170 157 collisions with them can thus be sometimes avoided. Populations of very small debris,o0.1 mm, can be experimentally studied and statistically characterized by bringing exposed surfaces back to Earth from orbit (e.g., LDEF, SMM, EURECA, SFU, and Space Shuttle) and examining collision effects [7]. These very small debris could cause degradation of surfaces and perhaps damage unprotected spacecraft components. This work concentrates on the population of debris with intermediate sizes from 0.1–10 mm. Such submillimeter (0.1–1 mm) and millimeter (1–10 mm) debris are typically too small to be detected by existing optical and radar means, and there are too few of them to be described by studying exposed surfaces. Because of this gap in observations [7,13,14], one has to rely on modeling of their populations based on difficult to accurately predict processes in breakup of complex bodies in high-velocity impacts. At the same time possible damage to spacecraft by submillimeter and millimeter debris ranges from surface degradation to possible loss of spacecraft capability or its components [3]. Atmospheric drag lowers orbital altitudes of space objects in LEO and causes their eventual reentry into the atmosphere. (Solar radiation pressure may also become important for some objects with large area-to-mass ratios.) Drag acceleration acting on an orbiting body is inversely proportional to its characteristic size. Consequently, the acceleration becomes higher and the lifetime in the orbit shorter with the decreasing size of the body. For example, drag would remove an aluminum sphere 1-mm in diameter from an initial 400-km altitude orbit within a couple of weeks. Such altitudes are especially important for human spaceflight (International Space Station) and atmospheric drag effectively reduces danger of small debris in such orbits. At higher altitudes from 700–1500 km with numerous application satellites, lifetimes of orbiting objects may exceed dozens or hundreds of years, resulting in accumulation of debris. Existing ground-based optical and radar systems have inherent limitations in measuring dangerous to satellites submillimeter and millimeter debris at these altitudes. First, only ground-based systems near the equator can probe space objects in low-inclination low-Earth orbits, limiting capabilities of facilities in the continental United States. In addition, ground optical systems can observe LEO objects only during short time intervals at local dawn and dusk. Second, for radar, the efficiency of electromagnetic radiation scattering, characterized by radar cross-sections, rapidly decreases for small objects, e.g., [15]. For example, the radar cross-section of a perfectly conducting sphere with a diameter at least three times larger than the wavelength approximately equals its geometric crosssection. For spheres with diameters less than one-third of the wavelength, radar cross-sections drop precipitously (known as Rayleigh scattering) with decreasing diameters. Most existing and planned (such as the upgrade of the Space Fence in the United States) radars operate in the S-band frequencies with wavelengths 6–15 cm. For example, a perfectly conducting 1-mm diameter sphere would have the radar cross-section almost 5 orders of magnitude smaller than its geometric cross-section for a typical S-band radar operating at the 3 GHz frequency (wavelength 10 cm). For an X-band radar at the 10 GHz frequency (wavelength 3 cm), the corresponding radar crosssection would be 800 times smaller. Atmospheric absorption (e.g., [16]) fundamentally limits possible increase of radar frequencies beyond the X-band required for further lowering the upper size limit of the Rayleigh scattering region. The German Fraunhofer Institute's Tracking and Imaging Radar (TIRA) near Bonn operating autonomously or in a bistatic mode (with the Max-Planck-Society's radio telescope at Effelsberg) could detect orbiting debris as small as 1–2 cm at altitudes up to 1000 km [5]. Most capable NASA radars today, the Haystack and Goldstone, can detect debris down to several millimeters in size at important for human spaceflight 400-km altitudes [7,17]. In recent years, the upgraded Haystack and Haystack Auxiliary (HAX) radars also demonstrated detection of such debris at 800 km altitude [18] where numerous satellites operate and where debris accumulate due to significantly diminished atmospheric drag. Satellite-based (space-based) optical sensors offer important opportunities especially for detecting large orbiting objects as was demonstrated, for example, by the SpaceBased-Visible sensor on the Midcourse Space Experiment, or MSX [19,20]. A number of feasibility studies examined capabilities of possible space-based passive optical sensors for measuring millimeter and/or centimeter size debris and larger in LEO [13,14,21–23] as well as observing objects in geostationary orbit [13,14,21,22,24,25]. (For the sake of completeness, we also note here an assessment of a space-based radar for detecting objects in LEO [26].) To the best of my knowledge no study has specifically looked at optical observation of submillimeter debris. At the same time, a recent NASA Handbook [7] especially emphasizes the importance of such submillimeter debris by stating that “from the safety and satellite operations perspective, there is an immediate need for a large and dedicated meteoroid and orbital debris sensor to monitor and update the populations between 0.1 and 1.0 mm.” A special feasibility study sponsored by the European Space Agency specifically focused on closing “the existing knowledge gap in the space debris population in the millimeter and centimeter regime” [14]. Scarcity of experimental data leads to a one-order-ofmagnitude disagreement among debris models (ORDEM, MASTER, SDPA) in prediction of fluxes of debris smaller than 1 cm [7]. This article describes a new space-based experimental and instrumental concept – the Local Orbital Debris Environment (LODE) detector – to observationally characterize 0.2–10 mm debris near the satellite path important for space system design and for mission assurance and safe operation. It concentrates on top-level system parameters and design tradeoffs, outlines an approach to identifying and extracting rare debris detection events from the background, and presents an example of performance characteristics of a LODE sensor with a 6-cm diameter aperture. Specific designs of the optical part of M. Gruntman / Acta Astronautica 105 (2014) 156–170 158 the sensor and its focal plane detector are beyond the scope of the article. While the LODE concept primarily focuses on observation of debris in LEO, it can also probe micrometeoroid fluxes, debris at geostationary orbit, and perhaps dust in the lunar environment. 2. LODE sensor concept The Local Orbital Debris Environment sensor is based on a passive optical photon-counting time-tagging imaging system deployed on a spacecraft and detecting solar photons (in the visible range) reflected by debris crossing its field of view (Fig. 1). The sensor would thus characterize the debris population locally within a certain distance from the satellite orbit path. Sensor pointing in a general antisolar direction on a spacecraft in dawn–dusk (equator crossing) sun-synchronous orbit offers most favorable observation geometry in LEO. The instrument could fly on a dedicated small satellite or as a hosted payload. Prior flown or studied space-based optical sensors for debris detection used CCD (charged-coupled device) detectors in the focal plane [13,14,19–22,24,25]. The CCDs are essentially frame detectors. A passage of an object across the field of view during the frame accumulation time interval results in an image of a streak across a pattern of fixed stars and diffuse background. Arrival of 20 or more photons into the instrument is usually needed for a signal above the intrinsic noise in a single CCD pixel. Reliable detection of a rare debris streak requires a sequence of multiple lighten-up pixels and correspondingly at least a few hundred debris-reflected photons. The LODE concept differs from considered in the past approaches by relying on a different type of focal-plane detectors that enable the desired capabilities. Specifically, the resulting reduced number of photons needed for detection of an object crossing the sensor field of view (FOV) opens a way for observing smaller submillimeter and millimeter debris. To the best of my knowledge, in Fig. 1. Local Orbital Debris Environment (LODE) concept for in situ characterization of debris population in the 0.2–10 mm size range near the satellite orbit by passive optical detection of debris-reflected solar photons in the visible spectral range. FOV – the field of view of the sensor. addition to purely astronomical applications, e.g., [27–30], such spaceand ground-based visible-range photoncounting time-tagging imaging systems were considered in the past only for imaging of complex three-dimensional objects (such as satellites) and non-local (i.e., at large distances) detection of centimeter size debris [23,31,32]. Nobody considered applications of such systems for detection of submillimeter and millimeter debris locally near the satellite path. The LODE instrument relies on a microchannel-plate (MCP) based position-sensitive detector (PSD) in the focal plane. Such detectors have been used in laboratory and space applications registering and determining coordinates of individual particles and photons since 1970s (see, e.g., reviews [33–39] and references therein). For example, two currently operating Hubble Space Telescope instruments include detectors of this type [29,30]. While MCP-based detectors are less common in satellite-based astronomical and astrophysical instruments, they are routinely used (though without photocathodes), in space physics experiments. Consequently, the associated technology readiness levels of MCP detectors, readout electronics, and high voltage (usually not exceeding 2–3 kV) power supplies are very high. Such photon-counting imagers are fundamentally different from CCD-based systems. In contrast to CCDs, an MCPbased PSD does not accumulate a frame image but determines in real time the coordinates (in the focal plane) and the exact detection time of each registered individual photon and stores them in a memory. A computer can then use such data in asynchronous post-processing to form an image accumulated during any desired time interval while preserving timing information, or time tags, of each and all registered photons. These unique capabilities of photoncounting time-tagging imagers enable the proposed concept. Fig. 2 shows the schematic of a LODE imager. It consists of an optical system, a simple telescope, with the entrance aperture diameter d0 and field of view angle ω0 and a photon-counting imager at the focal plane. A baffle reduces light from off-axis bright sources such as the Moon and atmospheric glow. Due to spacecraft orbital motion, the relative velocity of debris would be predominantly in the direction opposite to the spacecraft velocity vector. Therefore, pointing the LODE sensor normally to the orbital plane would maximize the number of crossings of the instrument FOV by debris. When a FOV crossing occurs, registered photons would form a well-determined sequence, or line (a “trajectory”), in three-dimensional x; y; t ð Þ “sensor space” of two spatial coordinates-directions x and y in the detector focal plane (similarly to a streak captured by CCD detectors) and additionally in the dimension of time. In contrast, background photons would form a random pattern of counts in spatial and time dimensions. Registration of only several debris-reflected photons is sufficient to reliably identify an object FOV crossing. Therefore, arrival of a relatively small number, only a few dozen, of photons into the LODE sensor with a realistic photon detection efficiency is needed. Consequently, this sensitive technique offers observational access to submillimeter and millimeter debris inaccessible to conventional radar and optical means. The observation Fig. 3. Schematic of a microchannel-plate position-sensitive detector. Such photon-counting time-tagging imaging detectors determine in real time the coordinates and moment of time of detection of each registered incident photon. The detector consists of a transparent window with the photocathode, microchannel plate stack, and anode-collector in a vacuum-sealed tube with connectors for high voltage power and for signal output. A photoelectron produced by an incident photon from the photocathode is accelerated toward the MCP stack that converts it into an avalanche of 10–10 electrons falling on the readout anode-collector. Fig. 2. Schematic of the LODE sensor: optical telescope with the aperture diameter d0 and field of view ω0 and a photon-counting time-tagging imaging detector in the focal plane. An object of a diameter a crosses normally the field of view at the maximum detection distance hM; also shown is a trajectory crossing the instrument axis at an angle β. M. Gruntman / Acta Astronautica 105 (2014) 156–170 159 of a FOV crossing event would constrain in an important way a combination of parameters characterizing this particular debris trajectory, size, albedo, and velocity. The count rate limitations of MCP-based PSDs results, as shown below in Section 3, in a relatively narrow field of view of the LODE sensor, a small fraction of one degree. In contrast, passive CCD-based optical sensors usually relied on relatively wide fields of view. For example, the MSX's Space-Based Visible sensor had a 15-cm aperture telescope with the 1.41 6.61 field of view [19,20] while other considered systems had, for example, 7.5-cm aperture and 201 FOV [21]; 10-cm aperture and 11 FOV [13]; and 20-cm aperture and 61 FOV [22]. The LODE imager could thus be, in an important sense, complementary to CCDbased optical sensors and perhaps they could be combined in a package offering unique cross-calibration and debrischaracterizing capabilities. 3. LODE sensor performance characteristics A photon-counting imaging detector (Fig. 3) consists of a photocathode followed by a stack of MCPs and an electron collector-anode with its readout circuitry determining the coordinates of each registered incident photon [33–39]. For visible spectral range, the photocathode is usually deposited on the back side of a transparent entrance window, with photoelectrons accelerated across a small gap (so-called proximity focusing) in a compact design towards the entrance of the MCP. The MCP stack of two or three plates converts the photoelectron into an avalanche of 10–10 electrons falling on the anode. A sealed vacuum tube confines the photocathode side of the window, the MCP stack and the collector-anode (Fig. 3). The electronic circuitry determining photon coordinates and detection moments of time as well as the high-voltage (2–3 kV) power supply are outside the sealed tube. A number of readout designs have been developed for PSDs based on charge division, multianode discrete elements, and delay lines [33–45]. Various approaches are particularly suitable for different applications and the design can be optimized to achieve specific LODE requirements. The PSD sensitive area can vary from 15–100 mm with spatial resolution up to 2000 pixels in each dimension. The detectors can also determine the moment of photon detection with accuracy as high as a fraction of 1 ns (e.g., [46,47]). The LODE sensor requires, as discussed below, much more modest spatial resolution and timing accuracy of the PSD.