A New Global Array of Optical Telescopes: The Falcon Telescope Network

A growing number of nations and commercial companies have developed the technology necessary to manufacture and launch their own orbital space systems. Although a few global powers have historically dominated this field, such as the United States, Russia, the European Community, China, India, Japan, and Canada, other countries are now becoming space-faring nations such as Mexico, Chile, Argentina, Taiwan, and Australia to name a few.

The fact that more and more countries have access to this type of technology has enabled significant advances in the field of communications, space research, and the study of the Earth. Nonetheless, this abrupt increase in the number of Earth-orbiting objects has also caused problems regarding managing and controlling the orbital population, resulting in collisions between satellites operating in low-altitude orbits. This scenario makes the development of new techniques to characterize artificial satellites a priority. These techniques would enable the development of tools to accurately determine both the number as well as the type of objects that are in various Earth orbits.

For several decades now, astronomy observation techniques have been applied in the precise characterization of artificial satellites located at different orbits. Early on, authors such as Tousey (1957) and Kissell & Tyson (1964) devoted themselves to studying the properties associated with the optical detection of satellites, such as visibility, photometric orbit determination, and the temperature that a satellite reaches as a result of radiation exchange with the environment. On the other hand, in the 1960s and 70s, researchers from the United States Air Force worked on characterizing satellites through the phenomenon of polarization (Stead 1967) and studying the visible reflectance spectra (Lambert 1971). At present, researchers from around the world have made a considerable effort to improve studies of space object properties through techniques that do not require high spatial resolution, known as non-resolving techniques (Payne et al. 2002, 2006; Schildknecht et al. 2005; Hall et al. 2007; Seitzer et al. 2007; Schildknecht et al. 2009; Tippets et al. 2015).

Thus, Seitzer et al. (2007) were able to characterize geostationary satellites using photometric observations obtained through the use of Morgan & Johnson system filters and also by using the Space Object Identification (SOI) in Living Color (SILC) technique, concluding that the different satellite structures (bus) possess distinctive photometric properties that can feasibly be classified. This was improved by Bédard et al. (2012), who, using more refined observational techniques, determined that although satellites with the same bus may have similar photometric characteristics, they can be differentiated by eliminating a possible cross-sectional characterization of satellites.

Along with photometric analysis, there have been breakthroughs based on spectroscopic studies of artificial satellites. One of the most used traditional techniques in spectral observations is "long-slit spectroscopy." This technique refers to the use of an elongated slit aperture located between the object signal and the diffraction grating, in order to select a bounded region of the incident light beam, which is then dispersed by a primary element in order to obtain the detail of its spectral signature. This type of spectral observation in artificial satellites has previously been carried out and reported; however, the instrumentation and the long-slit observation technique is complex, mainly due to the type of movement that these satellites involve. Lambert (1971), using a scanning spectrograph, was able to obtain reflectance spectra in satellites and to detect the existence of aluminum in their coverage. This is explained by the fact that these materials are developed with components that are capable of protecting existing instrumentation on the interior of the satellites from external elements, such as friction or large changes in the temperature, to which they are exposed. Likewise, Jorgensen et al. (2004) discovered a reddening effect of artificial space objects when compared to laboratory samples.

Thus, several studies have harnessed the spectroscopic observation technique in order to find unique characteristics in Earth orbit systems. However, traditional observation techniques require much work and are difficult to carry out regarding systematic observations, something that is essential for the characterization of artificial satellites. This is due to the fact that it is only possible to obtain spectroscopic properties of one object at a time, requiring large amounts of observation time to generate sufficient databases for characterization studies. These observations are also very risky, as precise positioning is necessary to place an object moving at high angular speeds across the sky, as is the case for low-altitude satellite observations, in a very small area on the observation equipment (slit aperture).

In 1970, Hoag and Schroeder introduced a new technique for spectroscopic observations: slitless spectroscopy. These authors showed that slitless observations have a number of advantages: first, the spatial observation coverage allows spectroscopic surveys of different objects simultaneously. Along with this, the object center does not need a high guidance accuracy, because the light coming from the object does not need to pass through a slit. Finally, its configuration is simple due to the use of simple diffraction gratings. Nonetheless, an important disadvantage to highlight is that the information obtained is of very low spectral resolution. Additionally, there is no kind of filtering in the observation of an object; therefore, the different spectral moments may be confused within a same observation. Recently, Tippets et al. (2015) and Dunsmore et al. (2016) showed results of using slitless spectroscopy on glinting communication GEO satellites providing direct evidence that the source of the satellite glints are the solar panels.

In addition to using photometric and spectroscopic techniques to characterize artificial satellites, carrying out simultaneous observations performed from different perspectives and lighting conditions to study satellites has been proposed (Hall et al. 2012). One of the most noteworthy studies in this regard is Fulcoly et al. (2012), who, using photometric modeling tools, were able to demonstrate the power of using simultaneous spatial photometry from multiple telescopes to determine the basic forms of a satellite.

These studies show that determining orbital element properties is fundamental for the development of technology in the short term, transforming this into one of the priority areas in relation to the study of extraterrestrial sciences.

Since the construction of the first optical systems for the study of the universe, researchers have become devoted to improving and expanding various tools in order to answer some fundamental questions. Engineering wonders have had to be developed in order to answer questions as essential as how was the universe created? or what were the initial conditions that formed the space that surrounds us today? These technological innovations have allowed us to detect energy coming from astronomical sources located at immense distances.

In this context, one of the methods designed to answer these questions was optical and infrared telescopes of ever-increasing size. During the first half of the 20th century, telescopes such as the Victor Blanco of the Astronomical Observatory in Cerro Tololo (Chile), with a 4 m primary mirror, the Hale telescope of the Mount Palomar observatory in the United States, with its 5.1 m primary mirror, or the 6 m Large Altazimuth Telescope, belonging to the Special Astrophysical Observatory (Russia), dominated observational astronomy as the largest optical telescopes in the world. However, these were quickly overshadowed by a new generation of much larger telescopes in the late 20th and early 21st century (e.g., 10 m Keck Telescopes, USA; 8 m Very Large Telescope (VLT) and Gemini in Chile, 10 m Southern African Large Telescope, among others). This new generation of telescopes brought human knowledge of the universe to levels that were unthinkable only 50 years ago with the discovery of, for example, more than 3000 exoplanets, or galaxies located at such distances that at the time of issuing the detected signal, the universe was no more than 400 million years old.

However, our search to understand our environment does not end at this point. Further, development in engineering, computing, and design have allowed us to develop a new generation of giant or extremely large telescopes: the 26 m Giant Magellan Telescope, located in the Cerro Las Campanas Observatory in the north of Chile, the 39 m European Extremely Large Telescope (E-ELT) of the European Southern Observatories also built in the Atacama desert of Chile, and the Thirty Meter Telescope project developed by a consortium of institutions, but still in the process of designating a final location.

Despite enormous technical and human advances employed in the construction of these high-level research centers, many current advanced systems remain insufficient to successfully pursue lines of research that require systematic and continuous temporal studies. Studies to detect extrasolar planets using the technique of stellar variability involve making temporal observations over hours or even days (Charbonneau et al. 2000). Defining supernovae light curves requires taking data night after night for months (Filippenko 1997). Furthermore, observing the optical counterparts of Gamma-ray bursts (GRBs) requires rapid responses from the telescopes as the brightness of these transient astronomical phenomena can diminish below detection limits in a matter of hours (Gehrels et al. 2009).

To respond to these events, a series of laboratories have been developed that are devoted to time domain observations. They seek to continuously collect data in the greatest possible time range through the construction of global arrays of telescopes. Recently, a series of these facilities have been established, some larger and more developed than others. Among these, we can highlight the Las Cumbres Observatory Global Telescope Network (Brown et al. 2013), composed of a network of 42 telescopes of various sizes throughout the world, and the Skynet Robotic Telescope Network which consists of 11 telescopes located at 17 observatories around the world, such as the PROMPT telescopes (Reichart et al. 2005), with locations in the United States, Canada, Italy, Australia, and Chile (https://skynet.unc.edu/). We can also add the Bell Observatory/Crimean Astrophysical Observatory/Kitt Peak National Observatory (BCK) Network of optical telescopes (McGruder et al. 2015), the Master Global Robotic Network (Lipunov et al. 2010), the Cooperation through Education in Science and Astronomy Research Telescope Network (CESAR, Cuesta & Vaquerizo 2014), among others, to these telescope networks. Finally, there is the International Scientific Optical Network (ISON), which is composed of approximately 80 telescopes worldwide dedicated primarily to observations and studies of manmade satellites, space debris, GRB afterglows, and asteroids (Molotov et al. 2008).

In this paper, we present a new global network of small aperture telescopes, developed by the Center for Space Situational Awareness Research of the Department of Physics at the United States Air Force Academy (USAFA), and dedicated to identifying and characterizing artificial satellites and studying the near universe: the Falcon Telescope Network (FTN). This is a more up-to-date paper than the conference proceedings of Chun et al. (2014). Thus, we have structured our study as follows. In Section 2, we describe the telescope network, with particular emphasis on the instrumental characteristics. In Section 3, we describe each of the sites that make up the global network, including their geographical characteristics and the local institutions that manage each telescope. In Section 4, we explain the control system designed for the remote control of each of the nodes. In Section 5, we present the first scientific results obtained to date, based on the data obtained by the nodes of the FTN that are already operating. In Section 6, we present a summary of the education and community outreach programs that have been designed by the FTN team, to be implemented with each of the nodes.

Even though there are currently various global arrays of optical telescopes, most of them are composed of systems with different optics or important instrumental differences. This may cause long-term temporal studies to present inhomogeneities that could affect their results. As one of the goals of the FALCON project is the simultaneous observation of the same object from distant locations, it was decided that the network would be composed of "cloned instrumentation," i.e., each FALCON node must be a true and accurate copy of the other nodes that make up the network (see Figure 2). In addition, each node must have the ability to be managed remotely from anywhere in the world or be autonomously tasked and controlled.

Zoom In Zoom Out Reset image size

Therefore, after considering factors such as the availability of instrumentation, exportability and ease of acquisition, geographic location, installation coordination and logistics, robotization, automation and remote control capability, and financing, the company Officina Stellare was selected as the main supplier. The selected telescope was the Pro RC 500 Ritchey–Chrétien model, which has a 0.5 m (20 inch) primary mirror, an f/8.1 focal ratio, and a 4000 mm focal length. The primary mirror is constructed using zero expansion Astro-Sitall glass, while the main tube is designed on the basis of a carbon and aluminum alloy. Focusing is achieved by moving the secondary mirror using the SkyX autofocusing feature for reaching optimal focus.

The telescope is on a Paramount ME2 (German equatorial mount) from Software Bisque. This type of mount locates the weight of the telescope and the counterweights in positions opposite to the main focus of the mount (or polar axis or right ascension axis), ensuring stability, versatility, and precision at the time of tracking.

The buildings that house the telescopes and control systems are covered by a 3.65 m diameter (12 foot diameter) fiberglass structure, with a clamshell aperture. The dome supplier is the company Astro Haven Enterprises. The building is the only component in the network that can be different. These are designed and built by the telescope host institutions, and respond to different needs and capabilities, mainly associated with the geography of each place or available financing and resources of the partner. For example, while it was chosen to build based on pre-formed concrete panels or with steel in some nodes, in Chile, due to its complex geography and high seismicity, it was selected to build based on reinforced masonry with an internal structure of steel and reinforced concrete beams.

Below is a list of hardware and software components for each FALCON telescope.

• (1)  
  Hardware.

  • (a)  
    Officina Stellare 20 inch, f/8.1 Ritchey–Chrétien.
  • (b)  
    Apogee Alta F47: 1024 × 1024, 13 μm pixel, 11 FOV (065 pixel⁻¹ or 50'' mm⁻¹).
  • (c)  
    Apogee 9-position Filter Wheel: B, V, R, g', r', i', z', 100 lines/mm diffraction grating, exoplanet filter.
  • (d)  
    Astro Haven 12-foot Clamshell Dome.
  • (e)  
    Boltwood Cloud Sensor II.
  • (f)  
    Symmetricom GPS Receiver (BC637PCIE): 170 ns rms accuracy (UTC), 100 ns clock resolution for time requests (sub-ms timing for computer applications).
  • (g)  
    Advantech Gen II Computers w/Intel SSD DC S3700 Series.
  • (h)  
    Synology NAS (8–20 TB).
  • (i)  
    Cisco ASA 5505 Router.
  • (j)  
    Cisco Small Business 200 Series Switch.
  • (k)  
    Tripp Lite Uninterruptible Power Supply (UPS) SMART1000/2000.
  • (l)  
    Foscam Pan/Tilt Internet Protocol (IP) Camera (FI8919W and FI9826W).
• (2)  
  Software.

  • (a)  
    Software Bisque TheSkyX Suite.
  • (b)  
    FalconExec.

The camera installed on all telescopes of the global FALCON network is an Apogee Alta SERIES model F47 brand camera. The camera is equipped with a 1024 × 1024 pixel E2V CCD47-10 sensor with back-illuminated architecture and a nominal distance of 13 μm between pixels. The sensor has a quantum efficiency of 52% at 400 nm and 96% at 500 nm. The sensor size is 13 × 13 mm (18.88 mm diagonal), where each pixel has a size of 13 μm. The sensor is cooled by thermoelectric coolers using a forced air system, reaching −25 °C. The typical plate scale of the system is 065 per pixel resulting in a field-of-view of approximately 11'.

Each telescope has an Apogee AFW50-9R filter wheel mounted on the Cassegrain focus of the telescope. The filter wheel structure allows the installation of 9 different circular filters of 50 mm in diameter. Selecting the best photometric system for a global array of telescopes is not a simple task. The global distribution of all FALCON nodes will allow the performance of a variety of astronomical projects, from searching for exoplanets by studying the curves of light in prolonged timescales, to searching for the electromagnetic triggers of gravitational waves. On top of this, we should add the capability of performing observations of the same object from different latitudes of the world co-temporally. This allows us to, for example, study the structural properties of geosynchronous satellites, thanks to the ability to detect various sections lit differently from symmetrically separated positions, due to the geometry of the rays that illuminate the satellites (see Section 4).

Based on these factors, a wide filter system was selected that is commonly used by the scientific community. This should allow us to easily calibrate the photometric data obtained by the different nodes, and create a homogeneous database in a relatively straightforward way. Two commonly used wide band photometric systems include: the Sloan prime photometric system u'g'r'i'z' and the Johnson–Cousins system. Considering the detector's quantum efficiency, we decided to use the following photometric filters: B, V, R, g', r', i', z'. The selection of these filters ensures acceptable transmission between 400 to 1000 nm. All the Johnson–Cousin and Sloan prime filters are 3 mm thick, thus minimizing the need to refocus the telescope when changing filters.

Additionally, we have an Astrodon ExoPlanet Blue-Blocking filter. This filter is also 3 mm thick and has anti-reflective layers on both sides of the filter. Furthermore, it is optimized for studying light curves in bright stars (brighter than magnitude 10), thanks to its ability to block UV radiation and the blue portion of the visible spectrum, generating excellent transmission at wavelengths greater than 500 nm. There are several benefits of using such a filter when conducting exoplanet transit photometry. First, mitigation of airmass extinction differences between target and comparison stars of different colors. Second, reduction of scattered moonlight into the optical path. Third, stellar limb darkening is weaker at longer wavelengths, yielding better estimates of second and third contact times during transit events.

The filter wheel also has a 100 lines per millimeter Richardson transmission diffraction grating (100 lines/mm, 46 blaze) in one of the slots. The diffraction grating's theoretical resolution is 103 < R < 147 for a wavelength range of 350 nm < λ < 700 nm. At first glance, slitless spectroscopy appears to be a rather unusual way to obtain satellite spectral information. Nonetheless, this method can be useful since it delivers many spectra in a single image. With our modest resolution, our spectrograph will deliver more usable information than images made using a set of BVR filters (10 times as many spectral channels since a filter in a BVR system typically has a full width at half maximum of 100 nm). We performed an analysis to determine the appropriate grating dispersion and considered the major sources of error. Since our cameras have a CCD chip size of 1024 × 1024 pixels with a 13-micron pitch, using a 100 lines/mm grating will allow us to image the zero order and the +1 order in a single frame. This allows co-temporal imagery of the spectrum from 350 to 850 nm.

3.1.1. OJC-Falcon Telescope

3.1.2. NJC-Falcon Telescope

3.1.3. PSU-Falcon Telescope

3.1.4. FLC-Falcon Telescope

3.1.5. MMO-falcon Telescope

3.1.6. CBR-Falcon Telescope

3.1.7. Pindan-Falcon Telescope

3.1.8. YDR-Falcon

The YDR-Falcon is located near the small town of Yoder, Colorado approximately 38 miles (61 km) south east of the USAFA Cadet Area. The YDR-Falcon is part of a larger optical telescope complex on leased land. Unlike many of the other FALCON telescope observatories, the dome for the YDR-Falcon is not constructed as a two-level observatory, rather the dome is on a 12 inch riser and the dome base ring has an access hatch for entry and exit. Another difference between this telescope and other FALCON telescope is that this telescope is mounted onto a 30 inch sand-filled Software Bisque riser; other systems are typically mounted onto a 12 inch riser. The YDR-Falcon achieved first light on 2018 March 8.

Zoom In Zoom Out Reset image size

3.2.1. FRH-Falcon Telescope

The FRH-Falcon telescope is located on a 650-acre recreational area maintained and operated by USAFA, and located approximately 7 miles (12 km) west of the USAFA Cadet Area in the Pikes Peak region of Colorado. The observatory is already built and the Paramount ME2 is installed (Figure 5). However, to become fully operational, the telescope and all the hardware accessories need to be installed. We anticipate that the FRH-Falcon telescope will become operational in 2018.

Zoom In Zoom Out Reset image size

3.2.2. CMU-Falcon Telescope

The Colorado Mesa University Falcon Telescope will be located at The Grand Mesa Observatory situated approximately 25 miles (19 km) southeast from downtown Grand Junction, Colorado and 181 miles (291 km) west of Denver. The observatory will be a shorter building constructed with concrete tiles; and instead of a concrete pier, we will use a 48 inch sand-filled riser attached to a concrete slab isolated from the building foundation (Figure 6). The construction of the observatory is underway and we expect it to be fully operational by 2018 August.

Zoom In Zoom Out Reset image size

3.2.3. TUBS-Falcon Telescope

The TUBS-Falcon telescope will be installed at $10\buildrel{\circ}\over{.} 55$ longitude and $52\buildrel{\circ}\over{.} 28$ latitude. The site is located north of Braunschweig, about 1.2 miles (2 km) from the Institute of Space Systems of the Technical University of Braunschweig. Due to the climatic conditions in northern Germany, the construction of a solid building is required. The building is currently under construction, as can be seen in Figure 7. You can see the reinforced concrete central body and the bottom panel, which are constructional decoupled from one another to reduce transmitted vibrations the telescope.

Zoom In Zoom Out Reset image size

The FTN was designed to achieve co-temporal, simultaneous observations of geosynchronous (GEO) satellites from multiple illumination geometries as illustrated in Figure 3. While the geosynchronous belt is becoming increasingly more populated, its importance to the national security of many countries is paramount. Due to the great distance at which the geosynchronous belt is located, satellites orbiting at that altitude are non-resolved, even using the largest telescopes. Therefore, to study the properties of these objects, techniques must be developed based on non-resolving observations and measurements using broadband photometry and low-resolution slitless spectroscopy. We have conducted some preliminary spectral observations of satellite glints (i.e., a specular reflection off of a satellite, usually its large solar panel arrays) from two different telescopes, the NJC-Falcon and a smaller 16 inch telescope located on the campus of USAFA (called USAFA-16). Figure 10 shows the raw images taken from the two telescopes of a cluster of GEO communication satellites seen in the skies above Colorado. The USAFA-16 (right panel image) has a slightly larger field of view than the NJC-Falcon and hence was able to capture four satellites in an image, whereas the NJC-Falcon only observed three satellites. The satellites in the NJC-Falcon image (left panel) from left to right are DirecTV-15, DirecTV-12, and DirecTV-10. The satellites in the USAFA-16 image (right panel) from left to right are DirecTV-15, DirecTV-12, DirecTV-10, and Spaceway-1. These measurements are now being analyzed by USAFA cadets (senior physics majors) as part of their capstone research project to determine relative power capacities of different satellite manufacturers (Dunsmore et al. 2016).

Zoom In Zoom Out Reset image size

In 2016 February, the National Science Foundation (NSF) conducted a press conference in conjunction with the Laser Interferometer Gravitational Wave Observatory (LIGO) Scientific Collaboration (LSC), a partnership consisting of more than 1000 scientists from 16 countries, announcing one of the most important scientific discoveries in history: the detection of gravitational waves coming from the collision of two massive black holes 1.5 million light years away. The detection of this phenomenon not only confirms Einstein's General Theory of Relativity, but it is also the first direct detection of two colliding black holes. LIGO is a network of advanced ground-based gravitational-wave (GW) interferometers consisting of two widely separated installations within the United States located at Hanford, Washington and Livingston, Louisiana. The observatory is designed to be capable of detecting GWs emitted by the mergers of neutron stars and/or black holes in binary systems out to distances of hundreds of megaparsecs (see Abbott et al. 2016 and references therein).

In 2013 June, the LIGO Scientific Collaboration (LSC), the partnership that control the LIGO observatories and consisting of more than 1000 scientists from 16 countries, make a worldwide call to participate in electromagnetic (EM) and multi-messenger observations of GW events recorded by their detectors, using a wide range of telescopes and instruments of "mainstream" astronomy.

The recent GW events are the impetus for the creation of international collaboration dedicated to the detection of the electromagnetic counterpart of gravitational wave triggers. In this context, a new collaboration was established: the Transient Optical Robotic Observatory of the South (TOROS), a partnership composed of scientists from the Gravitational Wave Astronomy at The University of Texas Rio Grande Valley and astronomers at Texas A&M University in the United States, the National University of Cordoba in Argentina, the University of La Serena (ULS) and the Pontifical Catholic University in Chile, and the National Institute of Astrophysics, Optics, and Electronics in Tonantzintla, Puebla, Mexico (Benacquista et al. 2014; Diaz et al. 2014). The main objectives of the collaboration is to participate in observations of future detections of events detected by LIGO using a network of optical telescopes located in the southern hemisphere of the world, and to develop new facilities such as an astronomical facility in Cordon Macon, a mountain located in the Atacama highlands of northwestern Argentina. In 2016 August, the Universidad de La Serena, host institution of the MMO-Falcon signed a memorandum of understanding with the TOROS collaboration to include that node as an official instrument thus making the entire FTN an active member of the TOROS collaboration.

On 2017 August 17 at 12:41:04 UTC, a binary neutron star system (BNS) merger candidate was identified in data from the LIGO Hanford (H1) detector. The Gamma-ray Burst Monitor (GBM) on board Fermi (Bissaldi et al. 2009) detected an event ∼2 s after the GW trigger (designation GRB170817A). The optical transient was independently detected by multiple teams within an hour. This event corresponds to the first multi-messenger observations of a BNS merger, the trigger of the Gravitational Wave GW170817 (Abbott et al. 2017). TOROS participated in the search for the EM counterpart of GW170817 starting just a few hours after the trigger (Diaz et al. 2017). On the nights of August 17 and 18, we surveyed 26 nearby galaxies contained in the initial localization region using the MMO-Falcon. Additionally, ULS served as the hosting center for data taken by other telescopes also associated with TOROS.

Employing the four sites that were operational in 2015–2016, a team of USAFA faculty members, cadets and civilian high school students explored the potential future utility of the FTN for detecting exoplanets via the transit method. This endeavor began with collecting photometric data over several hours during known transits of "hot Jupiters" with apparent magnitudes 9 < m_v < 13 and transit depths >1% (∼10 mmag). Performing differential photometry with various COTS astronomical software packages, several transit signals were recovered at greater than 5σ statistical significance including a pair for the exoplanet TrES-3b from two different Falcon telescopes (Figure 11). TReS-3b is a 1.3R_Jupiter planet in a 1.3 day orbit around an apparent magnitude m_V = 12.4 G-type star 400 parsecs from Earth. Note that the ordinate for the OJC plot is normalized flux (the required input of the EXOFAST fitting program), whereas the NJC ordinate is magnitude (as required by the JKTEBOP fitting program). As a result, the transit depth in normalized flux should be about nine percent smaller than the depth in magnitudes. We label the NJC ordinate as Δm to indicate that the magnitudes shown are differential magnitudes. The rms of the residuals is 0.0026 (normalized flux) for the OJC fit, and 2.7 mmag for the NJC fit. This exercise validated Falcon telescopes as a viable resource for follow-up observations on exoplanet candidates identified by ground-based wide-FOV surveys (e.g., TrES (Alonso et al. 2004), KELT (Pepper et al. 2007)), and the FTN has been identified as a potential future contributor to follow-up observations for the space-borne Terrestrial Exoplanet Survey Satellite (TESS; J. Pepper 2018, personal communication).

Zoom In Zoom Out Reset image size

Further data collection focused on smaller planets (1–3R_Earth) orbiting "ultra-cool dwarf" (UCD) stars (spectral type M4-M9), congruent with ongoing efforts using sub-1 m ground-based telescopes to detect terrestrial exoplanets residing in their host stars' habitable zone (e.g., MEarth Nutzman & Charbonneau 2008, TRAPPIST Gillon et al. 2016). The population of known planets with these characteristics is significantly smaller than that of hot Jupiters, limiting the sample set to a pair of objects discovered by MEarth: GJ1132b (1.3R_Earth, M4 host star) (Berta-Thompson et al. 2015) and GJ1214b (2.7R_Earth, M5 host star) (Charbonneau et al. 2009). Transit detection for GJ1132b proved unachievable with the FTN at this point, given observed transit depths of less than 5 mmag (Berta-Thompson et al. 2015). However, Figure 12 depicts the successful recovery of a 5σ transit signal for GJ1214b collected by OJC-Falcon. Predicted transit values are duration = 52.1 minutes, depth = 14.9 mmag Berta et al. (2011); transit values recovered from the OJC-Falcon data are duration = 55 minutes, depth = 23.1 mmag. The missing data and fit line before and during ingress, as well as post-transit, were due to passing clouds (and corresponding low SNR). The rms of the residuals for this fit is 3.75 mmag.

Zoom In Zoom Out Reset image size

Additional data was gathered from multiple FTN sites on a variety of different UCDs (not known to host exoplanets) to assess photometric precision and light curve stability over several hours of continuous observation. Using these results alongside theoretical analyses based on known system specifications, design parameters were developed for a hypothetical stand-alone UCD transit monitoring campaign. With optimistic assumptions regarding weather and dedicated telescope time, this study predicted that a campaign targeting M5-type UCDs with the full 12-node network would yield its first discovery of a new 1–2R_Earth exoplanet within approximately 80 nights of observing. Less optimistic assumptions extend that timeline by a factor of two or three. USAFA faculty and cadets researchers continue to refine both observational and analytical techniques in preparation for making substantive contributions to the exoplanet community as the FTN proceeds toward full operational capability.

UNSW Canberra Space, an Australian leader in Space Research has a wide portfolio of space missions and space research and development activities. The UNSW Canberra FTN node is central to the group's research in Space Situational Awareness. In 2017 November, The UNSW Canberra Space group and Defence Science and Technology Group (DSTG) launched the jointly owned and operated BRMM (Buccaneer Risk Mitigation Mission) CubeSat. The growth of this group and Australian space activities in general has been influenced by participation in the FTN. Two more small satellites missions are being designed and built by UNSW Canberra Space and scheduled to be launched during 2018 and 2019 sponsored by the Royal Australian Air Force (RAAF).

The Space Surveillance team of UNSW Canberra Space have been performing systematic and exhaustive observations of Buccaneer and other CubeSats (Figure 13). Observing CubeSats in Low Earth Orbit (LEO) is a challenging task due to their very small size and the strong uncertainties in their orbital information. They are affected considerably by space weather events and atmosphere density changes. As part of an ongoing research effort, UNSW Canberra Space has been able to measure attitude slew on specific CubeSats in the context of building up a photometry database on over 75 CubeSats. A particularly important example of this is the use of the CBR FTN to make first optical observation of the 2017 February launch of 104 satellites out of India into low earth orbit.

Zoom In Zoom Out Reset image size

Similarly observing from geographical location different to the other nodes, we are able to analyse satellites in GEO of the likes of AMC-9 and Telkom-1 which are non-consistent in their position, and hence visible at times from longitudes other than their apportioned slot in the GEO belt. Such opportunity sponsors international collaboration in the surveillance of space.

In order to combine the technology that we have available in the global FALCON network with the need to bring knowledge to the communities housing the different network telescopes, the FTN STEM First Light Project was designed. The First Light Project provides an opportunity for students, as part of a class or other educational group, to propose which objects will be the first outreach observations of an FTN node. It has been held for six of the eight currently operational sites, with the plan to conduct such an event for every FALCON telescope. A representative of each node works with the USAFA FTN outreach lead to identify teachers, schools, or organizations in their local area that focus on STEM related subject matter and invite them to participate. The project allows students to walk through the process of determining which astronomical objects would be interesting to observe and developing a technically sound proposal.

To provide one location where teachers can register for the event, acquire all of the needed event information, and submit their students' proposals, an online event software is used. The prospective teacher participants are provided with a link to the webpage where they can find all of the project details and guidance for submitting a proposal. For example, a document titled "How to Begin Your Proposal" helps teachers answer a series of standard questions such as is your target (or targets) visible at all from the designated FTN site? What time of year and what time of night is best for observing your target (or targets)? Is your target (or targets) well matched to the capabilities of the telescope and camera system? Also provided are a proposal template and a list of astronomy-related resources.

Once the FTN team receives the proposals, pre and postgraduate students along with FTN scientists evaluate the proposals and determine the feasibility of conducting them based on the characteristics of each site. After determining which proposed objects can and will be observed, FTN team members contact the students and teachers to discuss their proposals, provide interesting facts about their proposed objects, talk about the FTN, and answer their questions. To date, we have completed First Light Projects for six of the eight installed telescopes resulting in the participation statistics provided in Tables 2 and 3.

The proposed first-light objects fell within a variety of different categories, the most popular being solar system objects, as is seen in Table 4. Several of the proposals, especially during the projects using the Vicuña, Chile (MMO), Canberra, Australia (CBR), and Durango, CO (FLC) sites, indicated that in depth analysis would be completed upon receiving the data. Included in those studies were an exoplanet transit (WASP-43), orbital mechanics and laws of physics investigations based on multiple observations of the Saturnian and Jovian moon systems, open versus globular cluster studies, a comparison of stellar spectra, and galactic classification using the Hubble Classification Scheme. In addition, the international aspect of the projects allowed students to observe portions of the sky that may not typically be visible to them, for example, students in the US were able to observe Southern Hemisphere objects.

Examples of the resultant data images from the MMO and CBR first-light proposals that we provided back to the students can be seen in Figures 14–17. As we proceeded through all the sites' STEM First Light Projects and improved our capabilities, the provided data improved in quality and began including color-stacked images. Examples from the three prior projects can be found in a previous paper (Gresham et al. 2016). The data files sent to participating teachers are accompanied by an explanation of the filters used, viewing conditions, and any other information we feel the teacher could use to expand the students' understanding. We also offer to hold a teleconference with the classes to provide verbal feedback and to answer any questions they might have in regard to the images they received.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

To conclude each event, we ask the teacher participants to complete a voluntary survey on the educational effects and project design. Overall, the results have indicated an improvement in student interest and understanding while also meeting the teachers' educational needs. To summarize the educational results tallied for all the sites' First Light Projects,

• 1.  
  Over half of the responding teachers stated that the project helped them meet state or national education standards with about one third indicating they did not know.
• 2.  
  On a 1–5 scale (strongly disagree to strongly agree), the average score was a 4.29, between Agree and Strongly Agree, when asked if the project improved the students' interest in science, astronomy, or other STEM subjects.
• 3.  
  On a 1–5 scale (strongly disagree to strongly agree), the average score was a 4.29, between agree and strongly agree, when asked if the project met the teachers' expectations with respect to educating their students.

Previous project feedback pertaining to project design and the logistics of obtaining supplemental information and submitting proposals has been mixed. However, over time it has become more positive. We have improved user experience continually based on feedback from previous events. (Gresham et al. 2016).