Combating Threats Exchange (CTX) - Special Issue: Unidentified Anomalous Phenomena: Science and Analysis

Summary

This special issue of Combating Threats Exchange focuses on integrating scientific methodologies, sensor technologies, and standardized analytic frameworks into the study of Unidentified Anomalous Phenomena (UAP). It features insights from leadership at the All-domain Anomaly Resolution Office (AARO), research on event-based sensing, Ukrainian military observations of aerospace phenomena during wartime, and practical guidance for scientific-grade optical data collection.

Cover

CTX SPECIAL ISSUE Unidentified Anomalous Phenomena: Science and Analysis ISSN 3070-4529 (online) SPRING 2026

Editorial Information & Letter from the Guest Editors

EDITORIAL STAFF KATHLEEN S. BAILEY, Executive Editor AMINA KATOR-MUBAREZ, Editor MICHAEL THOMAS, Editor AMELIA WELD, Visual Content Curator

EDITORIAL REVIEW BOARD VICTOR ASAL, University at Albany, State University of New York ANDREW GARFIELD, Joint Special Operations University DEBORAH GIBBONS, US Naval Postgraduate School TIMOTHY JONES, Duke University CRISTIANA MATEI, US Naval Postgraduate School BRANDON NAYLOR, US Naval Postgraduate School MICHAEL OWEN, CAPT, US Navy, US Naval Postgraduate School IAN C. RICE, COL, US Army (Ret.), US Naval Postgraduate School HUGH SUTHERLAND, Joint Special Operations University SHYAM TEKWANI, Asia-Pacific Center for Security Studies CRAIG WHITESIDE, US Naval War College

Layout and design provided by US Naval Postgraduate School

Letter from the Guest Editors Unidentified Anomalous Phenomena (UAP) is the modern term used for what was traditionally known as Unidentified Flying Objects (UFOs). The term UAP first appeared in the late 1960s and, in recent decades, has gradually replaced UFO in US official and scientific contexts. This shift reflects an effort to move away from the culturally loaded, stigmatized, and often limiting connotations of UFO toward a more inclusive and precise definition. Unlike the earlier term, UAP encompasses reported anomalous detections across multiple domains, rather than being confined to aerial observations alone.

For decades, both trained observers and members of the public have reported objects exhibiting unusual and often unexplained characteristics. These observations have captured widespread attention. For some, national security concerns, safety, and scientific inquiry drive interest; for others, fascination centers on the possibility of unknown technologies or even extraterrestrial origins, reflecting a more speculative and curious approach to what may exist beyond our current understanding. Regardless of motivation, the undeniable reality is that UAP represent a phenomenon of growing significance and interest, making it essential to approach the topic with rigor, without stigma, and guided by evidence-based inquiry.

Growing institutional legitimacy has repositioned UAP from a marginal topic to a subject of serious inquiry. Organizations such as the US Department of Defense, the US Department of Homeland Security, NASA, and leading academic institutions (both within the United States and equivalent organizations internationally) have dedicated significant resources and personnel to studying these phenomena and addressing critical knowledge and analysis gaps. Sustained public interest and congressional hearings have reinforced transparency and accountability, helping normalize the study of UAP as a credible area for scientific and policy inquiry and encouraging more rigorous, open reporting and investigation. Recognizing the importance of this complex subject area, the US Naval Postgraduate School (NPS), in close partnership with the All-domain Anomaly Resolution Office (AARO), presents this inaugural special issue, which centers on integrating scientific methodologies into the study of UAP. Its goal is to highlight the importance of UAP through a focused, systematic, evidence-based approach; encourage stigma-free reporting and dialogue; examine current gaps and challenges; and promote collaborative strategies for the United States, its allies, and partner nations to advance rigorous study and best practices in addressing these phenomena.

Letter from the Guest Editors (Continued)

The issue opens with a Foreword by Dr. Michael Hesse, who offers valuable insights into the challenges of UAP, the NPS-AARO umbrella initiative, and effective strategies to address them.

In our CTX Interview, AARO Director Dr. Jon Kosloski provides a firsthand perspective on the organization’s mission, priorities, and strategic direction in advancing UAP-related efforts.

Dr. Randy Bostick discusses the importance of data vetting and accessibility in the study of UAP, emphasizing the need for high-quality datasets, rigorous standards, and well-validated methodologies.

Next, Kaylin Hagopian and Dr. Joshua Shank examine event-based sensing as an emerging technology for detecting, tracking, and characterizing UAP, highlighting its potential as a promising addition to the monitoring toolkit near military ranges, sensitive facilities, high-traffic air spaces, and other areas of critical infrastructure.

In the following essay, Drs. Artem Bilyk and Kyrylo Nikolaiev share lessons learned and best practices for addressing UAP-related challenges during the ongoing Russo-Ukrainian War. They underscore that studying UAP in the modern era is not only a matter of national security and defense but also a complex challenge for contemporary science.

Finally, J. Kevin Ryan provides a practical guide for everyday observers on capturing UAP events at night, emphasizing the critical role of citizen scientists in reporting and documenting them.

Our next call for papers is right around the corner. If you are interested in UAP-related topics and have valuable insights to share, we encourage you to visit the website for more information: https://nps.edu/web/ccht/forthcoming-call-for-papers

We express our sincere appreciation to all special issue contributors, including the editorial board, the CTX team, and our dedicated readers.

Tahmina Karimova and Lawrence Walzer Center on Combating Hybrid Threats Energy Academic Group US Naval Postgraduate School

NOTES 1 Merriam-Webster, “The Words of the Week (Feb. 10),” Merriam-Webster Wordplay, accessed March 24, 2026, https://www.merriam-webster.com/wordplay/the-words-of-the-week-feb-10

The Center on Combating Hybrid Threats coordinates and conducts interdisciplinary research, education programs, and outreach in order to enhance the Naval Postgraduate School students and partners with the strategic, operational, and technological means necessary to detect, deny, disrupt, degrade, defeat and ultimately deter hybrid threats. Learn more at nps.edu/ccht

Table of Contents (Inside)

Inside 01 Letter from the Guest Editors - Tahmina Karimova and Lawrence Walzer, Director, Center on Combating Hybrid Threats, US Naval Postgraduate School 06 Foreword - Dr. Michael Hesse, Vice Provost for Research and Innovation, US Naval Postgraduate School 08 The CTX Interview: Dr. Jon T. Kosloski - Interviewed by Lawrence Walzer, Director, Center on Combating Hybrid Threats, US Naval Postgraduate School 13 Data Vetting and Accessibility for the Study of UAP - Dr. Randy Bostick, Former Science Advisor, All-domain Anomaly Resolution Office 21 Event-Based Sensing for Unidentified Anomalous Phenomena Detection, Tracking, and Characterization - Kaylin Hagopian and Dr. Joshua Shank, Sandia National Laboratories 32 Ukrainian Military Observations and Analysis of Unidentified Aerial Phenomena during Defense against the Russian Federation - Drs. Artem Bilyk and Kyrylo Nikolaiev, Defense Intelligence Research Institute, Kyiv, Ukraine 42 Capturing Unidentified Anomalous Phenomena Events: A Practical Photography Guide for Everyday Observers - J. Kevin Ryan, Special Agent, US Air Force (Ret.) 59 Call for Submissions

About the Contributors

About the Contributors

Dr. Artem Bilyk is a Ukrainian scientist, engineer, and defense specialist. He is an associate professor at the Defense Intelligence Research Institute in Kyiv, Ukraine. Dr. Bilyk holds a PhD in engineering (Candidate of Technical Sciences) from Kyiv National University of Construction and Architecture.

Dr. Randy Bostick served for more than 20 years in the US Air Force, specializing in image processing and remote sensing, developing advanced methods for extracting actionable intelligence from Department of Defense and Intelligence Community sensor systems. As Science Advisor to the All-domain Anomaly Resolution Office, Dr. Bostick helped advance the US government’s approach to Unidentified Anomalous Phenomena studies. His work emphasized the application of rigorous analytic techniques, the effective use of existing government-owned sensor systems, and the design of dedicated sensors and collection strategies to improve data quality and scientific rigor. Now, as an independent consultant, Dr. Bostick remains committed to strengthening scientific methodologies and fostering collaboration across civilian, academic, industry, and government communities. He received his PhD in physics, with a concentration in electro-optical engineering, from the Air Force Institute of Technology.

Kaylin Hagopian works at Sandia National Laboratories and served as project lead for an effort that developed low-power event-based sensing algorithms for wide-area search-relevant applications. She is the primary author of the Wide Area Tracking and Characterization via Event-based sensing Reconnaissance pipeline and currently supports various efforts at Sandia that advance event-based sensing/neuromorphic capabilities, largely for nuclear nonproliferation applications, along with efforts that advance confidence calibration for automatic target recognition. She holds an MS in computer science from the Georgia Institute of Technology.

Dr. Michael Hesse serves as Vice Provost for Research and Innovation at the US Naval Postgraduate School (NPS), where he provides strategic leadership for the university’s research enterprise. In this role, he oversees the development and execution of research priorities that advance the US Navy, US Marine Corps, and Department of Defense. Prior to coming to NPS, Dr. Hesse was the Director of the Science Directorate at NASA’s Ames Research Center, a position in the Senior Executive Service. Before joining Ames, Dr. Hesse spent three years at the University of Bergen in Norway, where he held a professorship in physics. Dr. Hesse received his diploma and doctoral degree in theoretical physics from the Ruhr-Universität in Bochum, Germany.

Tahmina Karimova joined the US Naval Postgraduate School (NPS) in 2010. She serves as a faculty associate with the Energy Academic Group and as the deputy director for the Center on Combating Hybrid Threats. She holds a master’s degree in public administration from the Middlebury Institute of International Studies and an MA in Security Studies from NPS.

Dr. Jon T. Kosloski serves as the Director of the All-domain Anomaly Resolution Office. Prior to that, Dr. Kosloski held technical and leadership positions within the Research Directorate of the National Security Agency. In that capacity, he led advanced mission-oriented research in the fields of networking and computing and served as a subject matter expert in Free Space Optics, advising various Department of Defense (DoD) agencies. In addition to his optics and crypto-mathematics research, Dr. Kosloski invented an advanced language-agnostic search engine and served at the DoD Special Communications Enterprise Office. He received his PhD in electrical engineering from Johns Hopkins University.

Dr. Kyrylo Nikolaiev is an associate professor at the Defense Intelligence Research Institute in Kyiv, Ukraine. His research focuses on ecological security, sustainable development, and advanced monitoring technologies for studying Unidentified Anomalous Phenomena. Dr. Nikolaiev holds a Doctor of Sciences in public administration (state security) from Ukrainian Interregional Academy of Personnel Management.

J. Kevin Ryan is a retired federal special agent with more than 37 years of combined military and civilian service with the US Air Force Office of Special Investigations (OSI). Special agent Ryan served in more than 15 assignments with OSI including as Director of Counterintelligence for the Department of the US Air Force and US Space Force and concluded his federal career as Chief of Operations for the All-domain Anomaly Resolution Office. A lifelong amateur photographer, he brings both investigative rigor and a passion for the night sky to the study of unidentified anomalous phenomena. Mr. Ryan received an MA in criminal justice from Oklahoma State University and an MA in national security affairs, with a concentration in Western European studies, from the US Naval Postgraduate School.

Dr. Joshua Shank leads the event-based sensing working group at Sandia National Laboratories and leads several projects related to nuclear nonproliferation remote sensing, ranging across novel semiconductor devices, assembly and characterization of remote sensing systems, and algorithm development for data exploitation. Dr. Shank holds a PhD in electrical engineering from the Georgia Institute of Technology.

Lawrence Walzer is a retired US Marine Corps officer and a graduate of the US Naval Postgraduate School (NPS), where he earned a master’s degree in national security affairs. He has been a faculty member at NPS for eight years and currently serves as the Director of the Center on Combating Hybrid Threats.

COVER IMAGE The night sky seen during Exercise Black Jack 25 off the coast of Queensland, Australia, 17 July 2025. Black Jack is a biannual reconnaissance exchange that strengthens relationships and enhances the capabilities and interoperability between the US and Australian reconnaissance forces. (US Marine Corps photo by Lance Cpl. Sawyer Carleton)

DISCLAIMER This journal is not an official Department of Defense publication. The views expressed or implied within are those of the contributors and do not necessarily reflect the views of any governmental or non-governmental organization or agency of the United States of America or any other country.

TERMS OF COPYRIGHT Copyright © 2026. The copyright of all articles published in Combating Threats Exchange (CTX) rests with the author(s), unless otherwise noted. CTX is a peer-reviewed, biannual journal available free of charge to individuals and institutions. Copies of this journal and the articles contained herein may be printed or downloaded and redistributed for personal, research, or educational purposes free of charge and without permission, unless otherwise noted. Any commercial use of CTX or the articles published herein is expressly prohibited without the written consent of the copyright holder.

Foreword

Foreword By Dr. Michael Hesse, Vice Provost for Research and Innovation, US Naval Postgraduate School

Unidentified Anomalous Phenomena (UAP) represent a real and impactful domain-awareness challenge. They sit at the intersection of operational safety, emerging technology assessment, intelligence analysis, and scientific inquiry. Observations span air, maritime, space, and other operational environments. Some can be resolved through conventional explanations. Others remain unresolved—not because they defy physics, but because the information is incomplete, ambiguous, or insufficiently instrumented.

Reducing uncertainty in this domain requires a systematic approach. It requires calibrated sensors, standardized data architectures, rigorous analytic processes, and a culture that prioritizes evidence. Above all, it requires the systematic application of proper scientific methodology.

Recent years have seen significant progress in reporting structures and institutional coordination. Yet, the following persistent gaps remain: inconsistent metadata standards, limited sensor fidelity, uneven analytic frameworks, and cultural hesitancy in reporting. These are solvable problems. They demand dedicated investment in sensing technologies, cross-domain data fusion, reproducible analysis pipelines, and related research grounded in physics, engineering, statistics, and operational analysis.

This special issue of Combating Threats Exchange (CTX) is dedicated to strengthening that foundation. The objective is straightforward: bring scientific rigor to a problem set that has too often been characterized by fragmentation or speculation. Scientific inquiry—falsifiable hypotheses, calibrated measurement, uncertainty quantification, reproducibility—is the essential tool for further progress.

At the US Naval Postgraduate School (NPS), through the Center on Combating Hybrid Threats and in close partnership with the All-domain Anomaly Resolution Office, we are building an interdisciplinary framework to integrate operational data with scientific analysis. This effort includes collaborative research agreements, advanced modeling and sensing studies, classified analytic work where required, targeted experimentation, specialized publication, and tailored academic offerings. We are also expanding communities of interest across service components, fleet commands, allied institutions, and research partners.

NPS is uniquely positioned to contribute. As the Department of Defense’s graduate education and applied research institution, we operate at the nexus of theory and operational practice. Our faculty and students bring expertise in plasma physics, signal processing, aerospace engineering, data science, human systems integration, intelligence analysis, and policy. This cross-disciplinary environment is precisely what a multi-domain problem requires.

For the US Navy in particular, persistent global presence across all domains makes domain awareness essential. Unresolved anomalies—if not properly characterized—can obscure sensor limitations, mask emerging technologies, or introduce operational risk. While most cases are likely attributable to conventional sources—natural phenomena, sensor artifacts, commercial systems, or foreign technologies—we cannot assume adequacy of explanation without rigorous analysis. Strategic surprise often exploits ambiguity. Only a systematic approach can reduce it.

Equally important is the human dimension. Although progress has been made in normalizing UAP reporting, cultural reticence still exists. High-quality data begin with professional, stigma-free reporting channels supported by sound analytic feedback loops. Organizational behavior, cognitive bias, and decision science therefore matter as much as hardware and algorithms.

This is not solely a government challenge. Observations may occur near critical infrastructure, maritime corridors, industrial sites, or populated areas. A credible framework requires collaboration across governmental agencies, academia, industry, and allied partners. Shared data standards, interoperable metadata architectures, joint analytic methodologies, and coordinated research efforts will accelerate learning and strengthen attribution capabilities.

From my perspective as a physicist and former NASA research leader, the way forward is clear. Complex phenomena demand measurement. Measurement demands instrumentation. Instrumentation demands calibration. And analysis demands rigor. We must integrate operational awareness with the scientific method, close data gaps, quantify uncertainty, and progressively constrain the space of plausible explanations.

UAP-related challenges are global. Our allies face similar observational ambiguities. Strengthened international cooperation—focused on shared sensing strategies, analytic standards, and coordinated research—will enhance collective domain awareness and strategic stability.

This CTX special issue reflects a commitment to move the conversation from conjecture to disciplined inquiry. By embedding scientific methodology within operational frameworks, we strengthen safety, enhance attribution, and reinforce national and allied security in an increasingly complex technological environment.

ABOUT THE AUTHOR Dr. Michael Hesse is Vice Provost for Research and Innovation at the US Naval Postgraduate School. This is a work of the US federal government and is not subject to copyright protection in the United States. Foreign copyrights may apply.

The CTX Interview: Dr. Jon T. Kosloski

THE CTX INTERVIEW Dr. Jon T. Kosloski, Director, All-domain Anomaly Resolution Office (AARO) Interviewed by Lawrence Walzer, Director, Center on Combating Hybrid Threats, US Naval Postgraduate School

On 4 December 2025, Lawrence Walzer facilitated an interview with Dr. Jon T. Kosloski. The interview questions were tailored to address key issues surrounding Unidentified Anomalous Phenomena (UAP) for this special journal issue, offering valuable insights into AARO’s mission, priorities, and challenges. The discussion provided an important opportunity to hear directly from AARO leadership and explore ways to strengthen interagency collaboration and advance efforts related to UAP problem sets.

Walzer: To begin, could you explain the primary mission of AARO and how it fits within the broader national security and defense framework?

Kosloski: Certainly. AARO’s mission is to synchronize efforts across the Department of Defense and other federal agencies to better detect, identify, and attribute objects of interest, with a focus on national security sites and critical infrastructure. Unknown objects in any domain—whether in the air, sea, or space—pose potential threats to national security and warrant the attention of the US government. AARO is the first whole-of-government effort to advance the idea of “multi-domain awareness” for anomaly resolution as a core security objective. As we have all seen from conflicts around the globe, warfare is changing rapidly. We help the Department stay ahead of developments to ensure our warfighters have the knowledge and tools to mitigate risks and maintain our technological edge in any environment or operational context.

Walzer: That’s remarkable. What really stands out to me in AARO’s mission is the emphasis on whole-of-government efforts and the pursuit of multi-domain awareness. Those approaches highlight that your work spans all domains—such as land, sea, air, and space—while applying scientific methods to understand potential national security issues. I also appreciate how central collaboration is to your mission, as well as your commitment to reporting findings to Congress and the public.

Considering your multi-domain work, I am wondering what are the biggest challenges your office faces when investigating UAP?

Kosloski: Far and away, AARO’s biggest challenge is a lack of high-quality data. US military sensors, both the technology and the human operators, are second-to-none at doing their specific mission, but they aren’t general purpose scientific instruments. So, when one of these highly specialized sensors detects something potentially anomalous, the data captured is often not well-suited for rigorous scientific investigations.

This mismatch can lead to incorrect initial assessments of an object’s size, speed, or appearance. AARO sees this play out in its findings: the overwhelming majority of UAP reports we’ve resolved turn out to be misattributions of ordinary, prosaic (or everyday) objects. Importantly, we do not have a conclusion in mind before we start an investigation. AARO will follow the data wherever it leads, but we need good data to follow.

Walzer: How does AARO handle public interest and skepticism/stigma around UAP reports?

Kosloski: This is a crucial point. The US government takes UAP seriously, and there should be no stigma associated with making a UAP report. Detailed reports of UAP observations are essential to AARO’s work, and withholding a report could potentially endanger US national security. When a UAP is reported, our team seeks detailed information about its physical characteristics, performance, and behavior. We are also interested in information about the observer and the environment to support a full analysis and, when possible, resolution. AARO focuses its efforts on the national security concerns associated with domain awareness gaps; the stigmatization of UAP reporting actively hinders our mission. We appreciate public interest in UAP and examine each case with an open mind. We always start from a position of curiosity and, from there, we simply follow the data.

Walzer: Are there specific innovations or tools that you believe will significantly enhance AARO’s capabilities?

Kosloski: Above all, we need more high-quality sensor data, and we need better-trained observers. We need the ability to access the many existing sensor systems operated by the US government (USG) while building new, bespoke sensors as needed. As for better-trained observers, we find that a big part of our work focuses on education and awareness. AARO publishes educational materials to help familiarize USG personnel and the public with the common sources of UAP misattributions—things like parallax, satellite flaring, or even in-camera artifacts. Our objective is to improve the quality of the incoming reports so we can focus our resources on the UAP events that may demonstrate truly anomalous characteristics.

AARO is also looking into bringing new analytic capabilities online using artificial intelligence-enabled algorithms to filter vast data sets for potential anomalies.

Walzer: What is your approach to interagency information sharing and fostering cooperation internationally?

Kosloski: Interagency cooperation is essential to our work. AARO works closely with partners across the federal government, who bring their own unique expertise to bear in the effort to better understand and respond to UAP sightings. We have also participated in several international engagements with US security partners to share insights and best practices for UAP analysis. Meanwhile, we work with groups outside of government, especially in the scientific and academic communities, which we rely on to provide feedback on our assessments, conduct scientific research, and ensure our approach remains grounded in a peer-reviewed process. We try to share as much information as possible at the unclassified level, recognizing that significant expertise resides outside of government.

Walzer: I could second that. Recognizing how essential interagency cooperation is to your work, our center is eager to join forces with NPS faculty and students (as well as international partners) and help drive collaboration across multiple lines of effort—academic research, education, experimentation, peer-reviewed publications, CRADAs [Cooperative Research and Development Agreements], and more.

Beyond partnerships with academia, interagency teams, and non-governmental organizations, how can the Services—particularly the US Navy—better support your mission?

Kosloski: The best way the Services can support AARO is by continuing to promptly report UAP incidents and retain any technical data in accordance with the February 2025 Joint Staff GENADMIN [General Administration] guidance on UAP reporting and materiel disposition. That guidance lays out the best way to preserve and convey all the important information AARO needs to begin an investigation. There should be no stigma associated with reporting UAP; reporting is in the national security interest.

Walzer: What are the priority research focus areas that may best lead to increased capability and capacity to detect, identify, and attribute UAP?

Kosloski: This is an area where our scientific and academic partners are key. AARO’s top priority is facilitating the integration of the nation’s diverse, multi-domain sensing capabilities into a single, unified architecture. We need the ability to access all the existing civilian and military systems we have, making sure all of our systems can talk to each other.

We also need to develop novel detection and filtering algorithms. We are dealing with vast data sets, so we need to find better ways to discriminate potential anomalies from all that noise, particularly when we’re dealing with multi-sensor fusion and differing data types. This is important because sensors of different kinds may have observed the same incident from a different perspective, and integrating those perspectives could help AARO in its analysis.

Walzer: What are your long-term goals for the office, and how do you measure success?

Kosloski: Our long-term goal is to ensure that the Department of Defense, the intelligence community, and all the relevant civilian agencies are equipped to efficiently detect, track, analyze, and manage anomalous phenomena. We want them to be using standardized tools and tradecraft, so everyone has a shared awareness using shared language.

Measuring success can be challenging. This isn’t a project with a finish line; it’s an ongoing mission. However, AARO has already shown success in applying industry-standard intelligence and scientific techniques to reduce the amount of time between receiving a report and rendering a finalized assessment. That improvement in efficiency shows that our value to national security only continues to grow as we hone our tradecraft. Each time we enable a partner to make sense of an observation or improve their domain awareness, that keeps them better informed and safer. We consider that a success.

Walzer: Of course, I have to ask you this next question: in all of AARO’s investigations to date, have you found any evidence, data, or credible information that would suggest any UAP are extraterrestrial in origin?

Kosloski: That’s the most common question I get, and I am glad to answer it directly. To date, AARO has found no verifiable evidence suggesting any UAP incident we have investigated is extraterrestrial in origin or represents extraterrestrial technology. AARO will not assert a conclusion without evidence to support it, and we will not speculate about a phenomenon’s nature or origin based on inconclusive data. As the saying goes, “Extraordinary claims require extraordinary evidence.” From what I’ve seen as director, the data we’ve analyzed to date simply does not point to that conclusion. We do have many unresolved UAP cases, and we will continue to follow the data wherever it leads.

Walzer: You’ve stated most resolved cases are “prosaic objects.” Of the unresolved cases, what is your working assessment? Are we primarily talking about sensor anomalies, or is there a genuine concern that these represent advanced and unknown technological capabilities from adversaries?

Kosloski: This is precisely the question AARO was established to answer. Many of our cases result from sensor anomalies and other mundane explanations, but there is also the potential for strategic surprise from an adversary, perhaps using a new technology we are not familiar with, or an old technology used in a new way. We simply cannot have “unknowns” operating in our airspace, or any domain for that matter.

Walzer: Given the sensitivity of your work, I understand that certain data must be handled within specific classification boundaries. How do you balance public interest in UAP with the need to protect classified information? Can you explain what prevents AARO from releasing clearer data or video, and how you are working to declassify information so the public and scientific community can analyze it?

Kosloski: AARO is constantly balancing its commitment to transparency with the need to protect national security information. We understand that public interest in UAP is high. We launched our public website, www.aaro.mil, to share UAP imagery, trends, reports, and educational material at the unclassified level.

However, much of AARO’s data comes from highly sensitive military and intelligence systems. Importantly, AARO doesn’t directly own these systems, so we work closely with the reporting service to declassify as much as we can while protecting our national security. Releasing unredacted video or military platform data could inadvertently reveal sensitive US information. That’s a compromise we simply cannot make. It would be handing a playbook to our adversaries. So, our job is to be as transparent as possible by sharing our conclusions and declassifying imagery only when we are certain it won’t damage national security.

When we release data or video, we do so at the media’s operational resolution without adulteration, with the only modifications being precise redactions to protect military or intelligence capabilities and vulnerabilities. Many of the systems from which AARO obtains data are not designed to produce high-resolution imagery. The videos that we post on our website are, unfortunately, as good as these reports get. It’s part of the reason we continually emphasize the need for more and higher-quality data associated with UAP reports.

Walzer: The government has a history of seeking to address this topic, often with inconclusive or dismissive results. What makes AARO fundamentally different from past efforts, and why should the public trust this time will be different—more transparent, definitive?

Kosloski: AARO is the first whole-of-government effort, established in law by Congress, with the authority to synchronize this work across the Department of Defense, the intelligence community, and other federal agencies. We are an integrated part of the national security framework and report directly to the Deputy Secretary of Defense and the Principal Deputy Director of National Intelligence. We are the first UAP effort to have the full suite of investigative resources required to execute this authority, and unfettered access to programs across the federal government, ensuring that we can be confident in our conclusions.

Our methodology also sets us apart. We have built upon our predecessors’ work, and continue to develop a rigorous, data-driven framework with credentialed scientists, investigators, and intelligence analysts. We are systematically applying serious, proven scientific and intelligence tradecraft to a large volume of data to make robustly informed assessments. It is this combination of broad authority, unprecedented access, robust resourcing, and a rigorous analytical process that makes AARO’s efforts different from anything that has come before.

Walzer: Recognizing the importance of minimizing technical and intelligence surprises, I’m curious about surprises at a more personal level. After reviewing all these data—reports from pilots, sensor readings, and videos—what, if anything, has personally surprised you the most about the UAP phenomenon?

Kosloski: I have found it surprising that, in this age of ubiquitous sensor coverage, it is still so difficult to get high-quality, actionable data suitable for resolving, or even just advancing our understanding of some of the more intriguing cases. That said, I have also found that many of these initially baffling reports are fully explainable once you apply a rigorous, scientific process. It is easy to look at a strange video and jump to a conclusion. But time and time again, when our team of analysts and scientists dig in, we find the answer. It has been a powerful reminder of how important it is to stick to the data and not let assumptions get ahead of the evidence, regardless of how compelling a good mystery can be. In spite of all the noise, I always try to stay focused on the cases that may demonstrate true anomalies.

Walzer: Absolutely. The public fascination with UAPs makes it even more important to remain grounded in the data. It’s a reminder that, no matter how compelling or mysterious a phenomenon may seem, rigorous scientific methods and evidence-based reasoning must guide our conclusions. Any final thoughts?

Kosloski: At its core, AARO’s work is about resolving ambiguity in the name of national security. AARO is committed to a rigorous, data-driven approach as it tackles this complex challenge. We follow the evidence wherever it leads, without speculation or bias. We are here to ensure that operators, service members, and decisionmakers have the clearest possible understanding of their strategic and operational environments.

Walzer: I sincerely appreciate your time and the valuable insights you’ve provided. Your mission is incredibly important, and I wish you—and your Office—much success. Thank you!

ABOUT THE INTERVIEWER Lawrence Walzer is Director of the Center on Combating Hybrid Threats at the US Naval Postgraduate School. This is a work of the US federal government and is not subject to copyright protection in the United States. Foreign copyrights may apply.

Data Vetting and Accessibility for the Study of Unidentified Anomalous Phenomena

Data Vetting and Accessibility for the Study of Unidentified Anomalous Phenomena By Dr. Randy Bostick, Former Science Advisor, All-domain Anomaly Resolution Office

In the middle of the twentieth century, the US Air Force established several projects—Project Saucer, Project Sign, Project Blue Book—to investigate the origins and potential threats posed by unidentified flying objects, or UFOs. Most of the evidence available to support investigation came from single-source observers who were not well prepared to report strange or unidentified events. They described these UFOs using familiar references such as “cigars,” “wheels,” and, famously, “saucers.” Except for a few cases, the conclusions they provided were without supporting metrics or reasoning. Although technical collection with advanced (for the time) instruments such as radar began in the 1940s and 1950s, these technologies were not without shortcomings. The initial shortcomings of these new technologies, including system anomalies in the earliest hardware, undoubtedly contributed to detections being interpreted as fantastic and unexplainable objects. The inability to save raw data from these instruments prevented further analysis by systems experts. UFO reporting became a social phenomenon which in turn led to a proliferation of science fiction writing and movies (which fed back into the reporting frenzy). Any real, consequential evidence to support scientific analysis was likely lost in volumes of summary reports.

In the twenty-first century, sensing technologies have become ubiquitous, and their capabilities greatly improved. However, the reporting of UFOs, now referred to as unidentified anomalous phenomena (UAP), has not likewise improved. Despite the quantity and quality of data and the general public’s exposure to and familiarity with new air- and spacecraft, an era of quality reporting and data for scientific analysis has yet to fully emerge. In reality, the nature of the unknown and anomalous has evolved with advancements in technology, as there will always be objects observed that are outside of a particular sensor’s ability to make a quality measurement. Ring doorbells, for example, may detect insects in the near field that appear as far-off blurry points of light moving in strange patterns. The proliferation of sensors and the massive increase in airborne objects provide more ways to observe more phenomena in some peculiar manner. Though the data are there, they are still not always understood, and the quality of the recorded products, along with the rush to analysis and conclusions based on misunderstood data, threatens to become a barrier to scientific analysis and acceptance of UAP research by the broader scientific community.

Nevertheless, scientific study of UAP is in a better position today than it was in the days of Project Blue Book. The shift from a lack of objective sensor data to having too much data is at least a step in the right direction. There is promise in this wealth of sensor data, but a vetting process is needed to separate useful from useless data along with supplemental information on collection and quality. Dedicated collection with calibrated sensors is necessary to advance scientific analysis. The All-domain Anomaly Resolution Office (AARO), as the government lead for UAP reporting and data collection, is in a position to collaborate with the UAP community to ensure quality control and consistency of UAP data products and to promote guidelines for data collection. These are basic functions to enable scientific analysis, results, and conclusions on the nature of UAP. This article proposes a common approach to UAP reporting by offering a definition of analyzable data and associated quality that can support any derived results and conclusions in the study of UAP.

The Definition of Data The need for quality data and consistent definitions of “quality” has been asserted by AARO and widely recognized by others who study UAP. However, less discussion has been afforded to the meaning of “data,” a term that is often abused. While implying numerical or other quantifiable information, “data” in the context of UAP has come to be conflated with any general reporting that may be received, including deductions, explanations, and conclusions made while observing an event, monitoring an instrument, or describing sensor output. Determinations such as “It had a plasma field,” “It was very hot,” or “It exhibited an advanced form of propulsion” are not data—they are interpretations of what one has observed, often with no objective, sensor-collected information. When presented as such, even unintentionally, rejection by the scientific community follows, as well as an overall unwillingness to engage on the topic of UAP. Data must be objectively collected and presented, with no predetermined conclusions stemming from attempts to better understand UAP.

Many features or characteristics of an object can be learned by collection with remote measurement devices. Most of these are calculated or derived through analysis, not directly measured by the instrument. That which is directly measured is what should be the basic definition of “data.”

For example, a simple electro-optical (EO) system such as the WESCAM MX™-20 can measure the following: • Irradiance: the amount of power or energy incident per unit area on a detector or pixel array. • Time: the time at which a measurement of irradiance is made on the array. • Position: the location in the optical system image plane of an irradiance measurement.

A simple radar system can directly measure values such as the following: • Pulse transmission and arrival time: the time that an electronic signal is sent from the transmitter and collected at the receiver after bouncing off an object. • Frequency shift: in a Doppler radar, the shift in the frequency of the transmitted pulse due to reflection from a moving object. • Direction: the angular direction of the object from the radar receiver. • Received power: the strength of the signal received by the receiver.

Absolute position in space, size, speed, heading, pulse delay, and acceleration are characteristics and features that can be calculated from the above direct measurements. With assumptions about the object’s material composition, analysts can estimate characteristics such as temperature, emissivity, reflectivity, and radar cross-section. Calculating most of these derived quantities requires supporting data specific to the detection sensor or instrument, called metadata. Metadata may include sensor settings, platform location and motion, and other ancillary values. Providing metadata with measurements enables others to repeat and independently verify calculated values. These measurements and the resulting data and metadata are specific to the sensor and do not require new definitions for UAP collection. The science behind the data definitions and calculated values is well known and does not need to be reinvented just because the focus of study is UAP.

Sensor type is a key factor in UAP collection and interpretation, as the calculations that can be derived, and the types of errors that can occur, will vary dramatically between, for example, a passive optical sensor and one that is in motion. For example, the public is likely most familiar with the optical sensors in digital or cellphone cameras, and the images these sensors capture are commonly provided as evidence of UAP. However, most calculations or estimations from passive optical sensor measurements require knowledge of the distance to the target. The range to the target is usually impossible to calculate from a static, monocular collection device. Inaccurate range assumptions lead to poor calculations of UAP speed, acceleration, and size. For moving sensors, such as aircraft, further error is introduced by the effect of parallax (measurement inaccuracy introduced by shifting angles or positions).

For those interested in using optical sensors for UAP collection, deploying two (or more) sensors is highly recommended. Range can be estimated from triangulating lines of sight from multiple sensors separated by a baseline distance. Also, employing two sensors reduces the likelihood of reporting sensor anomalies or other near-field objects as UAP. For example, small bugs flying in front of a camera create pinpoints of light that are often mistaken for distant, erratically moving craft. If these lights appear on only one sensor, we can conclude that they are likely caused by bugs. If the lights appear on two sensors, we can conclude that they are worthy of further analysis.

As a final point, many of the videos provided to AARO (the “Go Fast” and “Mt. Etna” cases in particular) or posted on the internet as “data” have undergone image processing. The FLIR (forward-looking infrared) sensors on fighter aircraft, used to create infrared images, and the multi-spectral targeting systems on remotely piloted aircraft used for intelligence, surveillance, and reconnaissance missions, almost always have some contrast enhancement or edge detection applied to benefit the sensor’s primary mission, which is threat detection, not UAP detection. While the data from such systems provide important information, it is nearly impossible to estimate target brightness or even size for smaller objects. Care must be taken to distinguish the properties of the object being collected from the result of image processing.

The Need for Quality Data Any substantive UAP claims must be founded upon vetted, sharable data with a known and described quality. The requirement is not necessarily for high-quality data—only that the quality be known in order to determine the confidence in any results derived from it. The term “vetted” refers to data that has not been intentionally altered or artificially created to support a desired conclusion. “Quality” is usually expressed through figures of merit such as spatial resolution and radiometric accuracy, along with associated systematic errors. However, a quantity of lower-quality collections can yield a greater aggregate quality when there are many collections on the same object or phenomenon—rarely the case with UAP instrumented collections. This article focuses on data from instruments and sensors. However, vetted oral reporting can yield a quantity of data that may reveal some quality of patterns of activity, common descriptions, or other trends. Many documented studies related to Project Blue Book written accounts have been conducted in this way, as well as analyses of more recent reports. The point here is that if only one or very few sensors collect data, then it is important that the data be of high quality. However, if many lower-quality sensors collect data, quality can be made up for with quantity.

AARO has received many videos of purported UAP with descriptors such as “spherical,” “circular,” “triangular,” and “plasmoid” (a ball of light without sharp edges) that were “moving at great speeds.” While these descriptions may not be inaccurate or disprovable, they often cannot be verified given the quality of the source data. The shape may be the result of camera defocus, odd surface reflections, and apparent speeds from misjudging distance or camera motion and other effects not associated with the object of interest.

Consider the pictures in Figure 1 as a simple example. On the left is an image that cannot be immediately identified as a known entity but might be identified as a gelatinous creature of unearthly origin. We must know the quality of the data to understand whether the image correctly represents the subject or is due to an out-of-focus camera, poor optics, or perhaps image processing (in this case, the author has applied a filter to the image, which is shown in original form in the center of Figure 1). The center image is from higher-quality data, which allows for characterization of the entity with high confidence as a male human (not anomalous). This male human may still be unidentified, but not because of poor data. His face and features are clear; the image supports further scientific analysis to determine identity. On the right is a clear picture, but now the human has a third eye. This entity would be considered unidentified and anomalous as the third eye is not due to data issues and likewise would demand further study.

The definition of “sufficient quality” and requirements for sensors dedicated to UAP collection must be discussed and agreed upon by the scientific community and be highly dependent on assumptions about the UAP. However “sufficient quality” is defined, data products with metadata and other sensor phenomenologies must be provided to substantiate analysis. The void of no data or poor data cannot continue to be filled by conjecture, assumption, and imagination. The scientific community will not entertain results or claims with greater assurance than the evidence warrants.

The Influence of Analysis and Results on Data Collection Data from sensor collections provide the basis for analysis that leads to conclusions. Too often, results are presented without supporting data, with poor or undocumented analysis, or with conclusions the data cannot possibly support. The effect of such presentations is twofold: 1) the reporting stigma related to UAP continues due to continued false claims, and 2) the conclusions become more prominent and energizing than the source data such that, over time, “facts” evolve to support conclusions. Separating data from lore becomes an afterthought, or even impossible, after time has passed. This second effect exemplifies the need for more rigorous approaches to UAP study. However, the first effect may be more damaging as it, along with the conflation of the terms UAP and UFO, leads to suppression of reporting and recovery of data from legitimate sources. The term “unidentified anomalous phenomena” was introduced to describe an event that an observer could not readily identify. The intent behind establishing this new term was to allow pilots, military personnel, law enforcement officers, and others in sensitive positions to communicate relevant information prior to analysis in a noncontroversial way to inform objective assessments. The UAP designation should not be considered the resolution or the event itself and should not lead to quick conclusions and dismissal of data, as has happened with UFO reports. All UAP reports with supporting data must be taken seriously; serious analysis will lead to determination.

Finally, UAP reporters (and anyone interested in or pursuing UAP research) should not feel obligated or allow themselves to be coerced into providing an immediate analysis of what they saw or what their instruments detected. For example, the assessment of a hot, fast-moving object should be left to validated analysis and not an initial description. The self-assessment by an observer that something is “weird” or exotic threatens to lead to the UFO supposition and a reluctance to report and the much-discussed reporting stigma. The focus should be on reporting the observable characteristics that led the observer to deem the object as hot or fast moving and on providing any associated oral, written, or instrumented data. This information provides the basis for scientific investigation; the reporter should be required only to report, not to provide the assessment.

An approach to reducing the stigma of reporting is to use the term UAP in its proper context as an object that is literally unidentified and/or is behaving anomalously with no assumption of origin based on the initial sighting. To alleviate the implied association of UAPs with UFOs, the initial UAP report may be designated a “potential UAP,” indicating that data and information have been provided, but that further analysis is needed before concluding that it is truly unidentifiable or anomalous and why. This suggestion is analogous to a citizen reporting suspicious activity to law enforcement and letting those professionals investigate whether a crime is actually being committed and by whom. Perhaps the potential UAP ends up being identified as a balloon, drone, or something incredible, but that should not concern the observer making a report.

The Scientific Study of UAP There is a need for leadership in organizing and developing data standards to enable the scientific study of UAP. Well-understood data are the foundation of such an effort, whether that data is unintentionally or intentionally collected. The analysis of the data must be documented and repeatable by independent scientists. Any results and conclusions must be traced back to the data through the analysis and cannot assert anything more.

It is not a question of how to make the study of UAP scientifically valid and acceptable; it is a question of applying scientifically valid and acceptable methodologies to the study of UAP. The approach is the same as that for any other domain awareness problem such as countering unpiloted air vehicles or achieving space domain awareness. It is only after scientific analysis indicates that the collected object is not identifiable or is technologically anomalous that the disciplines diverge. Presenting UAP study as a wholly separate field gives the impression of a presumed or necessary outcome, which precludes scientific legitimacy.

To round out the application of the scientific method, a hypothesis is necessary. Historical investigations such as Project Blue Book and the Condon Report sought to disprove the idea that there is an aspect of the natural world not yet explored by science. The inferred hypothesis, and one commonly followed, is that if the UAP described in a report do exist, they must be of extraterrestrial intelligence and not of this world. This hypothesis, which requires only existence to be “true,” cannot be proven or disproven, as it requires no explanation of origin. Moreover, UAP, by definition, are not specific, definable entities. Rather, the term “UAP” describes phenomena that cannot be immediately explained, although they may eventually be explained—perhaps as birds, balloons, or other prosaic objects. A testable hypothesis is that the phenomena described in reports and collected in sensor data exist, and that they have physical, energetic, or other properties that current technology cannot explain (or the null hypothesis that these objects can be explained by existing technologies—a hypothesis easier to test because it lends itself to requirements for collection). The properties that UAP may display that are not presently explainable must be determined by the scientific community. The critical point is that the properties of UAP are not assumed to exist, but are proven to exist.

The Way Ahead Anecdotal reports and unverified claims are insufficient to meet the rigorous standards of empirical science and lead to skepticism and limited academic engagement. AARO is in position to take the lead in a field of research that struggles to gain acceptance within the broader scientific community. This leadership can be accomplished, as in other scientific, defense, or intelligence communities, by establishing data standards to ensure interoperability, discovery, assessment, and exploitation. Application schemas and metadata should also be standardized and documented.

Likewise, study and analysis should be aligned with the standards common to scientific methods and intelligence reporting. The guidelines for the study of UAP data collected by sensors (analytic approaches to verbal and written witness accounts are a separate area of study) should include the following:

• Objective and independent observation: A phenomenon, which should initially be referred to as a “potential” or “possible” UAP, must be recorded by a sensor-based detection system (e.g., radar, infrared sensors, visual recordings). The data from these recordings should be provided without subjective bias, preprocessing, or sensor error. While it is desirable that sensors be calibrated as well as possible, it is more important that the degree of calibration be known. Any preprocessing sensor error that cannot be avoided must be documented. Ideally, this documentation would include corroborative multi-sensor data from diverse detection methods (e.g., visual, electromagnetic, acoustic), ensuring that the phenomenon is not an artifact of a single type of sensor.

• Documented and transparent analysis: All data analysis methods must be well-documented and made available for scrutiny by the scientific community to ensure transparency and to eliminate errors. The analysis should be repeatable by independent entities.

• Analysis of alternatives: To maintain consideration as a “UAP,” the event must be thoroughly analyzed to exclude all known natural phenomena, human-made objects, and sensor anomalies as potential explanations. This analysis includes levels of confidence in ability to make a conclusion based on error analysis and quality of provided data.

• Conclusions and results: Results must be presented in a fashion that characterizes the observed phenomena in terms of physical, radiometric, or other quantitative means. The conclusions on identification and origin are then put forward to the scientific community to propose the employment of known or advanced technologies given the state of the art, potential function, and intended purpose.

AARO, in collaboration with defense, intelligence, academic, and industry partners, should consider these guidelines and agree upon vetted methodology for UAP study.

ABOUT THE AUTHOR Dr. Randy Bostick is the former Science Advisor for the All-domain Anomaly Resolution Office. This is a work of the US federal government and is not subject to copyright protection in the United States. Foreign copyrights may apply.

Event-Based Sensing for Unidentified Anomalous Phenomena Detection, Tracking, and Characterization

Event-Based Sensing for Unidentified Anomalous Phenomena Detection, Tracking, and Characterization By Kaylin Hagopian and Dr. Joshua Shank, Sandia National Laboratories

The term unidentified anomalous phenomena (UAP) refers to unknown aerial targets present in an airspace of concern. They can span a wide spectrum of target types, including birds, stars, planes, satellites, and drones. UAP have presented an increased concern in the last several years due to their presence near military ranges, sensitive facilities, and other high-traffic air spaces. While some phenomena have been identified as benign objects such as balloons or birds, some remain unidentified and may represent intentionally deployed technology. No matter their origin, these objects pose a threat to air safety, and in the case of intentional technology must be positively identified as such and distinguished from the benign objects. Many UAP are difficult to detect, display a wide range of characteristics, and operate at unpredictable times. Developing a highly capable sensing system to detect, track, and characterize UAP is one of the primary challenges for next-generation airspace monitoring sensors.

Maintaining custody of an airspace to ensure detection of UAP with many possible characteristics operating at unpredictable times requires persistent remote sensing. This persistent sensing layer must cover a large geographic area and be able to detect and track objects with a high probability of accurate identification. These sensors must have the following characteristics: • Fast response times to detect and continuously track the motions of individual objects • Low data bandwidth, producing sparse data that include only signals of interest to decrease processing and decision response times • High dynamic range to enable detection of bright and dark objects in challenging environments that include multiple objects approaching the sun’s position in the field of view and in deep shadows in the same scene • Data capture sufficient for accurate discrimination between real targets, false positives (e.g., birds), and a cluttered background • Low power for situations requiring off-grid operation for extended periods • Small and lightweight design

Conventional sensors, such as high-speed radars or traditional imaging cameras, typically possess a subset of these desirable characteristics. For example, high-speed radars tend to quickly detect and track multiple targets, but they can produce large volumes of data, require considerable power to actively illuminate the airspace, and struggle with cluttered backgrounds. Traditional imaging cameras typically are low power, but they also have slow response times, limited dynamic range, and require complex image processing to mitigate cluttered backgrounds. As such, remote sensing for UAP detection and tracking often requires operational tradeoffs among these desirable characteristics and accompanying drawbacks.

Event-based sensing is a novel technology that achieves all of these features simultaneously. Event-based sensors (EBSs) are becoming recognized as a promising technology for UAP detection and tracking because they are able to provide a persistent wide-area search capability for initial target detection and cue higher-value, nonpersistent remote-sensing assets for characterization and identification.

This article discusses the potential for using EBSs for UAP detection, focusing on the technology’s characteristics relevant to this application space. We then present a series of visual EBS signatures from various target types and discuss potential paths for detection, identification, and characterization of UAP.

Event-based Sensors for UAP Detection EBSs are a novel camera technology whereby each pixel acts as an independent, asynchronous optical change detector. In conventional camera technology, all pixels act in concert to record a scene and output related data; in event-based sensing, only pixels that detect a sufficient change in optical intensity report an “event,” while pixels that do not detect a change remain “silent.” This architecture consumes minimal power and data bandwidth, enabling EBSs to possess all the desirable characteristics listed above for UAP detection: fast response times, low data bandwidth, high dynamic range, sufficient data capture capability, and low power in a small and lightweight design.

Fast Response Times The sparse change detection schema enables EBSs to allocate data bandwidth only to dynamically changing areas of a scene, typically achieving 100-microsecond response times. The asynchronicity of EBS event reporting enables each pixel to respond at the local speed of the scene without the latency introduced by synchronizing all pixels. This architecture allows EBSs to dynamically react at the speed of the scene rather than at an a priori rate, simultaneously detecting and tracking very high- and low-speed objects. In addition, EBS data can be used to support initial object discrimination and characterization.

EBS response speed is fundamentally faster than traditional cameras because its asynchronous operation enables pixels to respond on an “as needed” basis as a scene evolves, whereas traditional cameras must synchronize all pixels to produce dense data frames at set intervals, independent of the scene’s actual dynamics. Additionally, conventional cameras require a priori knowledge of expected targets and dynamics for the configuring of parameters like frame rate, gain, and integration time. For example, to consistently detect a visible 10 Hz signal, the frame rate must be at least 20 Hz. This response often requires either biasing camera systems towards faster frame rates and higher data volumes or sacrificing detection performance by missing higher-speed dynamics.

Figure 1 shows the nearly instantaneous response time of EBSs in an extreme case. The data were recorded during a counter-unmanned aerial systems (CUAS) collect when a lightning strike occurred in the background. Pixels exposed to the lightning’s changing intensity responded so rapidly that they captured the successive stages of the strike: leader formation, primary channel formation, and return stroke. During this dazzling, split-second event, the sensor still detected the target UAS at its own response speed while largely suppressing the background that a traditional camera would transmit.

Low Data Bandwidth As EBSs report only data related to change, they require much smaller data bandwidth than traditional imaging cameras as they do not record redundant information associated with a scene’s static areas. Traditional imaging cameras output full frame information, which is generally greater than 1 million pixels at greater than 10 Hz. This rate produces tens to hundreds of megabytes of data per second, which must be continuously processed or transmitted to a remote location. Much of these data are related to unnecessary background information such as the sky, clouds, and buildings. Sandia National Laboratories has found that, for real-world scenes, EBSs typically produce 1–5 percent of the data volume a traditional framing camera produces.

For example, Figure 2 shows a swarm of Group 2 drones (enclosed in red circles) recorded by an EBS (top) and by a conventional framing camera (bottom). While the conventional camera reports the full scene, including static background features such as empty sky and mountain peaks, EBSs report only the data related to change (black values correspond to “silent” pixels that reported no events). Thus, since EBSs output sparse data compared to the conventional camera’s image frames, the sensor data are naturally more prominent. (Background features such as mountain peaks may still be visible due to air scintillation or camera motion.)

Image prominence to the human eye is not equivalent to detectability by an algorithm. As this example shows, event-based sensors concentrate information around change, providing a higher proportion of information relevant to detection and tracking algorithms than traditional cameras.

High Dynamic Range While the core architecture of EBSs does not intrinsically impact dynamic range (the ratio between the brightest and darkest parts of an image), the major commercial off-the-shelf (COTS) EBSs use log-amplifiers in their pixel circuitry, which yield a dynamic range of greater than 120 decibels (dB). This capability supports operation across a wide range of conditions, from nighttime low-light illumination to direct sunlight in the field of view.

In Figure 3, an event-based sensor was set up with the sun deliberately in the field of view. A ball was then thrown such that its arc approached the sun’s position in the field of view. As Figure 3 shows, the event-based sensor data enabled clear detection and tracking of the ball throughout its trajectory.

Data Capture Capability The high temporal responsivity of EBSs, coupled with their potentially high-definition (720 x 1280) resolution, enables them to capture signature information that supports discrimination among different types of targets. This combination allows EBSs to produce clear target signatures even when targets are moving or changing (e.g., blinking) rapidly—as opposed to conventional framing cameras, where motion blur may prevent the capability to distinguish among different target types.

Low Power While traditional passive imaging cameras are typically not high power, they produce dense image frames that require high-power processing to convert raw data into usable information. Scenes consist largely of redundant background information that must be removed to reveal targets of interest. Image processing of redundant information can be a key reason for high system power usage and reporting latency.

An initial layer of background subtraction is built directly into the pixel architecture of EBSs; pixels that do not detect change do not consume power by reporting static background information. This architecture saves power both through the sensor’s non-response and through the reduced quantity of data that systems must process.

Small and Lightweight Design COTS EBSs are small and lightweight—on the order of tens of millimeters in height, width, and depth, and weighing only tens of grams. Depending on use, the optics attached to the sensor are often a larger factor than the camera itself for determining the size and weight of the overall sensing system.

Processing EBS Data Although using EBSs for UAP applications is a promising and steadily growing field, their novel data format poses a key challenge. EBS output includes the pixel coordinates (x,y) related to physical location in the sensor’s field of view, the microsecond precision time (t) at which a change was observed, and the increase or decrease in polarity (p) associated with the change (brightening or dimming). These data are represented as a sparse list of asynchronous <x,y,t,p> tuples, requiring novel algorithms to perform detection and identification. EBS algorithms are still immature compared to algorithms for conventional framing systems.

One approach to processing EBS data commonly found in the literature involves converting EBS data to more conventional “frames” and leveraging existing framing algorithms on the data. Although this approach can produce results with little development effort, it sacrifices the full advantage of the fine-grained temporal information EBSs provide. As previously mentioned, temporal information enables the production of distinct signatures for different target types, which supports target recognition; this benefit diminishes when producing frames, which collapse the temporal information. Streaming algorithms that can operate on the data in sparse <x,y,t> or <x,y,t,p> event space must be developed.

Although EBS data presented in framed format by accumulating events over a time window is easier for humans to interpret, this representation is misleading and, as noted, diminishes the richness of information the sensors provide. An alternate way to visualize EBS data is through a 3-D scatterplot (spatiotemporal volume), which emphasizes the data’s sparse nature and the true “shape” of the signature, but is less easily interpreted by humans.

Event-based Sensing Signatures A UAP ultimately remains a UAP if there is mystery surrounding its identity. Solving that mystery requires the capability to rapidly and robustly distinguish among different types of known targets based on signature. EBS may afford such determinations. The next several subsections present EBS signatures for various UAP-relevant target types to further illustrate the uniqueness of the data generated and the potential of EBS.

• EBS Signature: Birds Birds tend to exhibit fluid and complex trajectories (especially compared to planes or drones). They can change velocity suddenly—which can lead to sporadic decreases and increases in event rate along their dynamic paths. These decreases and increases produce distinct signatures in event space.

• EBS Signature: Bugs Flying bugs can also have dynamic trajectories—often these are rapid and circular or oscillatory in nature. If the bug is close enough to the sensor, prominent wavelike patterns can be observed in its flight path (potentially associated with wing motion).

• EBS Signature: Commercial Planes Planes flying at night produce a distinct signature, as the Federal Aviation Administration requires commercial aircraft to activate anti-collision lights at night. These lights typically flash in a very periodic on/off illumination pattern and at a frequency of 1-2 Hz—which translates to very clear periodic groups of events. If a beacon’s light is occluded from the EBS, the observed frequency may drop below 1 Hz. A plane’s trajectory tends to be roughly linear and of constant velocity, which produces a signature distinct from those of birds and bugs.

• EBS Signature: Stars EBSs can also detect stars due to air scintillation, which causes apparent twinkling. Stars exhibiting less twinkling (i.e., dimmer stars) tend to produce fewer events than those exhibiting more prominent twinkling (i.e., brighter stars). As stars move slowly with respect to the Earth, the EBS signatures for stars, if observed over long enough durations, capture this motion.

• EBS Signature: Satellites Satellite signatures resemble those of planes and stars because their trajectories are roughly linear and constant in velocity, so they appear as straight lines in EBS data. However, their signatures deviate from those of planes in direction and strobe frequency: satellites tend to move more vertically, whereas planes tend to travel more horizontally, and satellite illumination is relatively consistent or aperiodic, whereas planes exhibit a periodic on/off illumination pattern at night. Satellite and star signatures are often similar, particularly for satellites and dim stars; however, they can be distinguished by their speeds, as satellites are typically recorded as faster than stars (shooting stars may be an exception, but their paths are much shorter than those of satellites).

• EBS Signature: Quadcopter Drones Quadcopter drone signatures are characterized by the drone’s ability to hover and maneuver; the signatures display agile, multidirectional flight. In this instance, the quadcopter starts by hovering, then shifts to the right (with respect to the x axis/sensor) and hovers for many seconds before once again moving to the right. This inconsistent, sporadic hovering and moving makes this signature distinct from those explored above.

Exploiting Event-based Sensing Signatures As the previous subsections show, multiple types of UAP-relevant targets can appear in EBS data, each with distinct signatures. Drones are an interesting target type for detection and characterization because their signatures are multifaceted, with their rotors producing periodic signals and their bodies producing non-periodic linear motion. This section focuses on detecting and characterizing drones using Sandia National Laboratories’ Wide Area Tracking and Characterization via EBS Reconnaissance (WATCHER) Pipeline. We present example algorithm output that demonstrates the ability to use EBS data to simultaneously detect multiple drones in a scene and, separately, characterize a drone’s rotor frequency by leveraging the high temporal responsivity of EBSs.

The WATCHER Pipeline Sandia National Laboratories’ WATCHER pipeline is a general-purpose feature-extraction capability able to detect, track, and characterize linear motion and periodic signals in a scene. Researchers have used this pipeline, in part, to investigate the use of EBSs for a variety of CUAS scenarios, ranging from distant point-like targets to closer, spatially resolved vehicles, using both stationary EBS and mobile EBS.

The WATCHER pipeline includes multiple filtering, clustering, characterization, and tracking algorithms. It is regularly stress-tested with new use cases and datasets, existing algorithms are refined, and capabilities are expanded to address multiple applications. As such, the pipeline is continually being improved. The long-term goal is to convert the entire pipeline to a streaming workflow; currently, some of the algorithms are streaming-capable, while others rely on processing batches of events.

To demonstrate the promise of EBSs for automatic UAP detection, tracking, and characterization, WATCHER was applied to two CUAS scenarios—small drone swarms at a distance, and a single large drone—both detected by a stationary EBS.

Small UAS Targets: Swarms at a Distance In the first case, the pipeline processed a dataset containing a swarm of Group 2 drones flying several kilometers away, such that the UAS targets spanned too few pixels across all devices to resolve features. Figure 12 compares output from an EBS (top) and a conventional framing camera (bottom).

As apparent in Figure 12 (and previous figures), event-based data can be noisy, though they still require significantly less bandwidth than framing cameras. This noise can stem from multiple factors including sensor-specific nuances like noisy pixels, selectable sensor settings, ambient light levels, and optics (lens selection, focus). The first portion of the WATCHER pipeline handles noise by running a series of filtering algorithms that leverage spatial, spatiotemporal, and polarity correlation event information. Next, WATCHER identifies clusters associated with periodic and non-periodic signals in the event data and characterizes them. This process begins with object clusters initially being unidentified and uncharacterized (i.e., they are UAP). An initial round of characterization separates “periodic” from “non-periodic” clusters, and a second round performs finer-level identification. Events within a cluster can be generated by unresolved targets (those without prominent shape or features) or resolved targets (those with observable shape and features). Unresolved targets are distinguished from noise based on their “bulk” signals (e.g., overall motion type, flashing lights). As targets become more resolved due to higher zoom optics or closer proximity to the sensor, “detail” signals (e.g., shape, periodic rotor signals) are used to support more precise identification.

Figure 13 illustrates the clusters identified in this drone swarm: the pipeline detected all drones, with only a few false positives. All clusters were identified as not including periodic signals (i.e., clusters were “non-periodic”) as the drone rotors were too small to resolve. Figure 14 shows velocity estimates for each cluster. These results indicate that the vertical line of drones (right side in both images) were moving rightward and slightly upward at a consistent rate, whereas the diagonal line of drones (left side in both images) were moving at variable speed to merge with the vertical line. In particular, drones in the upper-left portion of the diagonal line were moving to the right more quickly than those in the lower-right portion. Additionally, drones in the upper left of the diagonal line were shifting slightly upward, whereas those at the bottom were moving slightly downward. Taken together, these results suggest that the drones in the diagonal line were maneuvering to perform a “zipper” merge with the drones in the vertical line, which matches the observed behavior.

Large UAS Target: Stationary EBS In the second case, the WATCHER pipeline processed a dataset containing a single drone flying close to the sensor, such that the target’s rotors spanned enough pixels to resolve clearly. The EBS remained stationary for these data. As Figure 16 shows, a well-resolved drone at closer range than in the first CUAS scenario is separable into periodic (rotor) and non-periodic (drone body) components, providing quantitative information for target characterization. Pixels responding to the rotors operate at speeds fast enough to assess both rotor speed and phase, yielding detailed information on the drone’s lift capabilities and design. Rotor speed (frequency) changes as the drone maneuvers, and rotor phase establishes that the drone has two blades per rotor that are counter-rotating (i.e., each rotor spins opposite its neighbors).

The two CUAS cases above demonstrate the promise of EBS for detecting and characterizing multifaceted signatures. Future EBS data processing efforts will expand the algorithmic capabilities available, including robust classifiers that explicitly leverage EBS signature differences to distinguish among target types and convert more UAP into known targets.

Conclusion EBSs are a promising technology for UAP detection, tracking, and characterization. This novel sensing architecture enables faster response times on the order of kHz, with low data bandwidth, high dynamic range, and low power consumption in a physically small (tens of millimeters) and lightweight (tens of grams) design.

Signatures from multiple target types relevant to airspace monitoring appear prominently in event-based sensing data, each with distinct characteristics. These differences may be leveraged to develop new algorithms that rapidly distinguish among UAP types, converting them from unidentified phenomena to identified targets. Sandia National Laboratories’ general-purpose feature extraction WATCHER pipeline can be used for wide-area detection, tracking, and characterization of multiple objects in a single scene. Although WATCHER was not developed specifically for UAP application, we have demonstrated that it can detect drones, estimate vehicle velocities for tracking, and, when rotors are sufficiently resolved and in focus, measure rotor frequency for characterization.

We plan to extend the WATCHER pipeline to increase its utility for UAP applications, for example, by incorporating UAP-relevant classifiers, improving the robustness of existing algorithms to reduce noise and recognize confusers (such as reflections), and adding ego-motion removal (which separates camera motion from target motion). We also plan to assess the impact of sensor focus on pipeline performance and investigate techniques for automatic sensor adjustments, such as autofocus.

Overall, early results demonstrate the strong potential of EBSs for low-power, persistent monitoring systems to maintain custody of an airspace against unpredictable and potentially dangerous UAP, expanding the tools we have available to keep the nation, and its people, safe.

Acknowledgements The authors would like to thank Marlene Dugger for her support in analyzing and generating results illustrating signatures of various target types in EBS space, along with John van der Laan and Jeremy Wright for providing the data used to test the WATCHER pipeline on the rotor analysis UAS case. This work was partially funded by the US Department of Defense AARO, which funded some of the research associated with applying event-based sensing to the UAP application, taking advantage of the science and engineering being conducted for CUAS. This work was also partially funded by the US Department of Energy (DOE) National Nuclear Security Administration (NNSA), Office of Defense Nuclear Nonproliferation Research and Development.

ABOUT THE AUTHORS Kaylin Hagopian works at Sandia National Laboratories and is the primary author of the Wide Area Tracking and Characterization via Event-based sensing Reconnaissance pipeline. Dr. Joshua Shank leads the event-based sensing working group at Sandia National Laboratories and manages several projects related to nuclear nonproliferation remote sensing. This is a work of the US federal government and is not subject to copyright protection in the United States. Foreign copyrights may apply.

Ukrainian Military Observations and Analysis of Aerospace Objects and Phenomena during Defense against the Russian Federation

Ukrainian Military Observations and Analysis of Aerospace Objects and Phenomena during Defense against the Russian Federation By Drs. Artem Bilyk and Kyrylo Nikolaiev, Defense Intelligence Research Institute, Kyiv, Ukraine

Aerospace domain awareness is one of the main challenges for the national security of all countries. The widespread use of aerospace reconnaissance, communications, and navigation requires reliable monitoring, tracking, and analysis of aerospace objects and phenomena (AOP) and their actions in the aerospace domain (ASD). Timely detection and identification of AOP is an important task for improving situational awareness in the ASD to ensure airspace security.

Some AOP, after detailed analysis, are identified as drones, balloons, lanterns, or even clouds. However, some objects and phenomena cannot be immediately identified because they demonstrate characteristics incompatible with available data on known objects: high speeds, instantaneous acceleration, lack of visible signs of propulsion, non-trivial light effects, and so on. Some of these objects, which could be initially referred to as unidentified flying objects (UFOs), are eventually identified, but some are recognized as anomalous based on these factors and are formally categorized as Unidentified Aerial Phenomena (UAP). UAP pose a potential threat because they may collide with aircraft and because of the effects of their light, radiation, and electromagnetic emissions on technical systems, people, and the environment.

According to a US Department of Defense (DoD) report, “unknown drones” have been a growing threat to critical national infrastructure, military bases, and other facilities for the past few years. Undoubtedly, a large share of these reports reflects the wide availability and variety of civilian drones, as well as the smuggling, industrial espionage, and intelligence activities of other countries. Many of these “unknown drones” exhibit the unusual UAP characteristics mentioned above.

Leading countries such as the United States, France, Norway, and Argentina maintain specialized units and organizations responsible for centralized AOP data collection and identification. Identifying AOP is a key issue, as they can pose a threat to national security, especially during wartime. Reports published by US governmental organizations including the All-domain Anomaly Resolution Office (AARO) and the Office of the Director of National Intelligence (ODNI), as well as other organizations studying UAP, state that UAP pose a clear national security threat because their origins, motives, and capabilities are unknown. Therefore, monitoring and studying UAP is important due to the potential danger they pose and their constant presence in controlled and restricted ASD.

Russia’s war against Ukraine, which began locally in 2014, has been full-scale since February 2022. One feature of this war is the saturation of the operational environment with reconnaissance systems, the use of unmanned aerial vehicles (UAVs), and a wide variety of existing and newer weapons at an unprecedented scale. These factors have had a major impact on UAP observations. Ukrainian civilians and military personnel now pay far greater attention to objects in the sky because their safety and lives depend on it. A significant number of types of AOP associated with combat operations have emerged, and false identifications of civil aircraft have decreased because martial law prohibits civil aviation flights. Due to the full-scale aggression of the Russian Federation, hundreds of thousands of monitoring systems and tools operate continuously across various platforms to record events in the ASD, particularly in combat zones, while detecting potentially threatening enemy air attack and reconnaissance assets. As the number of ASD monitoring systems expands and the level of professionalism among personnel increases, both the number and the quality of AOP reports continue to rise.

The purpose of this article is to introduce the Ukrainian methodology for observing, detecting, and identifying AOP; to provide an overview of Ukrainian military UAP video-detection cases during the 2023-2024 full-scale invasion by the Russian Federation; to contribute to ongoing UAP research; and to help other countries improve their ASD security.

AOP Monitoring by the Armed Forces of Ukraine Figure 1 shows the process of observing, detecting, and identifying AOP by the Armed Forces of Ukraine. The main stages are: Observation -> Detection -> Primary Identification -> Data preparation for threat assessment / Data preparation for final identification -> Final identification.

Observation and Detection Observation of the ASD is carried out visually or through technical means. Detection of AOP is the process of establishing their presence through observation using various methods (visual, thermal, acoustic, radar, etc.). Monitoring equipment and devices include individual (micro-level), local (meso-level), and global (macro-level) electronic and mechanical systems that measure and record environmental conditions. Rapidly developing individual electronic and mechanical systems include mobile devices, smartphones, surveillance cameras, mini radars, and portable stations for measuring meteorological conditions and other environmental parameters. Local systems currently include departmental and private devices, aircraft, and other platforms. Global monitoring systems for ASD in Ukraine include satellites, meteorological stations, and radar stations.

Visual monitoring devices and instruments include cameras, night vision devices, telescopes, and binoculars. These devices can be portable, stationary, or mounted on moving vehicles and equipment. Because of the war, thousands of military and civilian cameras have been deployed along the frontline and across Ukraine, recording video and thermal imagery. Radio reconnaissance devices, widely used by the Ukrainian military, can detect AOP radio emissions and help to identify them. Radar detection uses specialized systems that vary in design, range, operational altitude, and frequency bands. Satellites and spacecraft monitor AOP from hundreds of kilometers above Earth, while space-based imagery is capable of providing highly accurate visual detection, thermography, and electronic reconnaissance. Nearly any type of AOP data-collection equipment can be installed on satellites and spacecraft. Ukraine, lacking its own reconnaissance satellites, relies on foreign partners (the United States, Europe, and others) and commercial satellite imagery for operational information.

Manned and unmanned aerial vehicles are the most common platforms for monitoring systems in the ASD, primarily due to their low cost. The active use of UAVs, especially near the war zone, makes it possible to monitor AOP in close to real time and to analyze threats. Manned vehicles can provide more detailed information and support faster decision-making when crews detect AOP, but they put those crews at risk. The low visibility and maneuverability of UAVs, along with some models’ ability to hover for extended periods over a point on the Earth’s surface, enable extensive data collection. Their high speed allows them to maneuver closer to AOP for better visibility. Thousands of Ukrainian military UAVs operate continuously, many of them provided by international partners.

Special stationary and mobile monitoring systems are among the most promising means of monitoring and detecting AOP because they are relatively inexpensive, support the deployment of monitoring networks, are easy to disguise, and can be quickly upgraded. Stationary systems are located on the ground, on towers and masts, on building elements, and in other locations. Mobile monitoring systems are deployed on vehicles, equipment, and specialized unmanned mobile platforms. Modern monitoring systems are typically multispectral, but require specialized hardware, data storage, and transmission systems. For example, since 2023, Ukraine has deployed a nationwide system of acoustic sensors aimed at detecting drones in real time based on the characteristic noise of their engines and to predict their trajectories.

Ukraine produces some technical monitoring systems domestically, but since 2022, it has received many through international technical aid or purchased them from foreign partners, including the United States, Norway, France, Great Britain, and others.

Identification Identification is the process of establishing the identity of AOP based on the parameters and features of known anthropogenic and natural AOP. Two types of identification are distinguished: primary and final.

Primary Identification Primary identification is carried out directly by personnel involved in monitoring and detecting AOP or is automated using specialized technical systems. It provides the data needed to assess threats and support rapid decisions on how to neutralize them, if necessary. The data for threat assessment is prepared by the personnel involved in monitoring and detecting AOP, or other specialists, who promptly transmit it to the appropriate officials for decision-making. Threat-assessment data include key AOP parameters such as direction of movement, trajectory, speed, and general description. These data should be sufficient to enable decisions on how to neutralize possible threats, which is the top governmental priority. Inability to identify an AOP during primary identification does not prevent the preparation of threat-assessment data or decisions on potential neutralization.

AOP can remain unidentified at the primary stage for many reasons: poor observation conditions; limited capabilities of monitoring instruments; variability in existing and emerging aerial technologies (aircraft, UAVs, missiles) used by both Ukrainian and Russian forces; specialized flight modes (e.g., very low altitudes, no lights or identification marks, loudspeakers, thermite payloads); and the use of fiber-optic–controlled drones by both sides, which can operate inside buildings or lie in ambush on the ground. The enemy is actively using camouflage and deception measures to avoid Ukrainian air defenses. These measures include special UAV coloring; radio-absorbing and thermal-insulating coatings; engine noise reduction; the use of false targets, such as air balloons and reflector-equipped UAVs; missiles with dipole reflectors (strips of foil or pieces of metallized fiberglass) and heat traps; and UAV camouflaged as birds.

Final Identification Final identification is aimed at conducting further analysis when an AOP remains unidentified after primary identification. In such cases, the primary identification data are sent to other specialists for final identification analysis. Data required for final identification include AOP parameters and features. AOP parameters describe the AOP quantitatively and can be expressed numerically (e.g., speed, size, brightness). AOP features describe AOP qualitatively and cannot be expressed numerically (e.g., color, shape). The parameters and features should comprehensively characterize the AOP to support identification with a high level of confidence; be measured using available tools and instruments; and have known limits of variability for different observation conditions. The final identification of AOP may require specific calculations, the use of specialized software, laboratory tests, field work, and so on. A designated official makes the final AOP identification decision.

Final identification analysis may show that initially unidentified AOP with unusual parameters and features are the latest developments in enemy aircraft or UAV, effects of weapons use, or other phenomena. Parameters and features that are highly unlikely to match those of known objects and phenomena are classified as anomaly factors. If anomaly factors remain after final identification, the AOP is recognized as anomalous and is categorized as UAP.

Table 1. Examples of UAP Anomaly Factors:

• Appearance: Sudden appearance; gradual appearance in clear skies
• Body: Shape uncharacteristic of known objects or phenomena; blurred shape
• Details: Lack of details inherent in known objects or phenomena; presence of unusual details
• Movement: Sudden acceleration or stopping; ultra-high velocity; turning at high velocity at sharp angles; abnormal trajectory; long hovers in one place; high-speed rotation; inappropriate location; non-trivial trajectory; trans-domain movement
• Transformation: Splitting into parts; sudden change of shape, metamorphosis; separation or joining of objects
• Effects: Color change, unusual colors; ultra-brightness, flashes, pulsations; high or low temperature; rays with finite length, fragmented rays; different types of radiation; non-trivial, remote influences
• Interactions: Landing or take-off in unsuitable locations; environmental, technical, and biological impacts; residual effects; environmental changes
• Disappearance: Sudden disappearance; visual invisibility; gradual disappearance in clear sky

Public and Military Study of Anomalous Phenomena Because of these threats, both the public and military in Ukraine are engaged in the study of anomalous phenomena, which also presents opportunities. In the public sector, the Ukrainian Scientific Research Center for the Analysis of Anomalies (“Zond”), a separate unit of the Aerospace Society of Ukraine, has conducted extensive research on anomalous phenomena since 2004.

In the military sector, the Armed Forces of Ukraine issued a special order in 2023 stating the need to track unidentified objects in the sky. In 2024, the Ukrainian Defense Forces established a special unit—a research laboratory that analyzes AOP reports and collects and investigates UAP cases. This unit has produced two manuals on AOP identification and UAP information analysis. It has also developed a questionnaire to collect detailed information from military eyewitnesses. To avoid misidentification, the unit accepts only reports that include photos, videos, or other material evidence. It has also developed a scientific UAP identification methodology, and has begun specialized military training to increase ASD awareness and prepare personnel to receive UAP observation reports.

At the same time, a special “ePPO” (“electronic air defense”) app supports users in both sectors. Voluntary programmers from the Technari group developed the app, which collects real-time coordinates of observed objects and phenomena in the ASD. A separate “UFO” button allows civilian eyewitnesses to report initially unidentified objects other than those that are clearly of artificial origin (drones, airplanes, helicopters, missiles, quadcopters, aerial explosions). This app currently has more than 800,000 users. Civilian reports support military identification of AOP, because of the large number of potential witnesses and their uniform distribution across the territory. Based on these data, ePPO processing is performed by artificial intelligence (AI), which identifies the objects and predicts their location and flight trajectories. If necessary, a shootdown is planned.

Examples of UAP Observed by the Ukrainian Military Table 2. Summary of Ukrainian UAP Observed by the Military (2023–2024):

• 15 Feb 2023: cigar-shaped vertical | Size: 0.3–0.5 m | Number: 1 | Velocity: 250 m/s | Altitude: 200…300 m | Reference: Figure 9.b (Bayraktar UAV in central Ukraine)
• 24 Feb 2024: disc/cigar-shaped | Size: 26.4–32.4 m | Number: 1 | Velocity: 0 m/s | Altitude: 105 m | Reference: Figure 8.b (Mavic 3T UAV)
• 26 Mar 2024: spherical | Size: 1.3–2.7 m | Number: 4 | Velocity: 20–42 m/s | Altitude: 50–100 m | Reference: Figure 10.b (Mavic 3T UAV)
• Feb or Mar 2024 (specific date unknown): spherical | Size: 1.2–6 m | Number: 5 | Velocity: 7.3–36.5 m/s | Altitude: 100–200 m | Reference: Figure 8.a (Mavic 3T UAV wedge formation)
• 28 Mar 2024: spherical | Size: 2–2.9 m | Number: 4 | Velocity: 76.2–110.8 m/s | Altitude: 100–150 m | Reference: Figure 9.a (Mavic 3T UAV in-line formation)
• 24 Dec 2024: spherical | Size: 0.3–0.65 m | Number: 5 | Velocity: 7.5–16.6 m/s | Altitude: 20–40 m | Reference: Figure 10.a (Mavic 3T UAV, visible on thermal but invisible to conventional camera)

Conclusion Due to the war in Ukraine, civilians, military personnel, and thousands of technical systems continuously conduct large-scale observations to detect AOP. This has increased the observation and reporting of unknown objects in the airspace, many of which are determined to be of natural or artificial origin, such as enemy drones and other reconnaissance and attack systems. However, some objects are classified as UAP because of their anomalous characteristics. For example, UAP often fly in groups, and may be visible on thermal imagers but invisible to conventional cameras.

Recognition and study of UAP are matters of national security as well as challenges for modern science. Effective cooperation among all components of the national security and civilian sectors is essential to detect, identify, and, if necessary, neutralize potential threats as quickly as possible. Coherent, coordinated ASD monitoring significantly reduces misidentification and related losses.

Timely, reliable AOP identification protects military personnel, prevents fratricide from mistaken shootdowns, conserves scarce air defense ammunition, and enables the collection of UAP characteristics for further scientific study of these potential threats and their technologies.

ABOUT THE AUTHORS Dr. Artem Bilyk is an associate professor at the Defense Intelligence Research Institute in Kyiv. Dr. Kyrylo Nikolaiev is an associate professor at the Defense Intelligence Research Institute in Kyiv. This is a work of the US federal government and is not subject to copyright protection in the United States. Foreign copyrights may apply.

Capturing Unidentified Anomalous Phenomena Events: A Practical Photography Guide for Everyday Observers

Capturing Unidentified Anomalous Phenomena Events: A Practical Photography Guide for Everyday Observers By J. Kevin Ryan, Special Agent, US Air Force (Ret.)

Recent attention from governmental bodies, including the All-domain Anomaly Resolution Office (AARO), the Department of Defense, and NASA, has underscored the national interest in documenting Unidentified Anomalous Phenomena (UAP). AARO’s findings have highlighted a critical shortage of high-quality visual data, particularly from consumer sources, and emphasized the need for validated methods to record these events. Historically, reports of UAP have consisted mainly of anecdotal narratives and vague imagery lacking the rigor required for scientific inquiry. Today, the ubiquity of high-resolution consumer cameras presents a pivotal opportunity—provided those cameras are used correctly.

The challenge is that most cameras are not designed to photograph UAP. Modern imaging systems are optimized for predictable subjects: portraits, landscapes, wildlife, and sports. Their automatic exposure, focus, and stabilization routines assume a known subject at a known distance behaving in a known way. By definition, UAP conform to none of these assumptions. They appear without warning, at unknown range, moving in ways that defeat autofocus and confuse metering systems. In this sense, the most capable UAP imaging platform is not the most sophisticated one—it is the most manual one, one in which the photographer, not the camera, controls every parameter of image capture.

The scientific study of UAP—objects whose origins and behaviors, often due to insufficient technical data, cannot be readily identified as conventional, human-made, or natural phenomena—requires rigorous, reproducible, and technically sound imaging data. Peer assessments consistently find that most UAP cases remain unsolved because of the chronic absence of high-quality imagery, particularly at night, when sightings are most frequently reported. If a phenomenon has no previously known cause, the evidentiary burden on the image is even greater.

Technical Requirements for UAP Imaging High-value UAP imagery differs fundamentally from “viral” footage intended for mass consumption. Organizations such as AARO seek high-resolution, in-focus, high-contrast, and contextual photos and videos. Higher resolution is achieved by getting physically closer to the object, using higher-quality equipment (larger, better lenses), and providing scene context through multiple frames. Scientific value is defined by the potential for the following: • Photometric analysis: Well-calibrated, unaltered images enable estimates of object brightness, color, and surface properties, which can reveal potential physical characteristics. • Triangulation: Precise spatial and temporal data enable researchers to reconstruct an object’s position and speed using geometry, provided multiple observers or devices record the object. • Chain of custody: Original images and video must retain intact metadata and be handled so that third parties can verify their provenance, supporting credibility in peer review and official investigations.

Table 1: UAP Documentation Formats and Values

• Unprocessed (RAW) (.DNG, .ARW, .CR3, .NEF) | Forensic Value: Highest | Scientific Rationale: Direct sensor output; contains the highest dynamic range and most complete metadata for astrometric calibration.
• High-Bitrate Video (.MOV / .MP4 ProRes, All-I) | Forensic Value: High | Scientific Rationale: Minimal inter-frame compression; preserves motion vectors and luminosity changes with high fidelity.
• Standard Video (.MP4 / .MOV) | Forensic Value: Moderate | Scientific Rationale: Subject to compression artifacts; useful for motion but less reliable for morphology (shapes and structure).
• Process Processed Still (.JPEG, .HEIC) | Forensic Value: Low | Scientific Rationale: 8-bit depth; “baked-in” sharpening and noise reduction destroy low-contrast data and faint signals.
• Proprietary Meta (.XMP, .LOG) | Forensic Value: Auxiliary | Scientific Rationale: Sidecar files (small text-based files that live alongside your raw image) that document every edit made to a RAW file, ensuring the audit trail is intact.

Timestamp Accuracy Synchronizing your devices to global time standards is essential. UAP observers should use devices synchronized via GPS, Network Time Protocol (NTP), or Coordinated Universal Time (UTC). Even a few seconds of timestamp error can confound later triangulation efforts.

Metadata Preservation and Provenance Metadata (“data about data”) is information stored within an image file that describes how, when, and where the camera captured the image. EXIF is the most common metadata standard. UAP photographers should keep a contemporaneous log that includes observer name, date, time (UTC), location, orientation, equipment, settings, and narrative.

Equipment Considerations • Camera Systems and Sensor Architectures: Full-frame (36×24 mm) vs. APS-C (~22×15 mm). Support RAW capture and understand sensor ISO invariance. • Lenses: Balance aperture and resolving power. Prime lenses recommended; mid-telephoto (35–135mm full-frame equivalent) provides standard balance. Use lenses with hard infinity stops or calibrated manual focus scales. • Tripods and Stabilization: Prevent motion blur using a high-mass tripod with fluid or ball head, center-column ballast hook. Static, fixed camera provides consistent spatial reference. • Power and Storage: Use V60 or V90 high-endurance memory cards to prevent buffer lockups/corruption. Use external USB-C PD power banks to counter cold-weather battery drops. • Optical vs Digital Zoom: Use optical zoom only; digital zoom is destructive interpolation.

Photographs vs. Video • Photos maximize spatial and photometric resolution and dynamic range. • Video provides continuous temporal context, acceleration estimates, and flash frequency. • Cinematic vs Forensic Video: Use 30–60+ fps, 180° shutter angle (e.g. 1/60s at 30fps), high bitrate, manual focus locked to infinity.

Smartphones: Data Preservation vs Computational Artifacts Bypass computational post-processing and night modes. Use apps like Halide or Open Camera.

Table 2: Smartphone Parameters, Settings, and Rationales:

• Capture Format: RAW / ProRAW / DNG (Prevents destructive smoothing and preserves metadata)
• Focus: Manual (Infinity Lock) (Prevents focus hunting in low-contrast night skies)
• Frame Rate: Fixed (e.g., 30 or 60 fps) (Enables precise calculation of object velocity and acceleration)
• Exposure: Manual / Locked (Ensures brightness changes are physical, not auto-adjustments)
• Zoom: 1x (Optical Only) (Avoids digital interpolation and loss of spatial resolution)

The Exposure Triad in Low-Light UAP Capture:

• Aperture: Widest practical aperture (f/1.4–f/2.8) while avoiding severe aberrations, or stop down to f/4 if depth of field is needed.
• Shutter Speed: Match to target motion; fast shutter (e.g., 1/125s+) freezes motion, longer exposures (1–2s) capture faint objects.
• ISO Setting: Keep within native range to keep signals above noise floor without clipping highlights.

Post-Capture Workflow and Forensic Integrity:

1. Hash files immediately using SHA-256 to establish an ironclad digital fingerprint.
2. Retain unmodified originals with intact EXIF metadata.
3. Make only global, non-destructive edits and preserve XMP sidecar files.
4. Follow standard submission protocols: original RAW/video, metadata logs with UTC timestamps, calibration/sky star frames, and SHA-256 hashes.

Conclusion When executed properly, consumer-grade nighttime imaging of UAP can produce credible evidence suitable for quantitative analysis. Although UAP appearances are unpredictable, consistent preparation, validated settings, and disciplined metadata retention can transform fleeting observations into durable, shareable data.

ABOUT THE AUTHOR J. Kevin Ryan is a retired federal special agent with more than 37 years of combined military and civilian service with the Air Force Office of Special Investigations. He concluded his federal career as Chief of Operations for the All-domain Anomaly Resolution Office.

Call for Submissions

Call for Submissions

Combating Threats Exchange (CTX) is a biannual peer-reviewed journal. We accept submissions of nearly any type, from anyone; however, submission does not guarantee publication. Our aim is to distribute high-quality analyses, opinions, and studies to military officers, government officials, and security and academic professionals in the irregular warfare community. We give priority to non-typical, insightful work and to topics concerning the most pressing terrorism and hybrid conflict concerns.

CTX periodically publishes special issues—such as this important edition—to highlight emerging challenges in the national security and defense domains. To learn more about our next special issue, please visit: https://nps.edu/web/ccht/forthcoming-call-for-papers

Submission Guidelines: For detailed submission guidelines, go to https://GlobalECCO.org and click on the CTX Home link. Then click on Submissions in the left menu bar.

CTX accepts the following types of submissions. Please observe the following length guidelines:

• academic analyses (up to 6,000 words)
• reports or insightful stories from the field (up to 5,000 words)
• photographic essays
• video clips with explanation or narration
• interviews with relevant figures (videos no longer than 15 minutes, transcriptions no longer than 6,000 words)
• book reviews (up to 2,000 words), review essays (up to 2,000 words), or lists of books of interest (which may include books in other languages)
• reports on any special projects
• Any kind of submission can be multimedia.

Submission Requirements:

• Submissions must be original and not have appeared in any other publication.
• The work must be copyedited for basic errors prior to submission.
• Citations should adhere to the Chicago Manual of Style.
• The work submitted may not be plagiarized in part or in whole.
• You must have permission to use any images, videos, or statements included in your work.
• You must agree to our Terms of Copyright, which appear at the front of the journal.
• Include a byline as you would like it to appear and a short bio as you would like it to appear.

Submissions should be sent in original, workable format (such as Microsoft Word—no PDFs, please) in English.

Contact: [email protected] CTX is online at globalecco.org