Currently, traditional methods such as short-term average/long-term average (STA/LTA) are used to detect arrivals in three-component seismic waveform data. Accurately establishing the identity and arrival of these waves is helpful in detecting and locating seismic events. Convolutional Neural Networks (CNNs) have been shown to significantly improve performance at local distances. This work will expand the use of CNNs to more remote distances and lower magnitudes. Sandia National Labs (SNL) will explore the advantages and limits of a particular approach and investigate requirements for expanding this technique to different types, distances, and magnitudes of events in the future. The team will describe detailed performance results of this method tuned on a curated dataset from Utah with its expert-defined arrival picks.
Through cyberattacks on information technology and digital communications systems, antagonists have increasingly been able to alter the strategic balance in their favor without provoking serious consequences. Conflict within and through the cyber domain is inherently different from conflict in other domains that house our critical systems. These differences result in new challenges for defending and creating resilient systems, and for deterring those who would wish to disrupt or destroy them. The purpose of this paper is to further examine the question of whether or not deterrence can be an effective strategy in cyber conflict, given our broad and varied interests in cyberspace. We define deterrence broadly as the creation of conditions that dissuade antagonists from taking unwanted actions because they believe that they will incur unacceptably high costs and/or receive insufficient benefits from taking that action. Deterrence may or may not be the most credible or effective strategy for achieving our desired end states in cybersecurity. Regardless of the answer here, however, it is important to consider why deterrence strategies might succeed under certain conditions, and to understand why deterrence is not effective within the myriad contexts that it appears fail. Deterrence remains a key component of U.S. cyber strategy, but there is little detail on how to operationalize or implement this policy, how to bring a whole-of-government and whole-of- private-sector approach to cyber deterrence, which types of antagonists can or should be deterred, and in which contexts. Moreover, discussion about how nations can and should respond to significant cyber incidents largely centers around whether or not the incident constitutes a "use of force," which would justify certain types of responses according to international law. However, we believe the "use of force" threshold is inadequate to describe the myriad interests and objectives of actors in cyberspace, both attackers and defenders. In this paper, we propose an approach to further examine if deterrence is an effective strategy and under which conditions. Our approach includes systematic analysis of cyber incident scenarios using a framework to evaluate the effectiveness of various activities in influencing antagonist behavior. While we only examine a single scenario for this paper, we propose that additional work is needed to more fully understand how various alternative thresholds constrain or unleash options for actors to influence one another's behavior in the cyber domain.
Safety-focused risk analysis and assessment approaches struggle to adequately include malicious, deliberate acts against the nuclear power industry's fissile and waste material, infrastructure, and facilities. Further, existing methods do not adequately address non- proliferation issues. Treating safety, security, and safeguards concerns independently is inefficient because, at best, it may not take explicit advantage of measures that provide benefits against multiple risk domains, and, at worst, it may lead to implementations that increase overall risk due to incompatibilities. What is needed is an integrated safety, security and safeguards risk (or "3SR") framework for describing and assessing nuclear power risks that can enable direct trade-offs and interactions in order to inform risk management processes -- a potential paradigm shift in risk analysis and management. These proceedings of the Sandia ePRA Workshop (held August 22-23, 2017) are an attempt to begin the discussions and deliberations to extend and augment safety focused risk assessment approaches to include security concerns and begin moving towards a 3S Risk approach. Safeguards concerns were not included in this initial workshop and are left to future efforts. This workshop focused on four themes in order to begin building out a the safety and security portions of the 3S Risk toolkit: 1. Historical Approaches and Tools 2. Current Challenges 3. Modern Approaches 4. Paths Forward and Next Steps This report is organized along the four areas described above, and concludes with a summary of key points. 2 Contact: rforres@sandia.gov; +1 (925) 294-2728