Force Development—Undersea Networks: Nord Stream Case Study

Executive Summary

As part of CNA's Force Development initiative, we developed a Monte Carlo simulation model that assesses the potential benefits of undersea networks that can provide communications, power, and sensing data. This model, similar to the other "building block modules," was designed to explore a particular component of force design—in this case, the promise and opportunities presented by undersea networks. In this paper, we use the model to assess a real-world event—the September 2022 sabotage of the Nord Stream pipeline—to analyze the capabilities that an in-place undersea network would have needed to provide timely warning of the sabotage.

The Nord Stream pipeline is approximately 660 nautical miles long and consists of two natural gas pipelines running from Russia to Germany through the Baltic Sea. For this study, we examined a scenario in which unmanned underwater vehicles (UUVs) are used to search for potential sabotage, with a series of underwater nodes to charge the UUVs and provide them with a communications capability to report back. For our baseline UUV, we used the General Dynamics Bluefin-21, which is used by the US Navy for a variety of missions and has a 25-hour endurance at 3 knots, an assumed maximum recharge time of 4 hours, and an assumed probability of detection (PD) of 0.7. Based on these inputs, we found the following:

• Many searchers and nodes are required for timely capabilities. An undersea network of at least 21 UUVs and 23 network nodes would have been required to discover and report an act of sabotage in under 24 hours on average (mean time to report (MTR)). Expanding the timeline to 48 hours enables different, slightly lesser, combinations of searchers and nodes (12 to 20 searchers with 23 nodes, 14 to 20 searchers with 18 nodes, and 20 searchers with 15 nodes).
• The ability to coordinate between searchers provides significant benefits, particularly in cases with fewer searchers and nodes. MTR for 21 searchers and 23 nodes with coordination is comparable to the random model results (22.3 hours versus 23.3 hours, respectively), but for 6 searchers and 12 nodes, the coordinated MTR is 74.1 hours and the random model MTR is 354.2 hours.
• Based on those two findings, focusing UUV development on speed, battery life, and coordination capabilities appears to have the highest return on investment if maritime domain awareness missions such as this are of interest.
• For 0.7 P_D', approximately 314 combined search hours are required to "cover" the pipeline once. This time drops to 244 hours for 0.9 P_D' and increases to 733 hours for 0.3 P_D'. MTR is inversely related to the capability of each searcher, but low individual capability can be compensated for by adding more searchers.
• The delay between detection and reporting is at least the time it takes the UUV to transit from the detection to the next node. Spacing the nodes less than half of the UUV’s endurance range apart hedges against the loss of a node and reduces the time between detection and reporting.
• Variation in the time to charge each UUV from 16 to 32 percent of search time (4 to 8 hours) affected MTR by 10 percent at most.

As indicated by the above findings, our undersea network model can be used to analyze realworld situations and has the flexibility to assess the effect of changes in system parameters and employment concepts.