Mexican Officials Make Record Fentanyl Seizure Days After Trump Tariff Warning

Mexican Officials Make Record Fentanyl Seizure Days After Trump Tariff Warning

Mexican Officials Make Record Fentanyl Seizure Days After Trump Tariff Warning

By Jack Phillips of The Epoch Times

Mexican security forces said on Dec. 4 that they had made the largest fentanyl seizure in the country’s history, impounding 1,100 kilograms (1.2 tons) of the synthetic opioid in the state of Sinaloa.

Mexico’s top security official, Omar García Harfuch, said in a statement that more than a ton of fentanyl was seized by officials in Sinaloa state. Several guns were also seized, and two men were arrested, he said.

“This is an investigation that has been going on for a long time, and yesterday, it gave these results,” Mexican President Claudia Sheinbaum said at a press conference on Dec. 4, referring to the fentanyl seizures.

Violence has worsened recently in Sinaloa, where factions of the Sinaloa Cartel have been engaged in bitter fighting that flared after the capture of kingpin Ismael “El Mayo” Zambada in July.

“These actions will continue until the violence in the state of Sinaloa decreases,” Harfuch said.

Sinaloa is home to the powerful drug cartel that bears the same name and was formerly headed by longtime drug lord Joaquín “El Chapo” Guzmán, who is currently incarcerated at the ADX Florence federal prison in Colorado.

Security forces found the fentanyl at two properties in the municipality of Ahome after intelligence work and tip-offs from the public led them to investigate there.

In one building, law enforcement found 800 kilograms (1,763 pounds) of fentanyl, some precursor chemicals, and four vehicles. In the other, they discovered 11 packages totaling about 300 kg (660 pounds) of fentanyl, as well as precursors, scales, and industrial mixers.

Former Mexican President Andrés Manuel López Obrador, who handed over power to Sheinbaum in October, repeatedly denied that Mexico was a center for the production of fentanyl despite significant evidence to the contrary.

U.S. President-elect Donald Trump recently threatened to levy a 25 percent tariff against Mexico and Canada if either country didn’t do enough to curb illegal immigration and fentanyl trafficking into the United States.

His warning prompted a phone conversation with Sheinbaum, with the Mexican president later saying that caravans of migrants will be stopped before they reach the U.S.–Mexico border. However, she denied Trump’s claim last week that the Mexican border was closed down.

This week, activists and a Mexican agency said a migrant caravan heading north was dissolved. The Mexican National Migration Institute denied claims that the agency used deceptive tactics and said it had not received “any complaints” from members of the caravan.

Meanwhile, Canadian Prime Minister Justin Trudeau met with Trump at his Mar-a-Lago residence in Palm Beach, Florida, before several top Canadian officials assured reporters that the country would improve its border security with the United States.

Continue reading at the Epoch Times

Tyler Durden
Fri, 12/06/2024 – 23:00

Data Centers Are Sending Global Electricity Demand Soaring

Data Centers Are Sending Global Electricity Demand Soaring

Data Centers Are Sending Global Electricity Demand Soaring

Authored by Felicity Bradstock via OilPrice.com,

  • The rapid growth of data centers to support AI is significantly increasing global electricity demand.

  • This surge in demand threatens to outpace the development of renewable energy sources.

  • International regulations are needed to ensure tech companies use clean energy and minimize their impact on climate goals.

The global electricity demand is expected to grow exponentially in the coming decades, largely due to an increased demand from tech companies for new data centers to support the rollout of high-energy-consuming advanced technologies, such as artificial intelligence (AI). As governments worldwide introduce new climate policies and pump billions into alternative energy sources and clean tech, these efforts may be quashed by the increased electricity demand from data centers unless greater international regulatory action is taken to ensure that tech companies invest in clean energy sources and do not use fossil fuels for power.

The International Energy Agency (IEA) released a report in October entitled “What the data centre and AI boom could mean for the energy sector”. It showed that with investment in new data centers surging over the past two years, particularly in the U.S., the electricity demand is increasing rapidly – a trend that is set to continue. 

The report states that in the U.S., annual investment in data center construction has doubled in the past two years alone. China and the European Union are also seeing investment in data centers increase rapidly. In 2023, the overall capital investment by tech leaders Google, Microsoft, and Amazon was greater than that of the U.S. oil and gas industry, at approximately 0.5 percent of the U.S. GDP.

The tech sector expects to deploy AI technologies more widely in the coming decades as the technology is improved and becomes more ingrained in everyday life. This is just one of several advanced technologies expected to contribute to the rise in demand for power worldwide in the coming decades. 

Global aggregate electricity demand is set to increase by 6,750 terawatt-hours (TWh) by 2030, per the IEA’s Stated Policies Scenario. This is spurred by several factors including digitalization, economic growth, electric vehicles, air conditioners, and the rising importance of electricity-intensive manufacturing. In large economies such as the U.S., China, and the EU, data centers contribute around 2 to 4 percent of total electricity consumption at present. However, the sector has already surpassed 10 percent of electricity consumption in at least five U.S. states. Meanwhile, in Ireland, it contributes more than 20 percent of all electricity consumption.

While the speed and manner in which AI use will grow remains uncertain, and efficiency improvements are expected to be made, electricity demand from data centers, cryptocurrencies, and AI could reach as much as 1,000 Terawatt Hours (TWh) in 2026 – roughly equivalent to the electricity consumption of Japan – compared to 460TWh today, the IEA predicts.

The organization calls for more public-private dialogue, with policymakers, the tech sector, and the energy industry coming together for discussions to manage both expectations and energy use. Greater international regulation of the tech sector is required to ensure that the growing electricity demand for data centers does not outweigh the green transition achievements currently being seen worldwide. 

There are growing fears that, if left unregulated, the electricity consumption of data centers could surpass the electricity demand of some U.S. cities or even states. Many data center developers are concerned about finding enough land to house new sites and enough clean power to run them. The facilities could increasingly require 1 GW of power or more, which is equivalent to around twice the 2023 residential electricity consumption of Pittsburgh

The president of Lancium, a company that secures land and power for data centers in Texas, Ali Fenn, explained that U.S. tech companies are in the “race of a lifetime to global dominance”. Fenn said, “They’re going to keep spending” because there’s no more profitable place to deploy capital. 

At the rate the advanced technologies are expanding, renewable energy sources will not be sufficient to meet the growing demands of the tech industry. Many tech companies are expected to use natural gas to power operations, particularly in the U.S. where the gas sector is set to continue expanding rapidly. 

Currently, many tech companies operate data centers with a capacity of around 40 MW. However, in the coming years, more firms are expected to invest in campuses of 250 MW or more. As a growing number of campuses of 500 MW or more emerge in the 2030s and 2040s, which is equivalent to the power needed for 350,000 homes, this could lead to a surge in demand for gas-generated electricity, following years of national investment in a green transition.  

While the U.S. is expected to see the greatest data center expansion in the coming decades, Europe’s data center power consumption is expected to nearly triple by the end of the decade. Meanwhile, China has invested over $6.12 billion in a national project to develop data centers in recent years, according to a senior government official. 

A joined-up approach to regulating the energy usage of data centers is required to prevent the anticipated rise in electricity demand from challenging the progress of the global green transition. Governments worldwide must establish clear regulations and limits on the energy use of tech companies for advanced technologies, such as AI, if they hope to meet Paris Agreement climate pledges. This may include requiring tech companies to fulfill their energy needs through clean energy sources, such as renewables and nuclear power, as well as slowing the pace of deployment of these technologies.  

Tyler Durden
Fri, 12/06/2024 – 18:00

CEO Assassin May Have Used ‘Covert-Type’ Weapon – Took Greyhound Bus To NYC Last Month

CEO Assassin May Have Used 'Covert-Type' Weapon - Took Greyhound Bus To NYC Last Month

CEO Assassin May Have Used ‘Covert-Type’ Weapon – Took Greyhound Bus To NYC Last Month

The NYPD’s principal “person of interest” in the assassination of UnitedHealthcare CEO Brian Thompson arrived in Manhattan late last month after a long bus ride from Atlanta — then stayed at a hostel ahead of Tuesday’s shocking act, law enforcement sources tell various outlets. Many more details are piling up — from the remarkable weapon cops think he used to the type of pricey backpack he seems to have ditched in Central Park — but there’s still no name to go with the smiling face captured by the hostel’s front-desk security camera.  

The person of interest in the assassination of UnitedHealthcare’s CEO revealed his face as he flirted with a female employee of the hostel where he stayed (NYPD)

Police say the man arrived in New York City on Nov. 24 — the Sunday before Thanksgiving — stepping off a Greyhound bus at the Port Authority bus terminal in Manhattan around 9pm. That particular bus route originates in Atlanta. Some reports say police don’t yet know where he boarded the coach, while others indicate investigators believe he was aboard from the start of the nearly 900-mile journey from Atlanta to New York, which can take 20 hours or more on Greyhound.  

Police sources say he then checked into the HI New York City Hostel  at Amsterdam Avenue and West 103rd Street, on the Upper West Side near Central Park. The revelation of his face appears to have come during a flirtatious moment of weakness as he interacted with a woman working the hostel’s front desk, lowering his mask to beam at her across the desk. The suspect left the hostel on Friday, Nov. 29, only to return via Uber to check back in the very next day.  

The suspect is believed to have stayed at the HI New York City Hostel on Manhattan’s Upper West Side 

He reportedly used a bogus New Jersey driver’s license to book his stays at the hostel, where he roomed with two apparent strangers. “That’s not clever, because there is recording, video all over the place,” John Nielsen, a visitor from Denmark who’s been staying at the hostel for the past week, told the Times.  

The suspect shared a hostel room like this one with two other people, whom police believe were strangers to him (via HI New York City Hostel)

Police have taken keen interest in a gun sale in Connecticut — involving a highly-uncommon weapon that appears similar to the one seen in the video of the killing. “It’s a covert-type weapon. It is not sold widely,” said CBS News law enforcement contributor Rich Esposito. “They’ll go to the dealers of that gun and try to establish if they know who this person is.” 

Police believe the killer used a B&T Station SIX-9 (“Silenced Project caliber 9”), which retails for around $2,100 

According to CBS News, police believe the murder weapon is a B&T Station SIX-9, which comes equipped with a sound-suppressor and retails for around $2,100. Note carefully how B&T describes its latest version: 

This updated, integrally suppressed pistol features a new grip and updated magazines, yet maintains its non-descript appearance, whisper-quiet sound signature and unique rotating bolt operation. This model in 9mm wears a 3-in. barrel. It finds its operational success as a magazine-fed, non-auto-ejecting single-shot functioning in a rotational bolt action. It is beautifully enhanced by wipe suppressor, and a baffled suppressor is an available option for those who prefer that design. 

The “non-auto-ejecting single-shot functioning” is consistent with what’s seen in the video of the shooting, as the assassin manipulates the pistol after firing each shot — something that both casual observers and experts had widely attributed to the pistol malfunctioning. Here’s how the operation of the Station SIX looks with the 45-caliber version: 

The suspect shot the 50-year-old Thompson — who was arriving at the New York Hilton Midtown for an investors’ meeting at 6:45 am — in the back and leg before leaving first by foot, then using an e-bike that he rode through Central Park. Police think he exited the park at West 77th Street — minus the backpack he wore during the killing and as he rolled into the park.

via New York Post 

Some internet sleuths think they’ve figured out the make and model of that backpack — concluding it’s an Everyday Backpack sold by Peak Designs between 2016 and 2019, TMZ reports. It’s not cheap: Marketed for use by photographers but also used by others, the latest version goes for about $250.  

Internet sleuths think the shooter was wearing a backpack model that retails for hundreds of dollars (via TMZ)

Investigators recovered 9mm shell casings and live rounds at the shooting that were adorned with the words “deny,” “defend” and “depose,” written carefully with a sharpieThose words are open to interpretation, but many are pointing to their similarity to the title of Jay Feinman’s book, Delay Deny Defend: Why Insurance Companies Don’t Pay Claims And What You Can Do About It.” Given that, police are looking for people who may have reason for outrage at an individual coverage decision rendered by the health insurance giant. That could be quite a long list of people: Recent data from Lending Tree shows UnitedHealthcare has some of the highest claim denial rates of any US insurance company —  a jarring 32% compared to 17% for Blue Cross Blue Shield. (They may also want to talk to former Washington Post writer Taylor Lorenz.) 

There are many more clues undergoing intense police scrutiny. Less than a half-hour before murdering Thompson, the suspect used cash to buy drinks and snacks at a Starbucks near the ambush site. Police think they’ve obtained a water bottle and candy wrapper he left at the store, along with a coffee cup he may have drank from. They also found a cell phone in the alleyway that comprised part of his getaway route — a route that, at least this far along, has proven quite effective. 

However, his flirty smile at a female hostel clerk and choice of a highly uncommon pistol may prove to be his undoing… 

Tyler Durden
Fri, 12/06/2024 – 06:55

Underwater Geopolitics

Underwater Geopolitics

Underwater Geopolitics

Authored by Carlo J.V.Caro via RealClearWire.com,

How China’s Control of Undersea Cables and Data Flows Reshapes Global Power

Cable Routing Protocols

The rapid construction of undersea cables has brought a hidden but crucial issue into focus: the manipulation of the protocols that control how data travels beneath the sea. These protocols determine the pathways internet data takes, influencing speed, costs, and even exposure to surveillance. Even small changes in these pathways can tilt the global balance of digital power. China’s increasing role in this area demonstrates how technology can be used strategically to reshape geopolitics.

At the heart of this issue is a technology called Software-Defined Networking (SDN). SDN allows data traffic to be managed and optimized in real time, improving efficiency. But this same flexibility makes SDN vulnerable to misuse. Chinese tech companies like HMN Tech (formerly Huawei Marine Networks), ZTE, and China Unicom are leading the way in SDN development. China also holds sway in international organizations that set the rules for these technologies, such as the International Telecommunication Union (ITU) and the Institute of Electrical and Electronics Engineers (IEEE). This influence gives China a hand in shaping global standards and governance.

Africa illustrates how this influence plays out. Chinese investments in digital infrastructure across the continent are massive. For example, the PEACE (Pakistan and East Africa Connecting Europe) cable, which links East Africa to Europe, was designed to avoid Chinese territory. Yet, thanks to SDN technology, its traffic can still be redirected through Chinese-controlled points. This redirection could introduce delays of 20 to 30 milliseconds per hop—not much for casual browsing, but a serious issue for latency-sensitive activities like financial trading or encrypted communication.

In Southeast Asia, similar risks are evident. The Southeast Asia-Japan Cable (SJC), which connects Singapore to Japan, relies on several landing stations influenced by China. During a period of heightened tensions in the South China Sea, some data intended for Japan was mysteriously routed through Hainan Island, under Chinese jurisdiction. Such cases suggest technical routing decisions may sometimes have political motivations.

These examples are part of a broader strategy. By exploiting SDN, China can turn submarine cables into tools for surveillance and control. Data traffic from Africa or Southeast Asia destined for Europe could be secretly rerouted through Shanghai or Guangzhou, exposing it to China’s advanced surveillance techniques like deep packet inspection. This threat extends to cloud computing, as major providers such as Amazon Web Services (AWS), Microsoft Azure, and Alibaba Cloud rely on undersea cables. With SDN, Chinese cloud providers—aligned with state interests—could redirect sensitive inter-cloud traffic, putting critical communications at risk.

Manipulating global data routes gives any actor significant geopolitical power. For instance, in a crisis, China could degrade or even sever internet connectivity for rival nations. In the Taiwan Strait, this could isolate Taiwan from global markets, disrupting financial transactions and trade. In Africa, where Huawei has built a significant portion of the continent’s telecommunications infrastructure—reportedly constructing around 70 percent of 4G networks—there is concern that this reliance could create vulnerabilities. If political tensions were to arise, China could cause slowdowns or disruptions to reinforce dependence, making countries more vulnerable in political standoffs.

The numbers highlight the stakes. Submarine cables carry 99 percent of international data traffic—over 1.1 zettabytes annually. Significant portions of intra-Asia-Pacific data flows pass through key submarine cable landing stations, including Hong Kong, which is under Chinese jurisdiction. With Chinese firms increasingly involved in substantial global submarine cable projects—such as those undertaken by HMN Technologies—Beijing’s influence over the internet’s physical backbone is growing.

The economic impact of internet disruptions on highly connected economies is substantial. For instance, the NetBlocks Cost of Shutdown Tool (COST) estimates the economic impact of internet disruptions using indicators from the World Bank, ITU, Eurostat, and the U.S. Census. According to data presented by Atlas VPN, based on NetBlocks’ COST tool, a global internet shutdown for one day could result in losses of about $43 billion, with the United States and China accounting for nearly half of this sum. Additionally, Deloitte has estimated that for a highly internet-connected country, the per-day impact of a temporary internet shutdown would be on average $23.6 million per 10 million population.

A deliberate attack on routing protocols could cause widespread financial and operational chaos. In today’s interconnected world, where digital infrastructure underpins economic stability, the ability to manipulate undersea cable traffic represents a subtle but powerful geopolitical weapon.

Addressing this threat goes beyond simply building more cables. It requires rethinking how routing protocols are governed. Transparent global standards must ensure no single country or company can dominate these systems. Routine independent audits should be conducted to detect anomalies that may signal interference. Efforts like the European Union’s Global Gateway initiative and Japan’s Digital Partnership Fund must focus on creating alternative routes to reduce reliance on Chinese-controlled nodes.

This issue highlights a new reality in global politics: control over data flows is becoming a key form of power. While most attention has been on building physical infrastructure, the quiet manipulation of routing protocols marks an equally profound shift in global influence. To protect the integrity of the internet, the world must act decisively at both technical and governance levels.

Fiber-Optic Cable Repair Networks

China’s disproportionate control over fiber-optic cable repair networks reveals potential vectors for intelligence dominance, coercive leverage, and disruption of digital sovereignty. Globally, an estimated 60 dedicated cable repair ships service the planet’s 1.5 million kilometers of submarine cables. China controls a substantial percentage of the fleet, including ships operated by state-affiliated enterprises like Shanghai Salvage Company and China Communications Construction Group. In contrast, the United States and its allies maintain a small patchwork fleet, mostly concentrated in the North Atlantic and lacking coverage in the Indo-Pacific, where over 50 percent of global internet traffic routes through key subsea cables.

China’s fleet is heavily concentrated in the South and East China Seas, regions critical to global connectivity due to chokepoints like the Singapore Strait and the Luzon Strait. With maritime exclusivity bolstered by China’s claims in disputed waters, its repair ships have nearly unrestricted access to monitor, repair, or potentially tamper with cables under the guise of routine maintenance.

Repair missions involve exposing critical cable infrastructure, including repeaters, amplifiers, and branch units—hardware that boosts signal strength over long distances but also represents points of vulnerability. Chinese vessels are equipped with advanced robotic submersibles and precision cutting-and-splicing technologies, designed for repairs but capable of installing signal interception devices. Such tools could include optical fiber taps capable of harvesting unencrypted metadata or capturing latency patterns to infer sensitive traffic flow.

China’s advancements in photonics and quantum communication technologies underscore its capacity to exploit these vulnerabilities. The Chinese Academy of Sciences has reported significant breakthroughs in quantum key distribution (QKD) systems, raising the possibility of developing quantum-based methods to crack encrypted data intercepted during repairs. Integration of AI-driven data sorting tools could automate the extraction and classification of intercepted information, rendering bulk data acquisition during repairs a strategic advantage.

The high seas, where many repair operations occur, are governed by fragmented international frameworks like the United Nations Convention on the Law of the Sea (UNCLOS), which inadequately regulate activities involving critical infrastructure. The International Cable Protection Committee (ICPC) provides voluntary guidelines for repair operations, but enforcement mechanisms are weak, leaving the system vulnerable to exploitation by state actors.

Repair missions are often classified as “emergency operations,” requiring expedited approvals that bypass detailed oversight. In one case, a cable break in the South China Sea in 2021 prompted Chinese repair ships to operate without transparency for over three weeks, raising concerns about potential covert activities. These incidents are rarely reported, as they fall outside the jurisdiction of most maritime monitoring bodies.

The lack of countermeasures by the United States and its allies amplifies the risks posed by China’s dominance. The U.S. Navy operates no specialized repair ships, relying on private operators like Global Marine Group, whose fleet is aging and ill-equipped for operations in contested waters. This contrasts with China’s state-backed model, integrating its repair fleet into broader maritime networks, providing dual-use functionality for civilian and military objectives.

The financial model of undersea cable operations further constrains Western responses. Submarine cables are predominantly privately owned, with firms like Google, Meta, and Amazon investing heavily in infrastructure but lacking incentives to prioritize geopolitical considerations. This privatization leaves strategic gaps in surveillance and monitoring, as governments must negotiate access to privately controlled repair missions.

To mitigate China’s strategic advantage, a multipronged response is essential. The United States and its allies must develop state-owned or state-subsidized repair fleets to operate in contested regions like the South China Sea and Indian Ocean. Enhanced maritime surveillance systems, such as underwater drones and sonar-based monitoring arrays, should be deployed to track repair ship movements in real time.

Revising international frameworks by expanding ICPC mandates to include mandatory reporting of repair operations could curb opacity. Collaboration with regional partners, particularly nations in the Quad (Australia, India, Japan, and the United States), could bolster collective maritime domain awareness and create redundancies in cable repair capabilities.

Maritime Data Through Automated Vessel Tracking

China’s exploitation of automated vessel tracking systems exemplifies a sophisticated component of its global digital strategy. At the heart of this initiative lies the Automatic Identification System (AIS), a maritime safety technology mandated by the International Maritime Organization (IMO) for vessels exceeding 300 gross tons engaged in international trade. While originally intended to improve navigational safety by broadcasting vessel identities, locations, courses, and cargo details, AIS has been effectively repurposed by Beijing into a dual-use asset that supports both economic intelligence gathering and military surveillance.

Chinese firms, including the BeiDou Navigation Satellite System and Alibaba Cloud, have developed advanced platforms that aggregate AIS transmissions from shipping lanes worldwide. These platforms integrate AIS data with artificial intelligence-driven predictive analytics, enabling Beijing to monitor and analyze global maritime chokepoints such as the Strait of Malacca, the Panama Canal, and the Suez Canal—key arteries of international commerce. By doing so, China gains critical insights into global shipping patterns, strategic trade routes, and supply chain dynamics. As of 2023, the global merchant fleet comprised around 60,000 ships.

During the 2021 Suez Canal blockage, Chinese logistics firms, leveraging real-time AIS data, rapidly identified alternative routes through the Arctic and along the Indian Ocean, allowing Chinese exporters to reroute goods while Western competitors faced delays. Similarly, in the Strait of Malacca, a waterway facilitating the transit of over 16 million barrels of oil daily and 40 percent of global trade, Chinese analysts have used AIS data to optimize resource flow, preempt congestion, and study vulnerabilities in energy supply routes.

AIS data plays a pivotal role in China’s military strategy, especially in the Indo-Pacific. By combining AIS information with satellite imagery and data from undersea acoustic arrays, China has established a surveillance network capable of tracking naval deployments with precision. AIS data has been used to monitor patrol patterns of the U.S. Navy’s Seventh Fleet, revealing that over a third of its South China Sea operations in 2022 followed predictable routes. This surveillance allows the People’s Liberation Army Navy (PLAN) to anticipate U.S. Freedom of Navigation Operations (FONOPs) and position its assets accordingly.

China’s manipulation of AIS extends to conflict simulations and asymmetric warfare. During military exercises near Taiwan in 2023, Chinese forces reportedly deployed unmanned surface vessels programmed to mimic civilian AIS signals, complicating the identification of hostile assets.

Through its Digital Silk Road initiative, Beijing has exported various forms of maritime technologies that incorporate Automatic Identification System (AIS) capabilities. China often provides financial incentives to promote the adoption of its technologies abroad, which may enhance its access to regional maritime data. This asymmetry grants China an informational advantage and risks reshaping maritime transparency norms in its favor.

Rare Subsea Mapping Data

China’s increasing investment in subsea mapping has positioned it as a significant player in oceanographic intelligence, impacting scientific, commercial, and military domains. China has been actively mapping its claimed maritime territories using state-funded research vessels and autonomous systems. These efforts contribute to international initiatives like the Nippon Foundation-GEBCO Seabed 2030 project, which aims to map the entire global seabed by 2030 and had mapped approximately 23.4 percent as of June 2022 with international contributions. China’s activities extend to strategic regions in the Indo-Pacific, the Arctic, and the Indian Ocean, raising concerns over the dual-use potential of its data collection.

Subsea mapping data is critical for submarine cable routing, undersea infrastructure development, and naval operations. China’s repository of high-resolution bathymetric maps—including surveys of key chokepoints like the Strait of Malacca and the Bashi Channel—provides a tactical edge. These chokepoints are vital for global trade and serve as strategic naval passages for power projection and anti-access/area-denial operations. The People’s Liberation Army Navy uses seabed data to optimize the placement of undersea sensor arrays, critical for its “Great Underwater Wall” initiative, integrating hydroacoustic monitoring to detect foreign submarines.

China’s advancements in autonomous underwater vehicles (AUVs) enhance its capabilities. In 2021, the Hailong III and Qianlong II AUVs were deployed for deep-sea mapping missions in the South China Sea, gathering data at depths over 6,000 meters. These AUVs have multi-beam sonar systems achieving sub-meter resolution, surpassing commercial standards. Their ability to operate autonomously over long durations allows China to map intricate undersea topographies critical for resource exploration and undersea warfare.

China has used seabed mapping as a diplomatic tool to extend influence over smaller nations. Through its Maritime Silk Road Initiative, Beijing has signed agreements with over 20 countries, granting Chinese research vessels access to Exclusive Economic Zones (EEZs). Between 2015 and 2022, Chinese expeditions in Pacific Island nations’ EEZs often involved dual-use mapping activities.

In 2019, the Chinese survey vessel Haiyang Dizhi 8 conducted seismic surveys near the Vanguard Bank within Vietnam’s Exclusive Economic Zone (EEZ), collecting bathymetric data that aligns with key undersea routes potentially useful for submarine operations. This incursion led to a tense standoff with Vietnam, drawing international criticism over China’s assertive actions and raising concerns about the dual-use potential of the data collected. Similarly, in 2018, China’s proposed involvement in undersea cable projects connecting Papua New Guinea and the Solomon Islands through Huawei Marine raised significant security concerns. Fearing risks to the security of undersea communication cables and potential espionage, Australia intervened by funding and undertaking the projects themselves, highlighting apprehensions about granting Chinese entities access to critical seafloor data in the region.

China’s seabed mapping strategy has significant military implications, particularly in the South China Sea. In this region, where China has constructed artificial islands such as Fiery Cross Reef, Subi Reef, and Mischief Reef, high-resolution seabed data enables precise deployment of missile systems, naval patrols, and underwater drones. Detailed seabed mapping supports the construction and fortification of these islands, allowing for the installation of surface-to-air missiles, anti-ship cruise missiles, and the operation of military airstrips. Additionally, China’s deployment of unmanned underwater vehicles like the Sea Wing (Haiyi) gliders enhances their ability to collect oceanographic data crucial for submarine navigation and anti-submarine warfare. These activities have raised concerns among neighboring countries and the international community about the dual-use potential of China’s maritime endeavors and their impact on regional security.

By controlling seabed mapping data, China influences submarine cable networks, which carry 95 percent of global internet traffic and $10 trillion in daily financial transactions. China’s involvement in projects like the South Pacific Cable Project through state-owned China Mobile led to concerns over data interception capabilities. Its presence in Arctic seabed mapping, facilitated by icebreaker vessels like Xuelong 2, underscores ambitions to secure alternative maritime routes and resources under the guise of scientific research.

China’s approach to subsea mapping data has raised concerns about transparency and shared access in the global community. While international initiatives like the Seabed 2030 Project encourage open sharing of ocean floor data to advance scientific research and environmental understanding, China has been criticized for not fully sharing the extensive seabed data it collects. For example, much of the data gathered by Chinese vessels in international waters is not readily available in global databases like those managed by the International Hydrographic Organization (IHO) or the General Bathymetric Chart of the Oceans (GEBCO). This selective sharing limits other nations’ ability to leverage valuable information and contrasts with global norms promoting cooperation and transparency in oceanographic research.

Tyler Durden
Thu, 12/05/2024 – 23:25