Biaix House / GRATSave this picture! Simone MarcolinHousesBarcelona, SpainArchitects: GRATYearCompletion year of this architecture project Year: 2024 PhotographsPhotographs:Simone MarcolinMore SpecsLess SpecsSave this picture!Save this picture!Text description provided by the architects. Casa Biaix is located along the Aubareda Path, a trail that winds through fields of wheat, rapeseed, and Mediterranean forests. It sits atop a small hill, amidst a pine forest and an olive grove, just a few meters away from one of the oldest oaks in Barcelona.Save this picture!Save this picture!Save this picture!The design preserves the historical exterior shape of the farmhouse while introducing a new formal and material logic inside. It frames views of the regenerating landscape, partially transformed by a recent wildfire. An austere materiality of lime mortar is applied throughout the project, reintroducing a material that was traditionally used in rural houses. The local limestone, characteristic of the interior of Catalonia, defines the color palette of the design.Save this picture!Save this picture!Save this picture!The daytime program on the ground floor is organized to enhance visual connections with a landscape that changes throughout the seasons. On the upper floor, the rooms are arranged along a diagonal hallway that shapes the distinctive interior geometry. That oblique form is used to optimize service spaces and accommodate an additional room for the family.Save this picture!Save this picture!The adjacent annex becomes a new living space, connected to the main house by a new spiral staircase. The rear facade opens up, making the house completely accessible and exposed to the elements.Save this picture!Project gallerySee allShow lessAbout this officeGRATOfficePublished on February 05, 2025Cite: "Biaix House / GRAT" [Casa Biaix / GRAT] 05 Feb 2025. ArchDaily. Accessed . <https://www.archdaily.com/1026360/biaix-house-grat&gt ISSN 0719-8884Save!ArchDaily?You've started following your first account!Did you know?You'll now receive updates based on what you follow! Personalize your stream and start following your favorite authors, offices and users.Go to my stream

Apple MacBook Pro users often find it difficult to switch to Windows laptops due to the deep integration of the Apple ecosystem, which offers seamless connectivity between devices like iPhones, iPads, and Apple Watches. Features such as AirDrop, Handoff, iMessage, and iCloud synchronization create a frictionless workflow thats difficult to replicate on Windows.Additionally, macOS is known for its stability, optimized performance, and clean user experience, making the transition to Windows where driver issues, inconsistent UI, and bloatware can be concerns feel clumsy. Many MacBook Pro users are also heavily invested in Apple-exclusive software like Final Cut Pro, Logic Pro, and macOS-specific productivity tools.Hardware aesthetics and longevity also play a role; Apples unibody aluminum design, high-resolution Retina displays, and industry-leading battery life set a high standard. Ultimately, switching from a MacBook to Windows isnt just about a laptop its about leaving an entire ecosystem that enhances convenience and productivity. But lets focus on the laptop itself.Lenovos latest ThinkPad X1 Carbon Aura Edition is giving me pause. The Aura Edition moniker signifies advanced AI features, including intelligent modes for privacy, productivity, distraction management, and premium customer support.It also highlights a powerful neural processing unit (NPU) within Intels latest Lunar Lake processor, delivering a more innovative and personalized user experience than standard Lenovo models.I dont say this lightly. In the ever-evolving landscape of premium laptops, the Lenovo ThinkPad X1 Carbon Gen 13 and Apples 14-inch MacBook Pro with the M4 chip stand out as top contenders. Both devices cater to professionals seeking performance, portability, and advanced features.However, the ThinkPad X1 Carbon Gen 13, especially with Intels Lunar Lake processor marketed under the Core Ultra series, offers distinct advantages, making it a compelling choice for discerning users.The Lenovo ThinkPad X1 Carbon Aura Edition features a lightweight design and vibrant display, making it a strong alternative to the MacBook Pro. (Photo by Author)ThinkPad vs. MacBook: Design and Build QualityThe ThinkPad X1 Carbon Gen 13 Aura Edition continues Lenovos legacy of combining durability with elegance. Weighing a mere 2.17 pounds, it is notably lighter than many competitors, including the MacBook Pro. The X1s carbon fiber and magnesium alloy construction ensure sturdiness without compromising on weight, making it exceptionally portable for professionals on the move.In contrast, the 14-inch MacBook Pro with the M4 chip maintains Apples signature aluminum unibody design. While aesthetically pleasing and sturdy, it is considerably heavier than the ThinkPad at 3.3 pounds, which might be a consideration for users prioritizing ultra-lightweight devices and that additional 1.1 pounds feels much heavier when youve been on the road for a few weeks.Performance ShowdownAt the heart of the ThinkPad X1 Carbon Gen 13 is Intels Lunar Lake processor, designed for efficiency and performance. It enables smooth multitasking and handles demanding applications with ease. The ThinkPad X1 Carbon Gen 13 features integrated Intel Arc Graphics 140V, providing solid performance for professional workloads and light creative tasks.The MacBook Pros M4 chip is built on a 3-nanometer process, delivering significant performance improvements over its predecessors. It excels in single-core tasks and provides solid performance for creative applications.However, for users who rely on software optimized for Windows or require specific enterprise solutions, the ThinkPads architecture may offer better compatibility and flexibility especially since its x86-based. This reduces the risk of software compatibility issues, a potential concern with Arm-based laptops like those featuring the Qualcomm Snapdragon X Elite chip.ThinkPad OLED vs. MacBook XDR: Display QualityThe ThinkPad X1 Carbon Gen 13 features a stunning 14-inch 2.8K OLED display that delivers deep blacks, high contrast, and vivid colors. With HDR 500 certification and a 120Hz refresh rate, it offers smooth visuals and excellent clarity for both work and media consumption. The anti-glare, anti-reflection, and anti-smudge coatings enhance visibility, making it well-suited for varied work environments. The 14-inch MacBook Pro features Apples Liquid Retina XDR display, known for its extreme brightness (1,000 nits sustained, 1,600 nits peak in HDR) and exceptional color accuracy. The M4 series improves glare reduction while maintaining high contrast and true-to-life colors, making it an excellent choice for creative professionals working in different lighting conditions.Both displays offer outstanding image quality, but the ThinkPads OLED panel excels in contrast, while the MacBook Pros XDR display leads in sustained brightness and color accuracy. Users prioritizing high peak brightness for HDR content may prefer Apples offering, while those seeking deep blacks and OLED vibrancy may lean toward the ThinkPad.Battery Life and ChargingBattery performance is crucial for professionals on the go. The ThinkPad X1 Carbon Gen 13 stands out with respectable battery life, lasting through a typical workday under moderate use. While Apples MacBook Pro (and Air) with Apple Silicon set a high bar for battery endurance, Lenovos latest optimizations bring the X1 closer in efficiency, making it a more competitive option.While it may not surpass the MacBook Pros impressive battery life, the X1 compensates with rapid charging, minimizing downtime. The MacBook Pro with the M4 chip boasts extended battery life, often exceeding 16 hours for standard tasks. This longevity is beneficial for users who require prolonged unplugged use.ThinkPad vs. MacBook Keyboard ExperienceLenovos ThinkPad series is renowned for its exceptional keyboards, and the X1 Carbon Gen 13 is no exception. It offers deep travel and tactile feedback, ensuring a comfortable typing experience. Additionally, the iconic TrackPoint provides an alternative navigation method, which many professionals appreciate.One huge plus for the ThinkPad is that it uses Lenovos TrackPoint for precise cursor control without lifting your hands off the keyboard, speeding up workflow for power users. Its small footprint doesnt take up valuable trackpad space, making it perfect for confined spaces or on-the-go productivity. Plus, the tactile feel and responsiveness make it a favorite for those who prefer a more tactile navigation experience over a traditional trackpad.Apples MacBook Pro features the Magic Keyboard, which has been well-received for its improved key stability and responsiveness. While it provides a satisfactory typing experience, users who prefer deeper key travel might favor the ThinkPads keyboard.Connectivity and PortsThe ThinkPad X1 Carbon Gen 13 excels in connectivity options. It includes multiple USB-A and USB-C ports, HDMI, and a headphone jack, allowing users to connect various peripherals without needing adapters.While the MacBook Pro offers Thunderbolt ports, HDMI, and a headphone jack, it may require users to utilize adapters or dongles for USB-A and other connections, which can be less convenient for those with diverse peripheral needs.The ThinkPad X1 Carbons left side includes one USB-A and two USB-C (Thunderbolt 4) ports, offering greater connectivity than most ultrabooks. (Photo by Author)The ThinkPad X1 Carbons right side features HDMI, a headphone/mic combo jack, and another USB-A port, ensuring broad compatibility with peripherals. (Photo by Author)Windows vs. macOS: Software and CompatibilityThe ThinkPad X1 Carbon Gen 13 runs on Windows, providing access to various software applications, particularly those tailored for enterprise environments. This compatibility is essential for professionals who rely on specific Windows-based tools.The MacBook Pro operates on macOS, which is known for its stability and seamless integration with other Apple products. While it supports a wide range of applications, some specialized software may not be available or optimized for macOS, which could be a limitation for specific professional workflows.Price and Business ValueLenovos ThinkPad X1 Carbon Gen 13, starting at $2,267, is positioned as a premium device, reflecting its build quality, performance, and features. It offers substantial value for professionals seeking a lightweight, durable, and versatile laptop.The MacBook Pro with the M4 chip, which costs $2,199 when configured with similar storage and memory, also falls into the premium price bracket. While it delivers strong performance and battery life, users should consider whether its ecosystem and software compatibility align with their needs.Closing ThoughtsThe Lenovo ThinkPad X1 Carbon Gen 13 and the 14-inch Apple MacBook Pro with the M4 chip are formidable laptops catering to professionals.However, the ThinkPad X1 Carbon Gen 13 distinguishes itself with its ultra-lightweight design, superior keyboard and TrackPoint experience, diverse connectivity options, and the flexibility of the Windows operating system. These attributes make it a compelling choice for users seeking a reliable and versatile tool for their professional endeavors. While pricing is comparable, the ThinkPad X1 Carbon offers advantages such as frequent discounts from online retailers, customizable configurations, and enterprise-level security features tailored for business users. Additionally, its broader selection of ports and upgrade options reduces the need for costly dongles or proprietary accessories, further enhancing its appeal for professionals who prioritize versatility and practicality.The ThinkPad X1 Carbon Gen 13 is a powerhouse for professionals who need a lightweight, well-built laptop with top-tier input devices and strong connectivity. Lenovos focus on enterprise-friendly features, customization, and practicality makes it an excellent choice for those prioritizing versatility over ecosystem lock-in.For more details on the Lenovo ThinkPad X1 Carbon Aura Edition, visit the Lenovo store.

Large language models (LLMs) are increasingly being deployed on edge deviceshardware that processes data locally near the data source, such as smartphones, laptops, and robots. Running LLMs on these devices supports advanced AI and real-time services, but their massive size, with hundreds of millions of parameters, requires significant memory and computational power, limiting widespread adoption. Low-bit quantization, a technique that compresses models and reduces memory demands, offers a solution by enabling more efficient operation.Recent advances in low-bit quantization have made mixed-precision matrix multiplication (mpGEMM) viable for LLMs. This deep learning technique allows data of the same or different formats to be multiplied, such as int8*int1, int8*int2, or FP16*int4. By combining a variety of precision levels, mpGEMM strikes a balance among speed, memory efficiency, and computational accuracy.However, most hardware supports only symmetric computationsoperations on data of similar formatscreating challenges for mixed-precision calculations during General Matrix Multiplication (GEMM), a critical operation for LLMs. Overcoming these hardware limitations is essential to fully benefit from mpGEMM and support asymmetrical computations.To unlock the potential of low-bit quantization on resource-constrained edge devices, hardware must natively support mpGEMM. To address this, we developed the following three approaches for computing kernels and hardware architectures:Ladder data type compiler: Supports various low-precision data types by converting unsupported types into hardware-compatible ones without data loss, while also generating high-performance conversion code.T-MAC mpGEMM library: Implements GEMM using a lookup table (LUT) approach, eliminating multiplications to significantly reduce computational overhead. Optimized for diverse CPUs, T-MAC delivers several times the speed of other libraries.LUT Tensor Core hardware architecture: Introduces a cutting-edge design for next-generation AI hardware, tailored for low-bit quantization and mixed-precision computations.The following sections describe these techniques in detail.Ladder: Bridging the gap between custom data and hardware limitsCutting-edge hardware accelerators, such as GPUs, TPUs, and specialized chips, are designed to speed up computationally intensive tasks like deep learning by efficiently handling large-scale operations. These accelerators now integrate lower-bit computing units, such as FP32, FP16, and even FP8, into their architectures.However, constraints in chip area and hardware costs limit the availability of these units for standard data types. For instance, the NVIDIA V100 Tensor Core GPU supports only FP16, while the A100 supports int2, int4, and int8 but not newer formats like FP8 or OCP-MXFP. Additionally, the rapid development of LLMs often outpaces hardware upgrades, leaving many new data formats unsupported and complicating deployment.Additionally, while hardware accelerators may lack direct support for custom data types, their memory systems can convert these types into fixed-width data blocks that store any data format. For instance, NF4 tensors can be converted into FP16 or FP32 for floating-point operations.Building on these insights, we developed the Ladder data type compiler, a method to separate data storage from computation, enabling broader support for custom data types. It bridges the gap between emerging custom data formats with the precision types supported by current hardware.Ladder offers a flexible system for converting between algorithm-specific and hardware-supported data types without data loss. For low-bit applications, it optimizes performance by translating low-bit data into the most efficient formats for the hardware being used. As shown in Figure 1, this includes mapping low-bit computations to supported instructions and efficiently managing data storage across the memory hierarchy.Figure 1: The Ladder architectureEvaluating LadderEvaluations of Ladder on NVIDIA and AMD GPUs show that it outperforms existing deep neural network (DNN) compilers for natively supported data types. It also handles custom data types not supported by GPUs, achieving speedups of up to 14.6 times.As the first system to support custom low-precision data types for running DNNs on modern hardware accelerators, Ladder provides researchers with flexibility in optimizing data types. It also enables hardware developers to support a wider range of data types without requiring hardware modifications.T-MAC: Table-lookup for mpGEMM without multiplicationDeploying low-bit quantized LLMs on edge devices often requires dequantizing models to ensure hardware compatibility. However, this approach has two major drawbacks:Performance: Dequantization overhead can result in poor performance, negating the benefits of low-bit quantization.Development: Developers must redesign data layouts and kernels for different mixed precisions.To address these challenges, we introduce T-MAC, a novel LUT-based method that enables mpGEMM without dequantization or multiplication.T-MAC replaces traditional multiplication operations with bit-wise table lookups, offering a unified and scalable solution for mpGEMM. It incorporates techniques to reduce the size of tables and store them directly on the chip, minimizing the overhead of accessing data from memory. By eliminating dequantization and lowering computational costs, T-MAC enables efficient inference of low-bit LLMs on resource-constrained edge devices. Figure 2 illustrates T-MACs architecture.Figure 2. Overview of the T-MAC systemEvaluating T-MACPerformance evaluations of T-MAC on low-bit models demonstrated substantial benefits in efficiency and speed. On the Surface Laptop 7 with the Qualcomm Snapdragon X Elite chipset, T-MAC achieved:48 tokens per second for the 3B BitNet-b1.58 model30 tokens per second for the 2-bit 7B Llama model20 tokens per second for the 4-bit 7B Llama modelThese speeds far exceed average human reading rates, outperforming llama.cpp by 45 times and doubling the speed of a dedicated NPU accelerator.Even on lower-end devices like the Raspberry Pi 5, T-MAC made it possible for the 3B BitNet-b1.58 model to generate 11 tokens per second. It also proved highly power-efficient, matching llama.cpps generation rate while using only 1/4 to 1/6 of the CPU cores.These results establish T-MAC as a practical solution for deploying LLMs on edge devices with standard CPUs, without relying on GPUs or NPUs. T-MAC allows LLMs to run efficiently on resource-constrained devices, expanding their applicability across a wider range of scenarios.LUT Tensor Core: Driving hardware for mpGEMMWhile T-MAC and Ladder optimize mpGEMM on existing CPU and GPU architectures, improving computational efficiency, they cannot match the performance of dedicated hardware accelerators with built-in LUT support. Achieving significant improvements in performance, power, and area (PPA) requires overcoming four key challenges:Table precompute and storage: Precomputing and storing LUTs add overhead, increasing area usage, latency, and storage requirements, which can reduce overall efficiency gains.Bit-width flexibility: Hardware must support various precision levels, such as int4/2/1 for weights and FP16/8 or int8 for activations, along with their combinations. This flexibility is crucial for accommodating diverse model architectures and use cases.LUT tiling shape: Inefficient tiling shapes can raise storage costs and limit reuse opportunities, adversely affecting performance and efficiency.Instruction and compilation: LUT-based mpGEMM requires a new instruction set. Existing compilation stacks, designed for standard GEMM hardware, may not optimally map and schedule these instructions, complicating integration with LLM inference software.In response, we developed LUT Tensor Core, a software-hardware codesign for low-bit LLM inference. To address precomputation overhead in conventional LUT-based methods, we introduce techniques like software-based DFG transformation, operator fusion, and table symmetrization to optimize table precomputation and storage. Additionally, we propose a hardware design with an elongated tiling shape to support table reuse and a bit-serial design to handle various precision combinations in mpGEMM.To integrate with existing GPU microarchitectures and software stacks, we extended the MMA instruction set, added new LMMA instructions, and developed a cuBLAS-like software stack for easy integration into existing DNN frameworks. We also created a compiler for end-to-end execution planning on GPUs with LUT Tensor Core. This design and workflow, illustrated in Figure 3, enabled the quick and seamless adoption of LUT Tensor Core.Figure 3. The LUT Tensor Core workflowEvaluating LUT Tensor CoreTesting LUT Tensor Core on low-bit LLMs, such as BitNet and Llama, showed significant performance gains, achieving 6.93 times the inference speed while using just 38.3% of the area of a traditional Tensor Core. With nearly identical model accuracy, this results in a 20.9-fold increase in computational density and an 11.2-fold boost in energy efficiency. As AI models grow in scale and complexity, LUT Tensor Core enables low-bit LLMs to be applied in new and diverse scenarios.We believe the LUT technique could drive a paradigm shift in AI model inference. Traditional methods rely on multiplication and accumulation operations, whereas LUT implementations provide higher transistor density, greater throughput per chip area, lower energy costs, and better scalability. As large models adopt low-bit quantization, the LUT method could become the standard for system and hardware design, advancing the next generation of AI hardware innovation.Unlocking new possibilities for embodied AILow-bit quantization improves the efficiency of running large models on edge devices while also enabling model scaling by reducing the bits used to represent each parameter. This scaling enhances model capabilities, generality, and expressiveness, as shown by the BitNet model, which starts with a low-bit configuration and expands.Technologies like T-MAC, Ladder, and LUT Tensor Core provide solutions for running low-bit quantized LLMs, supporting efficient operation across edge devices and encouraging researchers to design and optimize LLMs using low-bit quantization. By reducing memory and computational demands, low-bit LLMs could power embodied AI systems, such as robots, enabling dynamic perception and real-time environmental interaction.T-MAC (opens in new tab) and Ladder (opens in new tab) are open source and available on GitHub. We invite you to test and explore these innovations in AI technology with Microsoft Research.Spotlight: blog postGraphRAG auto-tuning provides rapid adaptation to new domainsGraphRAG uses LLM-generated knowledge graphs to substantially improve complex Q&A over retrieval-augmented generation (RAG). Discover automatic tuning of GraphRAG for new datasets, making it more accurate and relevant.Read moreOpens in a new tab Opens in a new tab

Were only a few days into Get Fit February, and Im still finding my rhythm. I did get out on the bike over the weekend, thoughjust me, two water bottles, a long playlist, and the Bose Ultra Open Earbuds. Im not a big fan of being alone with my thoughts, but Im also not a big fan of being cut off from the sounds of cars and kids that might make me crash. I am a fan of saving you money, though, so I recommend you snap up a pair of these refurbished earbuds for $199 while supplies lastthats $100 off the retail price and $30 less than the best price Ive seen in the past. And act now because there are limited amounts of these inspected and tested earbuds available. Bose Ultra Open Earbuds $199 (was $299) See ItWhat I like about the Ultra Open Earbuds is, first and foremost, the form. The unique design grips cuff-like around the helix (the edge) of the ear; then, the speaker end settles right above, instead of in, the ear canal. This lets it beam the beat right into your ear while letting you stay aware of your environment. Thats practical when running or biking, so you can hear music or an instructors directions/motivation but also hear cars or other cyclists coming behind you. Thats one reason we named them one of the best headphones for working out. The other reason is that there is ample bass response and a crisp Immersive Audio mode, so sound appears to surround you as you can keep the pace while keeping safe. And if all that sounds good to you, act fast; these certified refurbished Ultra Open Earbuds are available in black and white smoke until tomorrow or until theyre sold out, whichever comes first.More Bose flash dealsBose QuietComfort Ultra Earbuds Refurbished $259 (was $299)Bose Smart Soundbar 900 Refurbished $599 (was $899)Bose QuietComfort Headphones Refurbished $299 (was $349)Bose Portable Smart Speaker Refurbished $289 (was $399)The post The earbuds that help me keep my heart rate up are down to their lowest price appeared first on Popular Science.

Cat lovers in Scotland can breathe a collective sigh of relief. The Scottish government will not be banning felines in the country home to an estimated 840,000 pet cats. First Minister John Swinney issued a statement after a report issued on January 27 by the Scottish Animal Welfare Commission said that cats were a threat to the countrys wildlife. The commission suggested measures to reduce their damageincluding restrictions on ownershipwere interpreted by some as an all out-ban on cats and rumors quickly spread online.Theres a report being produced by an external organization, which has come into the government for consideration, Swinney said via radio. Let me just clear this up today. The governments not going to be banning cats or restricting cats. We have no intention of doing so and well not be doing it.Outdoor cats can be destructive to the environment by killing native bird species and smaller mammals like squirrels and mice and can be considered an invasive species. They can also spread diseases including rabies and parasitic infections. Cats are the primary hosts of the parasite Toxoplasma gondii. Most become infected by accidentally ingesting the parasite through contact with cat feces. The majority of people who become infected with Toxoplasma gondii do not experience symptoms, but severe infection can cause damage to the brain, eyes, and other organs.In 2022, Poland officially listed domestic cats as an invasive alien species, due to their environmental impacts. Cats are believed to have caused the most destruction in Australia than any other continent, where they are estimated to have driven 27 native animals to extinction since colonization began in 1788.However, some countries including Italy, Austria, and Japan have laws on the books that protect cats, particularly feral cat colonies. The United States is home to an estimated 76.5 million cats, with several households owning more than one. Rules regarding pet ownership, vaccination, and spaying/neutering vary by state, with Rhode Island being the only state that requires cat licensing.In their report, the Scottish Animal Welfare Commission said cats annually kill at least 27 million birds and other animals throughout the United Kingdom. It advised that the government consider keeping domestic cats on leashes or indoors as one way to protect endangered species including Scottish wildcats. Some of the various cat containment measures were listed as restrictions on introducing cats to households in vulnerable areas.[ Related: Why do cats make biscuits? ]Cats Protection, the largest cat charity in the UK responded to the containment measures stressing simple solutions like keeping cats indoors at dawn and dusk to help balance the needs of both wild animals and domestic cats.Scotland is a nation of cat lovers, Alice Palombo, Cats Protections advocacy and government relations officer for Scotland, told the Associated Press. Cats are great pets for all sorts of reasons, whether its providing companionship for elderly people or those living alone, comfort for people with health conditions or helping children learn important lessons in caring for others, she said. We believe everyone who is able to care for a cat should be able to enjoy these benefits.According to the College of Natural Sciences at the University of Massachusetts Amherst, cat owners can lessen the impacts an outdoor cat can have on the environment, by training their felines to walk on a leash. They also suggest an enclosed catio spaced outside with plenty of toys and structures to give the cats exercise, play, stimulation, fresh air, and sun.The post Relax, Whiskers. Scotland is not banning cats appeared first on Popular Science.

Only four ships have ever visited the place where Thwaites Glacier pours off the coast of West Antarctica. This swath of ocean resembles a rugged, white desert a plain of wind-sculpted ice dotted with sheer-sided mesas that tower seven to 10 stories above the surrounding terrain.Those mesas are icebergs larger than aircraft carriers. They break from the glacier itself and from the rest of the West Antarctic Ice Sheet, a dome of ice as large as Mexico slowly oozing seaward like a heap of frozen custard.As the winds and ocean currents push the icebergs around, they plow through the meter-thick sea ice that covers the water, as if it were the fragile skin that forms on a cooling bowl of tomato soup.In the coming century, a pivotal drama between humans and nature could play out here. In a quest to slow down sea level rise, a few researchers are sketching out massive engineering and construction projects that could block ocean currents, alter the flow of some of the worlds largest glaciers and potentially delay or prevent a major collapse.Even compared to the Great Wall of China, some of the proposed projects would be just enormous, says Christian Rodehacke, a glaciologist at the Alfred Wegener Institute for Polar and Marine Research in Bremerhaven, Germany. With structures potentially as tall as the Empire State Building, such a project could be the largest effort ever undertaken by humans to modify Earth.The West Antarctic Ice Sheet is stabilized by undersea mountains that rise to form a jagged dike beneath its outer edges, but low spots in the dike provide gaps where gigantic corridors of fast-moving ice slide into the ocean. Thwaites, reaching more than 400 kilometers upstream into the heart of West Antarctica, is the most vulnerable of these glaciers, and the widest glacier in the world. Its coastal outlet is 130 kilometers across and dips as far as 1.2 kilometers below sea level exposing it to warm, dense, salty ocean currents that flow like rivers along the seafloor.Thwaites holds 480,000 cubic kilometers of ice. Its losing about 80 cubic kilometers per year a sixfold increase since the 1990s and its rate of loss is expected to increase more. As the glacier thins, it lifts off the seafloor, gradually losing its connection with the dike below. Because of this, the speed of the glaciers western branch has accelerated by more than 70 percent since 1973, reaching 4 kilometers per year. It is losing volume ever more quickly as it melts and sheds icebergs.The events unfolding here present humankind with a range of possible futures: At the optimistic end, the glacier keeps its death grip on the protective dike long enough for us to dramatically curb greenhouse gas emissions and remove planet-warming carbon dioxide from the atmosphere. In this case, Antarctica contributes only about 10 centimeters of sea level rise by 2100 and roughly a meter by 2300, with the rate of rise gradually tapering off over the next few centuries.The front edge of Thwaites Glacier is one of the most inaccessible places on Earth.D. FoxAt the other extreme, Thwaites thins enough to lose its grip, further accelerating and fragmenting into icebergs. This could unleash an irreversible collapse across the West Antarctic Ice Sheet, with the continent contributing 20 to 50 centimeters of sea level rise by 2100, and up to four to seven meters by 2300, drowning population centers in California, Florida, Louisiana, South Carolina, the Netherlands, Pakistan, Bangladesh, Vietnam, Thailand, the Philippines and many other places.In all of these scenarios, many more centimeters of sea level rise will come from other sources, like thermal expansion of oceans and melting Arctic glaciers, further worsening the situation.There is no consensus on when Thwaites will start to collapse or what temperatures will trigger it. Some scientists believe that Thwaites has already entered the early stages of its demise. Others think it will cross that invisible threshold in the coming decades. This is prompting some researchers to step beyond the role of simply documenting the demise of glaciers and make audacious plans to intervene, to save a handful of crucial glaciers that could set off runaway sea level rise.Sponsor MessageNormally staid and cautious scientists are now willing to speak aloud the things that theyve been quietly thinking for years. A prominent science-fiction novelist has launched the idea into public consciousness. And tech entrepreneurs have volunteered to find ways to fund the research, which government agencies consider too controversial to touch.The proposed ideas would be expensive, logistically challenging and legally fraught. The mere idea of engineering glaciers has provoked heated disagreement among well- respected scientists who have worked together for decades, reflecting deep philosophical and political divides over how society should respond to climate change.Not talking about this because it makes people uncomfortable feels like a dereliction of duty for scientists who are taxpayer-funded, says Slawek Tulaczyk, a glaciologist at the University of California, Santa Cruz who has long thought about engineering glaciers. Maybe the conclusion will be that we should not do it, he says, but to shut it down before real research has happened feels politically motivated.Erin Pettit, a glaciologist at Oregon State University in Corvallis, sees it differently. Pettit, who has studied West Antarctica for two decades and has collaborated with Tulaczyk, says glacial engineering is a Band-Aid that could divert money and attention from addressing the underlying problems of climate change. If people are working to stop sea level rise from Antarctica, she fears, were not going to care as much about solving the problem by stopping carbon emissions. She also fears that engineering glaciers or ocean currents could alter the environment in unexpected ways.Even proponents admit that glacial engineering will address only one symptom of climate change sea level rise while leaving problems such as heat waves, permafrost thaw, intensified hurricanes and ocean acidification unchecked. Whats more, unlike in Antarctica, glaciers in Greenland and the Arctic are experiencing massive surface melting, making them a bit less likely to respond to the same interventions.Yet glacial engineering might become necessary. Even optimistic scenarios for global warming, in which CO2 levels peak and level off by 2070, might still lead to Thwaites collapse. Every ice dynamicist on the planet ought to be looking at this, says John Moore, a glaciologist at the University of Lapland in Finland.A stuck glacier inspires glacial geoengineering ideasPeople have talked for decades about engineering ice to serve human purposes in one grand way or another. During World War II, the Allied countries considered building artificial icebergs reinforced with sawdust to create aircraft carriers impervious to German torpedoes. In the 1960s, Soviet scientists proposed building an 80-kilometer barrier across the Bering Strait, from Alaska to Russia, to change ocean currents, reduce sea ice and open vast swaths of Arctic permafrost for farming. During the 1970s, a pair of physicists suggested that nuclear waste could be stored deep in ice sheets. And in the 70s and 80s, engineers in Saudi Arabia suggested that icebergs towed from Antarctica could provide freshwater to places lacking it.Douglas MacAyeal was a graduate student at Princeton in 1983 when he read about those proposals to use icebergs as a source of freshwater. Inspired, he submitted a brief abstract to a scientific meeting suggesting a way to prevent glacial flow from speeding up in the face of a warming climate: Large amounts of seawater pumped onto the floating fronts of glaciers would freeze there, thickening the ice and causing it to rest more heavily on submarine mountains beneath. Anchoring the floating ice to submerged mountains would help it buttress and slow the glacier flowing from behind.MacAyeal never pursued the idea. That was a time of life for me when I have to write papers that are taken seriously, so I can get a job, he says. This silly idea wouldnt get him there. He eventually landed at the University of Chicago, rejoining the conversation on glacial engineering only in 2023, as he was retiring.For Tulaczyk, the roots of his own idea for glacial engineering began in the 1990s while working on his Ph.D. He was studying the Siple Coast of West Antarctica, roughly 1,200 kilometers east of Thwaites, where six massive glaciers called ice streams ooze off the coastline.Thwaites and Pine Island are West Antarcticas fastest-shrinking glaciers. If they collapse, it could doom the whole ice sheet and cause devastating sea level rise.T. Tibbitts, adapted from A. Shepherd, D.J. Wingham and J.A.D. Mansley/GRL 2002Thwaites and Pine Island are West Antarcticas fastest-shrinking glaciers. If they collapse, it could doom the whole ice sheet and cause devastating sea level rise.T. Tibbitts, adapted from A. Shepherd, D.J. Wingham and J.A.D. Mansley/GRL 2002These glaciers generally slide 300 to 700 meters forward per year. But scientists found that one, the Kamb Ice Stream, flows only about one-fiftieth that speed. Though Kamb used to move as quickly as its neighbors, it ground to a near-halt around 150 years ago.Some scientists, such as Tulaczyk, attributed this slowdown in part to the loss of lubrication that normally allows glaciers to slide easily over their rough, gravelly beds. Most glaciers have a thin layer of liquid water beneath them. Its produced as the bottom of the ice slowly melts, a few penny thicknesses per year, from both the heat of friction and the heat trickling out of the earth. The water beneath Kamb seemed to have migrated beneath a different glacier, like a river jumping its banks, causing Kamb to stagnate.By the late 2000s, Tulaczyk had started wondering if it might be possible to slow down other glaciers intentionally, by mimicking what might have happened at Kamb. He imagined drilling a narrow hole through a glacier and pumping out the water beneath. The glacier might eventually freeze to its bed, as Kamb had, remaining stalled for decades or centuries.Tulaczyk presented his idea to a small gathering of climate scientists in 2008. But when he asked the U.S. National Science Foundation to fund a workshop so scientists could further discuss it, he was sharply rejected. Tulaczyk suspects the agency was uncomfortable supporting such a seemingly radical idea because it might provoke negative reactions from a public that ordinarily supports climate research, or draw unwelcome attention from Congress, which approves funding for the agency.Could we freeze glaciers to the seabed to slow sea level rise?A pair of coincidences helped revive the idea a decade later.Around 2018, Tulaczyk received an email from the sci-fi novelist Kim Stanley Robinson, who happened to be one of the few nonscientists in the audience when Tulaczyk gave his talk in 2008. Robinson chatted with him about the idea, and later included it in his 2020 novel, The Ministry for the Future, in which humans successfully respond to climate change and sea level rise. In the book, a glaciologist named Slawek with Tulaczyks OK concocts the same strategy for slowing and stabilizing glaciers.Later, in 2023, Tulaczyk received an email from Alex Luebke, who had spent years founding tech companies, developing satellites and running advanced projects at Google X. Luebke had read Robinsons book, loved the idea of engineering glaciers and wanted to talk with Tulaczyk about how to advance it.Luebke, glaciologist Kenneth Mankoff, a former graduate student of Tulaczyks, and several other people convened a workshop at Stanford University in December 2023 aimed at mapping out how to test the feasibility of glacial engineering and how to find private funding.The very idea provoked strong reactions among the people invited. My first reaction was, This is crazy, admits Martin Truffer, a glaciologist at the University of Alaska Fairbanks who has known Tulaczyk for 30 years. I was really hesitant whether I should go or not.But Mankoff, of NASAs Goddard Institute for Space Studies in New York City, persuaded Truffer and many others. Fifty scientists from around the world showed up. When the group was asked on the first morning who was against glacial engineering, roughly half the hands went up. When asked who was undecided, the other half went up, including Tulaczyks. And when the moderator asked who currently supported the idea, a couple of hands might have wavered, but none went up.To Tulaczyk, it was a good start. Nobody is saying that we are ready to do anything at scale, he says. This current effort is about calling for this type of research to become a legitimate research area.Mankoff, Truffer and Tulaczyk are now making plans to test Tulaczyks idea along with Christine Dow, a subglacial hydrologist at the University of Waterloo in Canada, and Jenny Suckale, a geophysicist at Stanford. The team, which is looking for funding from private foundations, might conduct its first small experiments as soon as next summer, at a small glacier in Alaska. The researchers will use a jet of hot water to melt a bowling ballsized hole through the ice and then pump water out from underneath it for the next month or two.Even if this never leads to glacial engineering, it would still be useful science, Truffer says. It could answer key questions about how basal water controls a glaciers movement, improving the models that are used to predict how quickly glaciers will accelerate as temperatures rise.These experiments would also hint at how many holes might be needed to slow down a massive glacier like Thwaites. Roughly 1 to 3 cubic kilometers of subglacial water flow out from beneath Thwaites each year, according to one estimate. If the holes were strategically placed in an area where the glacier slides over rough bedrock, Tulaczyk speculates that removing just 1 to 3 percent of that water might vastly increase the drag, slowing the glacier. It might mean as few as 10 holes, each with a pump pulling out 100 liters of water per second, which existing well pumps are capable of. At the other extreme, it could mean 100 holes and 10 times as much water.Other researchers are developing variations on Tulaczyks idea. Brent Minchew, a glaciologist at MIT, has suggested that removing heat from the glacial bed would cause the subglacial water to freeze, accomplishing the same thing. He would do this using something called a thermosiphon. Such siphons create a convection current, with warmer gas bubbles constantly rising in a sealed pipe of condensed CO2 and colder liquid constantly sinking to take its place at the bottom. The heat from the gas bubbles would escape through the top of the pipe and into the environment.Thermosiphons are used along the Trans-Alaska Pipeline to prevent permafrost beneath the pipe from thawing and sagging (which could destabilize the pipe). They are also being studied for possible use in geothermal power plants to transport heat from several kilometers underground. At Thwaites, thermosiphons would drain the heat slowly over a period of years before the glacier started freezing onto its bed. But that would be enough, Minchew says: Slow and steady wins this particular race.Removing Thwaites lubrication was one of two major strategies kicked around at the December 2023 workshop. The second was newer and complementary.Could we block warm ocean currents that melt glaciers?The Southern Ocean surrounding Antarctica is known for its rough seas. Nowhere else on Earth can westerly winds circle the globe without encountering land. These winds pile the water into waves that can reach the height of a four-story building and drive the most powerful ocean current on Earth the Antarctic Circumpolar Current.The westerly winds are strengthening due to climate change, causing the circumpolar current to intensify and shift south, in toward the edges of Antarctica. As the current ruffles along the perimeter of Antarcticas continental shelf, the resulting turbulence causes a steady stream of water to billow up from more than a kilometer below the oceans surface onto the edge of the shelf. This circumpolar deep water, owing to its high salt content, is several hundredths of a percent denser than the cooler, less salty water on the shelf. That slight difference is enough to guide the warm, dense water into a deep groove called the Pine Island Trough, which dips several hundred meters below the rest of the shelf.This trough was carved by Thwaites and Pine Island glaciers as they advanced across the continental shelf during the last ice age. Today, the trough provides an easy path for dense, salty water to flow inland and access the fronts of those glaciers.Though the water that reaches the front of Thwaites is just 2 to 3 degrees Celsius above its freezing point, a vast amount of it flows through, roughly 2,800 cubic kilometers per year. Thats nearly enough water to fill Lake Ontario twice. It delivers 900 billion watts of thermal power to the front of Thwaites year-round similar to the output of 450 nuclear power plants.Michael Wolovick thought often about these currents during his early career in glaciology. In 2017, during a postdoc at Princeton, he gave public voice to an idea that hed been mulling for years. While attending the European Geosciences Union meeting in Vienna, he presented a research poster suggesting that massive dikes built across seafloor troughs like the ones in front of Pine Island and Thwaites glaciers could block warm currents and delay glacial retreat.The meeting space was huge, with thousands of posters displayed. Wolovicks was an oddball in a section otherwise devoted to standard measurements of ice shelves and tidewater glaciers. Sometime in the early evening, a scientist named John Moore showed up and introduced himself. He was excited about Wolovicks idea.I didnt realize that anybody else was working on this, says Wolovick, now at the Alfred Wegener Institute for Polar and Marine Research.In March 2018, he, Moore and colleagues published an essay in Nature calling for research into glacial engineering. It drew a sharp retort from seven prominent scientists, who warned that it was a fools errand, likely to distract from the ultimate goal of reducing greenhouse gas emissions.The next September, Moore and Wolovick proposed that an 80- to 120-kilometer-long dike in front of Thwaites could block incoming warm currents by cutting off several branches of the Pine Island Trough. The dike would rise 300 meters above the seafloor, topping out at 250 to 300 meters below the sea surface. Calculations suggested it might stabilize the glacier. But building it would consume 30 to 50 cubic kilometers of rocks, gravel and mud several dozen times the volume of material moved in the decade-long digging of the Suez Canal.Among the people who read those papers was Bowie Keefer, an engineering physicist living near Canadas Vancouver Island. He had worked on desalination, renewable energy and harnessing tides to generate electricity. He loved the idea of blocking ocean currents and contacted Wolovick and Moore to suggest a design more likely to survive the harsh environment.Anything constructed on the seafloor would be under constant threat from the icebergs that crowd the front of Thwaites. Some are up to 400 meters thick. Their undersides frequently scrape and gash the seafloor. If you build a barrier, Keefer says, you have to configure it so icebergs can go over it without destroying it. He imagined something akin to the flexible streamers of kelp that his kayak frequently slid over as he paddled the waters near his home.In March 2023, Wolovick, Moore and Keefer published two papers in PNAS Nexus rolling out a new design: a series of thin, buoyant sea curtains anchored on the seafloor. The curtains would easily bend as icebergs drifted over, while still blocking the dense, salty, bottom-hugging currents. Keefer imagines a modular design, composed of a couple thousand overlapping panels, each about as wide as a football field, that could be replaced individually if damaged.Much of the West Antarctic Ice Sheet is grounded below sea level. That makes glaciers like Thwaites vulnerable to melting from below. Warm, dense, salty currents snake through grooves in the seafloor like submarine rivers. They erode the glaciers foothold, destabilizing it. One engineering plan calls for building sea curtains made of overlapping panels (yellow) anchored to the seafloor to block this warm water (red arrow).T. TibbittsExamining the layout of Pine Island Trough and its various seafloor branches, the team plotted four sections of curtain that would protect Thwaites and Pine Island glaciers, as well as several other nearby glaciers. These flexible barriers would top out a little more than 500 meters below the sea surface just high enough to block the warm bottom currents. But in spots where the troughs are especially deep, that means the curtains would have to reach as much as 250 to 450 meters above the seafloor, equaling the height of the Empire State Building in some places.The estimated total building cost is $40 billion to $80 billion; on a per kilometer basis, thats similar to the cost of building some large bridges. Maintenance might cost another $1 billion to $2 billion per year. These costs might seem astronomical at first glance. But they could be small compared with the cost of building and maintaining dikes to protect coastlines from rising seas, estimated at $20 billion to $55 billion per year every year if global temperatures rise by 3 degrees.But theres some question about whether a barrier might simply redirect warm currents and thus melting to other glaciers farther west along the coast. In 2019, Rodehacke and colleagues published a rough analysis suggesting that, for an 800-kilometer oceanic barrier, that might be the case.Yoshihiro Nakayama, an oceanographer at Dartmouth College, is doing more detailed simulations of how a shorter barrier a 260-kilometer curtain blocking the main trunk of Pine Island Trough would impact the region.Researchers are still considering what building materials to use. Constructing curtains with smooth plastics could allow icebergs to slide harmlessly overtop. But that would release microplastics into some of the worlds most pristine ocean waters a prospect that Keefer does not welcome.Rather than a synthetic curtain, Ole Wroldsen, a marine civil engineer with Entr, the consulting arm of the company Aker Solutions in Fornebu, Norway, envisions other potential options, including a net fabricated from natural plant fibers. Over time, it would become encrusted with sponges, corals, mollusks and other marine animals, increasing its ability to block currents. The goal is to create a living structure that is acting with nature rather than against nature, he says. That would be a perfect match.The first curtain-related field tests could start in a year or two once private funding is arranged. A series of small, 10-meter-long curtains built with different materials could be rooted in a Norwegian fjord, then monitored to see how quickly they deteriorate. Larger versions might later be installed, temporarily, in a glacial fjord in Greenland or the Arctic archipelago of Svalbard, to study the impacts on ocean currents.But even if these small field trials succeed, other political and societal challenges lie ahead.Glacial engineering could have global side effectsLessons from another type of geoengineering suggest that glacial engineering is likely to encounter opposition. For two decades, scientists have used computer models to study the idea of injecting millions of tons of sulfate aerosols into the stratosphere to reduce warming from the incoming sunlight.Proponents see sulfate aerosols as a way to forestall climate catastrophe for a century or two, giving humans time to stop emitting CO2 and remove it from the atmosphere. But because of concerns about unintended side effects, researchers have not succeeded in getting a single field study off the ground. Many people distrust the notion that a handful of wealthy nations having already messed up the planet by emitting greenhouse gases will now fix the problem for everyone by messing with the planet some more.Glacial engineering is far more targeted, geographically, than stratospheric aerosols. But it still may have unintended effects.Sharon Stammerjohn, a sea ice scientist at the University of Colorado Boulder, believes that sea curtains, for example, might cause turbulence that mixes deep, warm water into the upper layers of the coastal ocean. The extra warmth could result in less sea ice production in those areas, she says.A drop in sea ice could disrupt the interplay among photosynthetic plankton, which depend on the melting ice in spring and summer, the krill that eat them, and the penguins and whales that eat the krill. Theres winners and losers, says Stammerjohn.Pettit does not distrust the intentions of people like Tulaczyk, whom she has known for decades. But she and some others are uncomfortable with the idea that at least in the near term, glacial engineering research might be funded, in part, by billionaires from Silicon Valley. She worries whether they will have the humility to refrain from steering the research toward their own preferred outcomes. Any money should have no strings attached, with groups of scientists reviewing and deciding which experiments will be funded, she says. If this is going to happen, we need to make sure theres some reasonable heads involved.Theres also concern that a strong move toward glacial engineering could upset the delicate geopolitics of Antarctica. It is by far the largest piece of land on Earth that is not owned by a particular nation. Though various countries made vast, overlapping territorial claims on the continent during the mid-20th century, the signing of the Antarctic Treaty in 1959 put these claims on hold. Countries pledged to limit their Antarctic activities to scientific research.But tensions persist, says Klaus Dodds, a professor of polar geopolitics at Royal Holloway, University of London. The stations maintained by over two dozen countries there dont merely serve science: They maintain a national presence in Antarctica, allowing those countries to have a seat at the table if the continent is ever divvied up.The Antarctic is already in quite a precarious state at the moment, geopolitically, Dodds says. He believes that any serious glacial engineering effort would be undertaken by a small group of allied countries say, the United States, the United Kingdom, Australia and New Zealand, or perhaps Russia, China and India. It would involve creating new infrastructure on the continent, and this could be viewed as a thinly veiled land grab, he says.Construction in Antarctica presents logistical challengesIf glacial engineering turns out to be scientifically and politically feasible, its ultimate fate will hinge on the ability of humans to build large and complex structures in an area that is notoriously harsh, even for Antarctica.Thwaites Glacier sits midway along a 7,000-kilometer stretch of coastline (roughly the distance from Seattle to Quito, Ecuador) without a single permanent outpost. No human laid eyes on the glaciers ice front until the 1940s, when a U.S. naval plane flew over. No human stood on its heavily crevassed ice shelf until 2019, when a plane first landed. And although heavily armored icebreakers began plying Antarctic waters in 1946, not until 2012 did an icebreaker come within sight of Thwaites Ice Shelf. Even today, the icebreakers that are sent to conduct research have a 50-50 chance of getting there any given year.The region owes its extreme inaccessibility to a quirk of geography that causes ice to pile up. During winter, sea ice up to 1.5 meters thick often extends more than 600 kilometers off the coastline. As the ice breaks up in spring, strong winds gather the ice and push it into the bay where it is compacted and piled up to 10 meters thick in some places. That jumbled ice might continue drifting west, but its stopped by a submarine ridge that extends 100 kilometers off the coast just west of Thwaites.Hundreds of icebergs run aground on that ridge, creating a huge log jam, Stammerjohn says. The stacks of sea ice and iceberg fragments the size of small apartment buildings pile up behind the bergs. The area in front of Thwaites and Pine Island is often choked with ice, even in summer.This satellite photo shows the mlange of sea ice and icebergs that piles up in front of Thwaites. Navigating this jumble is one logistical hurdle to any glacial engineering project that requires the use of ships.LAUREN DAUPHIN/NASA, U.S. GEOLOGICAL SURVEYThis satellite photo shows the mlange of sea ice and icebergs that piles up in front of Thwaites. Navigating this jumble is one logistical hurdle to any glacial engineering project that requires the use of ships.LAUREN DAUPHIN/NASA, U.S. GEOLOGICAL SURVEYA powerful, agile icebreaker working in this environment often has to abandon planned operations. Thats because simply lingering for 20 hours in one place, to launch a submersible or drill a core from the seafloor, is sometimes too risky amid the drifting sea ice and hundred-million-ton icebergs. But a ship towing a 1- to 5-kilometer-long segment of preassembled sea curtains would present a much larger, and less nimble, target for drifting bergs. Workers would rush to lower the new curtain and cinch it to the seafloor over a period of several days, as the crew anxiously monitored the shifting ice.People working 30 to 100 kilometers inland from the coast, drilling through Thwaites Glacier to pump out water, would face different challenges.Subglacial water would have to be pumped from the boreholes year-round, transported through hoses and misted into the air to create snow that would settle harmlessly on the glaciers surface. The entire system would require constant heating, to prevent the kilometer-deep, water-filled holes from freezing shut and to prevent water inside the hoses and pumps from freezing and rupturing the equipment. The power requirements for pumping 1,000 to 10,000 liters of water per second could range from 480,000 to 4.8 million watts similar to the electricity consumption of 400 to 4,000 American households.Solar power is not an option during the long, sunless winter, Truffer says. Wind turbines are used in some parts of Antarctica and five to 50 of them could probably supply the energy for pumping Thwaites Glacier but they would have to be built to survive hurricane-strength gales. If the power were generated by burning diesel on the other hand, it could require 260,000 to 2.6 million gallons of fuel per year, enough to fill as many as 250 semi tanker trucks.This might sound like a ridiculous amount of pumping and energy. But its a mere drop in the bucket, representing no more than 2 percent of the water pumped from all the wells in California each year and a tiny percent of the energy used. Its main significance is that it would require major logistics in a remote region.Hundreds of thousands of kilograms of food, gear and potentially fuel would likely be delivered annually by an icebreaker to an accessible spot on the coastline, 1,000 kilometers northeast of Thwaites. Supplies would be loaded onto convoys of shipping containersized sleds and towed by tractors, traveling on routes carefully surveyed for crevasses, which the British Antarctic Survey often uses in this region.Constant snowfall would bury pumping equipment. Alternating thaws and cold snaps could permeate the snow with layers of rock-hard ice. The machines would have to be dug out yearly with chainsaws, or mounted on stilts that could be raised above the new snow each year. The operation would be extremely massive, Truffer says.In Robinsons novel, scientists eventually use engineering on 30 glaciers. But the real-world scientists imagine targeting no more than a few: maybe Thwaites alone, maybe also Pine Island Glacier, and if things continue to worsen, maybe a couple of others in East Antarctica. If you had to protect glaciers around the whole continent, itd be simpler to build seawalls around cities instead.The key benefit of glacial geoengineering is that you can do this in a very limited geographic area and get big bang for your buck, MacAyeal says.For Tulaczyk, now 58 years old, it has been a long journey since those early days. After the sharp rejection on his funding proposal, he decided to put glacial engineering on the back burner for fear that it would endanger the funding of his other research. But now, Im going to retire, he says. As a result, he feels more free to speak openly. I want to stick to this issue.Hes pleased to see younger scientists like Wolovick and Mankoff also getting involved. They might live long enough to see the results, he says. For a young scientist to be doing this, when theyre trying to start a career, while theres so much hostility, Tulaczyk says, its amazingly courageous.

Nature, Published online: 05 February 2025; doi:10.1038/s41586-024-08387-9A multilayer-stacked two-dimensional polyaniline crystal shows high electrical conductivity and unique out-of-plane metallic transport behaviour, indicating potential for strong electronic coupling beyond in-plane interactions and three-dimensional metallic conductivity.

Nature, Published online: 05 February 2025; doi:10.1038/s41586-024-08380-2By etching cavities into a thin lubricant spacer and encapsulating it with functional flakes, electric and mechanical switching of single-crystal, van der Waals stacking configurations is demonstrated.

Maps created by combining different models of glaciers and ice sheets reveal the way water is flowing deep beneath Antarctica's ice.

Neurons found in one of the oldest regions of the brain may control when we stop eating a meal, new research in mice hints.