Amazonia is the largest1 and the most species-rich tropical forest region on Earth2, where hundreds of Indigenous cultures and thousands of animal species have interacted over millennia3,4. Although Amazonia offers a unique context to appraise the value of wildlife as a source of food to millions of rural inhabitants, the diversity, geographic extent, volumes and nutritional value of harvested wild meat are unknown. Here, leveraging a dataset comprising 447,438 animals hunted across 625 rural localities, we estimate an annual extraction of 0.57 Mt of undressed animal biomass across Amazonia, equivalent to 0.34 Mt of edible wild meat. Just 20 out of 174 taxa account for 72% of all animals hunted and 84% of the overall biomass extracted. We show that this amount of wild meat can meet nearly half of protein and iron dietary requirements for rural peoples, along with a substantial portion of their needs for B vitamins (18-126%) and zinc (23%). However, wild meat productivity is likely to have decreased by 67% in nearly 500,000 km² of highly deforested areas of Amazonia. Furthermore, the availability of wild meat per capita decreases significantly in areas with higher human population, greater proximity to cities, and more extensive deforestation. These findings highlight the urgent need to preserve the forest to safeguard biodiversity and traditional wild meat food systems, which will be essential for ensuring Amazonian peoples' well-being and achieving several of the United Nations Sustainable Development Goals5.

The discovery of Shimao city (around 2300-1800 BCE1), a premier state-level Neolithic fortified settlement in Shaanxi, China2, played an important role in helping us understand the emergence of socially stratified urban societies. However, key questions remain regarding how ancestry and kinship shaped the hierarchy of this class-based society characterized by human sacrifice. The origin of the founding populations of Shimao and other Loess Plateau settlements, and their interactions within the broader ancestral landscape, have yet to be determined. Here we present, by sequencing 144 ancient genomes from Shimao city and its satellites, pedigrees among tomb owners spanning up to four generations. These findings reveal a predominantly patrilineal descent structure across Shimao communities, and possibly sex-specific sacrificial rituals. We also characterize the population history, revealing that Shimao culture-related populations originated mostly from a Yangshao culture-related population present at least 1,000 years earlier, and the lasting inflow of Yumin-related populations from Inner Mongolia did not interrupt regional genetic continuity. Broader genetic influence from southern mainland ancestry over Shimao culture-related populations supports evidence of rice farming expanding further north than previously expected. Together, these results uncover fine details of the regional peopling and social structure of early state establishment.

The naming of Australopithecus deyiremeda1 from Woranso-Mille (less than 3.59 to more than 3.33 million years) indicated the presence of a species contemporaneous with Australopithecus afarensis in the Ethiopian Afar Rift. A partial foot (BRT-VP-2/73)2 and several isolated teeth from two Burtele (BRT) localities, however, were not identified to the species level. Recently recovered dentognathic specimens clarify not only the taxonomic affinity of the BRT hominin specimens but also shed light on the diet and locomotion of A. deyiremeda. Here we present a comparative description of these specimens and show that they are attributable to A. deyiremeda. We also find it parsimonious to attribute the BRT foot to this species based on the absence of other hominin species at BRT. The new material demonstrates that overall, A. deyiremeda was dentally and postcranially more primitive than A. afarensis, particularly in aspects of canine and premolar morphology, and in its retention of pedal grasping traits. Furthermore, the low and less variable distributions of its dental enamel δ13C values are similar to those from Ardipithecus ramidus and Australopithecus anamensis, indicating a reliance on C3 foods. This suggests that A. deyiremeda had a dietary strategy similar to the earlier A. ramidus and A. anamensis. The BRT foot and its assignment to A. deyiremeda provides conclusive evidence that arboreality was a significant component of the positional behaviour of this australopith, further corroborating that some degree of arboreality persisted among Pliocene hominins1,3-7.

Glasses-free three-dimensional (3D) displays provide users with an immersive visual experience without the need of any wearable devices1,2. To achieve high-quality 3D imaging, a display should have both large linear dimensions and a wide viewing angle. However, the trade-off between spatial extent and bandwidth of optical systems, the space-bandwidth product, conventionally constrains the simultaneous maximization of the two. The two most common approaches to 3D displays are holographic3,4 and automultiscopic1,5,6, which, respectively, sacrifice either scale or viewing angle. Recently, some implementations enhanced by artificial intelligence have shown directions to mitigate these constraints, but they still operate within a set space-bandwidth product7,8. As a result, it remains challenging to fabricate large-scale wide-angle 3D displays9. Here we report the realization of a large-scale full-parallax 3D display with seamless viewing beyond 100°, maintained at over 50 Hz and 1,920 × 1,080 resolution on a low-cost light-field delivery setup. This device, called EyeReal, is realized by accurately modelling binocular view and combining it with a deep-learning real-time optimization, enabling the generation of optimal light-field outputs for each of the eyes. Our device could potentially enable applications in educational tools, 3D design and virtual reality10,11.

NAND flash memory is essential in modern storage technology, amid growing demands for low-power operation fuelled by data-centric computing and artificial intelligence1,2. Its unique 'string' architecture3, where multiple cells are connected in series, requires high-voltage pass operation that causes a large amount of undesired power consumption4. Lowering the pass voltage, however, poses a challenge: it leads to an associated reduction in the memory window, restricting the multi-level operation capability. Here, with a gate stack composed of zirconium-doped hafnia and an oxide semiconductor channel, we report ultralow-power ferroelectric field-effect transistors (FeFETs) that resolve this dilemma. Our FeFETs secure up to 5-bit per cell multi-level capability, which is on par with or even exceeds current NAND technology, while showing nearly zero pass voltage, saving up to 96% power in string-level operations over conventional counterparts. Three-dimensional integration of FeFET stacks into vertical structures with a 25-nm short channel preserves robust electrical properties and highlights low-pass-voltage string operation in scaled dimensions. Our work paves the way for next-generation storage memory with enhanced capacity, power efficiency and reliability.

Mechanical transmission is essential in force-related activities ranging from the daily tying of shoe laces1 to sophisticated surgical2 and robotic operations3,4. Modern machines and robots typically use complex electronic devices designed to sense and limit force5, some of which still face challenges when operating space is limited (for example, in minimally invasive surgeries)6 or when resources are scarce (for example, operations in remote areas without electricity). Here we describe an alternative slipknot-based mechanical transmission mechanism to control the intelligent operation of both human and robotic systems. Through topological design, slipknot tying and release can encode and deliver force with a consistency of 95.4% in repeating operations, which circumvents the need for additional sensors and controllers. When applied to surgical repair, this mechanism helped inexperienced surgeons to improve their knotting-force precision by 121%, enabling them to perform surgical knots as good as those of experienced surgeons. Moreover, blood supply and tissue healing after surgery were improved. The mechano-intelligence exhibited in slipknots may inspire investigations of knotted structures across multiple length scales. This slipknot-gauged mechanical transmission strategy can be widely deployed, opening up opportunities for resource-limited healthcare, science education and field exploration.

Detecting individual spins-including stable and metastable states-represents a fundamental challenge in quantum sensing, with broad applications across condensed matter physics1,2, quantum chemistry3 and single-molecule magnetic resonance imaging4,5. Although nitrogen-vacancy (NV) centres in diamond have emerged as powerful nanoscale sensors, their performance for single-spin detection remains constrained by substantial environmental noise and restricted sensing volume6,7. Here we propose and demonstrate an entanglement-enhanced sensing protocol that overcomes these limitations through the strategic use of entangled NV pairs. Our approach achieves a 3.4-fold enhancement in sensitivity and a 1.6-fold improvement in spatial resolution relative to single NV centres under ambient conditions. The protocol uses carefully engineered entangled states that amplify target spin signals through quantum interference while suppressing environmental noise. Crucially, we extend these capabilities to resolve metastable single-spin dynamics, directly observing stochastic transitions between different spin states by identifying state-dependent coupling strengths. This dual functionality enables simultaneous detection of static and dynamic spin species for studying complex quantum systems. The achieved performance establishes entanglement-enhanced sensing as a viable pathway towards atomic-scale characterization of quantum materials and interfaces.

Semiconductors are among the most promising platforms to implement large-scale quantum computers, as advanced manufacturing techniques allow fabrication of large quantum dot arrays1. Various qubit encodings can be used to store and manipulate quantum information on these quantum dot arrays. Regardless of qubit encoding, precise control over the exchange interaction between electrons confined in quantum dots in the array is critical. Furthermore, it is necessary to execute high-fidelity quantum operations concurrently to make full use of the limited coherence of individual qubits. Here we demonstrate the parallel operation of two exchange-only qubits, consisting of six quantum dots in a linear arrangement. Using randomized benchmarking (RB) techniques, we show that issuing pulses on the five barrier gates to modulate exchange interactions in a maximally parallel way maintains the quality of qubit control relative to sequential operation. The techniques developed to perform parallel exchange pulses can be readily adapted to other quantum-dot-based encodings. Moreover, we show the first, to our knowledge, experimental demonstrations of an iSWAP gate for exchange-only qubits and of a charge-locking Pauli spin blockade (PSB) readout method. The results are validated using cross-entropy benchmarking (XEB)2, a technique useful for performance characterization of larger quantum computing systems; here it is used for the first time on a quantum system based on semiconductor technology.

Lightning is among the most energetic manifestation of electrical activity in planetary atmospheres, with documented observations not only on Earth but also on Saturn and Jupiter1. On Mars, the existence of electrical activity has long been suspected2,3 but never directly demonstrated. The dusty atmosphere of Mars undergoes aeolian processes, ranging from wind-blown dust and sand, metre-to-hundred-metre-sized dust devils to thousand-kilometre-scale dust storms4, which, in Earth's deserts, can become electrified through triboelectric charging5-7. For this reason, electric fields have been predicted to build up on Mars8-10, but with no measurement of Martian atmospheric electrical activity so far. Here we report in situ detections of triboelectric discharges, identified by their electrical and acoustic signatures captured by the SuperCam microphone aboard the Perseverance rover11,12. Fifty-five events have been detected over two Martian years, usually associated with dust devils and dust storm convective fronts. These serendipitous observations demonstrate that Martian electric fields can reach the breakdown threshold of the near-surface atmosphere of Mars, predicted to be on the order of several tens of kV m-1. Such electrical activity could affect dust dynamics13,14 and potentially fuel a reactive electrochemical environment enhancing the oxidizing capacity of the atmosphere, with consequences for the preservation of organic molecules15,16. This in situ evidence may have implications for surface chemistry, habitability and human exploration.

Nitrogen vacancy (NV) centres in diamond are widely deployed as local magnetic sensors, using single-qubit control to measure both time-averaged fields and noise with nanoscale spatial resolution1. Moving beyond single qubits to multi-qubit control enables new sensing modalities such as measuring nonlocal spatiotemporal correlators2 or using entangled states to enhance measurement sensitivity3. Here we describe protocols for using optically unresolved NV centre pairs and nuclear spins as multi-qubit sensors for measuring correlated noise at nanometre length scales. For noninteracting NV centres, we implement a phase-cycling protocol that disambiguates magnetic correlations from variance fluctuations, leveraging the presence of a third qubit, a 13C nucleus, to effect coherent single-NV spin flips and enable phase cycling even for co-aligned NV centres that are spectrally unresolved. For length scales around 10 nm, we create maximally entangled Bell states through dipole-dipole coupling between two NV centres and use these entangled states to directly read out the magnetic field correlation, rather than reconstructing it from independent measurements of unentangled NV centres. Importantly, this changes the scaling of sensitivity with readout noise from quadratic to linear. For conventional off-resonant readout of the NV centre spin state (for which the readout noise is roughly 30 times the quantum projection limit), this results in more than an order of magnitude improvement in sensitivity. Finally, we demonstrate methods for detecting high spatial- and temporal-resolution correlators with pairs of strongly interacting NV centres.