87191 research outputs found
Sort by
Chronology of our Galaxy from Gaia colour-magnitude diagram fitting (ChronoGal): IV. The inner Milky Way stellar age distribution
Get PDFThe Milky Way’s inner region is dominated by a stellar bar and a boxy-peanut-shaped bulge. However, which stellar populations inhabit the inner Galaxy or how star formation proceeded there is still unknown. The difficulty in studying these stars stems from their location in dense regions that are strongly impacted by extinction and crowding effects. In this work we used star formation histories computed in the solar neighbourhood via Gaia colour-magnitude diagram fitting to shed light on the evolution of the central regions of our Galaxy. For that, we obtained precise age distributions for the non-negligible amount of super-metal-rich stars ([M/H] ∼ 0.5) in the solar neighbourhood (more than 5% of the total stars within 400 pc of the plane). Assuming that these stars were born in the inner Galaxy and migrated outwards, those distributions should be indicative of the true stellar age distribution in the inner Galaxy. Surprisingly, we find that these age distributions are not continuous but show clear signs of episodic star formation (∼13.5, 10.0, 7.0, 4.0, 2.0, and less than 1 Gyr ago). Interestingly, with the exception of the 4 Gyr event, the timings of the detected events coincide with the formation of the primitive Milky Way and with known merging events or satellite encounters (Gaia-Enceladus-Sausage, Sagittarius dwarf galaxy, and the Magellanic Clouds), suggesting that these events could have triggered global star-forming episodes. These results are compatible with a scenario in which Gaia-Enceladus-Sausage is responsible for the formation of the bar 10 Gyr ago. However, we cannot associate any accretion counterpart with the event that occurred 4 Gyr ago, leaving open the possibility of a late formation of the bar, as previously proposed. The Auriga Superstars simulations also indicate that metal-rich stars in the solar neighbourhood-like regions formed at discrete times and migrated from the inner parts of barred galaxies, suggesting a possible link to bar dynamics and satellite accretion. This novel analysis allows us to indirectly witness the evolution of the inner Milky Way and constrain dynamical models of the Milky Way bar
The GECKOS survey: The formation history of a barred galaxy via structural decomposition and spatially resolved spectroscopy
Get PDFApproximating 3D bedrock deformation in an Antarctic ice-sheet model for projections
Get PDFThe bedrock deformation in response to a melting ice sheet provides negative feedback on ice mass loss. When modelling the future behaviour of the Antarctic Ice Sheet, the impact of bed deformation on ice dynamics varies but can reduce projections of future sea-level rise by up to 40%incomparison with scenarios that assume a rigid Earth. The rate of the solid Earth response is mainly dependent on the viscosity of the Earth’s mantle, which varies laterally and radially with several orders of magnitude across Antarctica. Because modelling the response for a varying viscosity is computationally expensive and has only recently been shown to be necessary over centennial time scales, sea-level projection ensembles often exclude the Earth’s response or apply a globally constant relaxation time or viscosity. We use a coupled model to investigate the accuracy of various approaches to modelling the bedrock deformation to ice load change. Specifically, we compare the sea-level projections from an ice-sheet model coupled to (i) an elastic lithosphere, relaxed asthenosphere (ELRA) model, with either uniform and laterally varying relaxation times, (ii) a glacial isostatic adjustment (GIA) model with a radially varying Earth structure (1DGIAmodel),and(iii) a GIA model with laterally varying earth structures (3D GIA model). Furthermore, using the 3D GIA model we determine a relation between relaxation time and viscosity which can be used in ELRA and 1D models. We conduct 500-year projections of Antarctic Ice Sheet evolution using two different climate models and two emissions scenarios: the high emission scenario SSP5-8.5 and the low emission scenario SSP1-2.6. Using a rigid Earth model, this results in ∼ 3–7.5m of barystatic sea-level rise with significant retreat in various basins due to marine ice sheet instability. The results show that using a uniform relaxation time of 300 years in an ELRA model leads to a total sea-level rise that deviates less than 40cm (6%) from the average of the 3D GIA models in 2500. This difference in the projected sea-level rise can be further reduced to 20cm (4%) by using an upper mantle viscosity of 1019 Pas in the 1D GIA model, and to 10cm (2%) in 2500 by using a laterally varying relaxation time map in an ELRA model. Our results show that the Antarctic Ice Sheet contribution to sea-level rise can be approximated sufficiently accurate using ELRA or a 1D GIA model when the recommended parameters derived from the full 3D GIA model are used
An Integrated Pyrolysis Approach for Hydrogen Production and Microplastic Elimination from Sewage Sludge: Experimental and Analytical Perspectives
Get PDFLatency-Constrained Resource Synergization for Mission-Oriented 6G Non-Terrestrial Networks
No full textFormats of representation in large language models
Get PDFThis paper argues for a pluralist approach to representation in large language models. There are two parts to this pluralism, the first is that we should recognise more than one vehicle of representation in transformer models. Call this vehicle pluralism. Rather than identifying the vehicles of representation with a single component of a system, e.g. individual neurons, patterns of activation, regions in the activation space, we should acknowledge multiple systems of representation within a network operating with different vehicles. The second claim is that we should recognise that there are different formats of representation in transformer models. Transformer models do not operate with a purely analogue, structural, or symbolic architecture but are a hybrid system of representation. Finally, I will discuss how this relates to several working hypotheses about representation that have become adopted in the field of mechanistic interpretability
EEG-Driven Intention Decoding: Offline Deep Learning Benchmarking on a Robotic Rover
No full textBrain–computer interfaces (BCIs) provide a hands-free control modality for mobile robotics, yet decoding user intent during real-world navigation remains challenging. This work presents a brain–robot control framework for offline decoding of driving commands during robotic rover operation. A 4WD Rover Pro platform was remotely operated by 12 participants who navigated a predefined route using a joystick, executing the following commands: forward, reverse, left, right, and stop. Electroencephalogram (EEG) signals were recorded with a 16-channel OpenBCI cap and aligned with motor actions at Δ = 0 ms and eight future prediction horizons (Δ > 0 ms). After data preprocessing, eleven deep learning (DL) models were benchmarked for the task of intent classification, across the Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), and Transformer architectural families. ShallowConvNet achieved the highest performance for both action prediction (F1-score 67% at Δ = 0 ms) and intent prediction (F1-score 66% at Δ = 300 ms), maintaining robust performance at future horizons. By combining real-world robotic control with multi-horizon EEG intention decoding, this study introduces a reproducible benchmark and reveals key design insights for predictive, DL-based BCI systems
Data sources:
Access Repository Dashboard
Do you manage Open Research Online? Become a CORE Member to access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard! 👇