Search

Educational research highlights that improved careers education is needed to increase participation in science, technology, engineering, and mathematics (STEM). Current UK careers resources concerning the space sector, however, are found to perhaps not best reflect the diversity of roles present and may in fact perpetuate misconceptions about the usefulness of science. We, therefore, compile a more diverse set of space-related jobs, which will be used in the development of a new space careers resource.

To maximize the economic benefits of geothermal energy production, it is essential to optimize geothermal reservoir management strategies, in which geologic uncertainty should be considered. In this work, we propose a closed-loop optimization framework, based on deep learning surrogates, for the well control optimization of geothermal reservoirs. In this framework, we construct a hybrid convolution-recurrent neural network surrogate, which combines the convolution neural network (CNN) and long short-term memory (LSTM) recurrent network. The convolution structure can extract spatial information of geologic parameter fields and the recurrent structure can approximate sequence-to-sequence mapping. The trained model can predict time-varying production responses (rate, temperature, etc.) for cases with different permeability fields and well control sequences. In the closed-loop optimization framework, production optimization based on the differential evolution (DE) algorithm, and data assimilation based on the iterative ensemble smoother (IES), are performed alternately to achieve real-time well control optimization and geologic parameter estimation as the production proceeds. In addition, the averaged objective function over the ensemble of geologic parameter estimations is adopted to consider geologic uncertainty in the optimization process. Several geothermal reservoir development cases are designed to test the performance of the proposed production optimization framework. The results show that the proposed framework can achieve efficient and effective real-time optimization and data assimilation in the geothermal reservoir production process., Comment: 37 pages, 24 figures

Blazars optical emission is generally dominated by relativistic jets, although the host galaxy, accretion disk and broad line region (BLR) may also contribute significantly. Disentangling their contributions has been challenging for years due to the dominance of the jet. To quantify the contributions to the spectral variability, we use the statistical technique for dimensionality reduction Non-Negative Matrix Factorization on a spectroscopic data set of 26 $\gamma$-ray blazars. This technique allows to model large numbers of spectra in terms of a reduced number of components.We use a priori knowledge to obtain components associated to meaningful physical processes. The sources are classified according to their optical spectrum as host-galaxy dominated BL Lac objects (BL Lacs), BL Lacs, or Flat Spectrum Radio Quasars (FSRQs). Host-galaxy sources show less variability, as expected, and bluer-when-brighter trends, as the other BL Lacs. For FSRQs, more complicated colour-flux behaviours are observed: redder-when-brighter for low states saturating above a certain level and, in some cases, turning to bluer-when-brighter. We are able to reproduce the variability observed during 10 years using only 2 to 4 components, depending on the type. The simplest scenario corresponds to host-galaxy blazars, whose spectra are reconstructed using the stellar population and a power law for the jet. BL Lac spectra are reproduced using from 2 to 4 power laws. Different components can be associated to acceleration/cooling processes taking place in the jet. The reconstruction of FSRQs also incorporates a QSO-like component to account for the BLR, plus a very steep power law, associated to the accretion disk., Comment: Accepted for publication in MNRAS

Recent advances in multimodal vision and language modeling have predominantly focused on the English language, mostly due to the lack of multilingual multimodal datasets to steer modeling efforts. In this work, we address this gap and provide xGQA, a new multilingual evaluation benchmark for the visual question answering task. We extend the established English GQA dataset to 7 typologically diverse languages, enabling us to detect and explore crucial challenges in cross-lingual visual question answering. We further propose new adapter-based approaches to adapt multimodal transformer-based models to become multilingual, and -- vice versa -- multilingual models to become multimodal. Our proposed methods outperform current state-of-the-art multilingual multimodal models (e.g., M3P) in zero-shot cross-lingual settings, but the accuracy remains low across the board; a performance drop of around 38 accuracy points in target languages showcases the difficulty of zero-shot cross-lingual transfer for this task. Our results suggest that simple cross-lingual transfer of multimodal models yields latent multilingual multimodal misalignment, calling for more sophisticated methods for vision and multilingual language modeling., Comment: Findings of ACL 2022

Reasoning over multiple modalities, e.g. in Visual Question Answering (VQA), requires an alignment of semantic concepts across domains. Despite the widespread success of end-to-end learning, today's multimodal pipelines by and large leverage pre-extracted, fixed features from object detectors, typically Faster R-CNN, as representations of the visual world. The obvious downside is that the visual representation is not specifically tuned to the multimodal task at hand. At the same time, while transformer-based object detectors have gained popularity, they have not been employed in today's multimodal pipelines. We address both shortcomings with TxT, a transformer-based crossmodal pipeline that enables fine-tuning both language and visual components on the downstream task in a fully end-to-end manner. We overcome existing limitations of transformer-based detectors for multimodal reasoning regarding the integration of global context and their scalability. Our transformer-based multimodal model achieves considerable gains from end-to-end learning for multimodal question answering., Comment: To appear at the 43rd DAGM German Conference on Pattern Recognition (GCPR) 2021

To study the nuclear ($\lesssim1\,$kpc) dust of nearby ($z<0.1$) type 1 Quasi Stellar Objects (QSOs) we obtained new near-infrared (NIR) high angular resolution ($\sim0.3$ arcsec) photometry in the H and Ks bands, for 13 QSOs with available mid-infrared (MIR) high angular resolution spectroscopy ($\sim7.5-13.5\,\mu$m). We find that in most QSOs the NIR emission is unresolved. We subtract the contribution from the accretion disk, which decreases from NIR ($\sim35\%$) to MIR ($\sim2.4\%$). We also estimate these percentages assuming a bluer accretion disk and find that the contibution in the MIR is nearly seven time larger. We find that the majority of objects ($64\%$, 9/13) are better fitted by the Disk+Wind H17 model \citep[][]{Hoenig17}, while others can be fitted by the Smooth F06 \citep[$14\%$, 2/13,][]{Fritz06}, Clumpy N08 \citep[$7\%$, 1/13,][]{Nenkova08a,Nenkova08b}, Clumpy H10 \citep[$7\%$, 1/13,][]{Hoenig10b}, and Two-Phase media S16 \citep[$7\%$, 1/13,][]{Stalev16} models. However, if we assume the bluer accretion disk, the models fit only 2/13 objects. We measured two NIR to MIR spectral indexes, $\alpha_{NIR-MIR(1.6,8.7\,\mu\text{m})}$ and $\alpha_{NIR-MIR(2.2,8.7\,\mu\text{m})}$, and two MIR spectral indexes, $\alpha_{MIR(7.8, 9.8\,\mu\text{m})}$ and $\alpha_{MIR(9.8, 11.7\,\mu\text{m})}$, from models and observations. From observations, we find that the NIR to MIR spectral indexes are $\sim-1.1$ and the MIR spectral indexes are $\sim-0.3$. Comparing the synthetic and observed values, we find that none of the models simultaneously match the measured NIR to MIR and $7.8-9.8\,\mu$m slopes. However, we note that measuring the $\alpha_{MIR(7.8, 9.8\,\mu\text{m})}$ on the starburst-subtracted {\it Spitzer}/IRS spectrum, gives values of the slopes ($\sim-2$) that are similar to the synthetic values obtained from the models., Comment: Accepted in ApJ

We study the problem of deconvolution for light-sheet microscopy, where the data is corrupted by spatially varying blur and a combination of Poisson and Gaussian noise. The spatial variation of the point spread function (PSF) of a light-sheet microscope is determined by the interaction between the excitation sheet and the detection objective PSF. First, we introduce a model of the image formation process that incorporates this interaction, therefore capturing the main characteristics of this imaging modality. Then, we formulate a variational model that accounts for the combination of Poisson and Gaussian noise through a data fidelity term consisting of the infimal convolution of the single noise fidelities, first introduced in L. Calatroni et al. "Infimal convolution of data discrepancies for mixed noise removal", SIAM Journal on Imaging Sciences 10.3 (2017), 1196-1233. We establish convergence rates in a Bregman distance under a source condition for the infimal convolution fidelity and a discrepancy principle for choosing the value of the regularisation parameter. The inverse problem is solved by applying the primal-dual hybrid gradient (PDHG) algorithm in a novel way. Finally, numerical experiments performed on both simulated and real data show superior reconstruction results in comparison with other methods., Comment: 34 pages, 13 figures

The magnetopause marks the outer edge of the Earth's magnetosphere and a distinct boundary between solar wind and magnetospheric plasma populations. In this letter, we use global magnetohydrodynamic simulations to examine the response of the terrestrial magnetopause to fast-forward interplanetary shocks of various strengths and compare to theoretical predictions. The theory and simulations indicate the magnetopause response can be characterised by three distinct phases; an initial acceleration as inertial forces are overcome, a rapid compressive phase comprising the majority of the distance travelled, and large-scale damped oscillations with amplitudes of the order of an Earth radius. The two approaches agree in predicting subsolar magnetopause oscillations with frequencies 2-13 mHz but the simulations notably predict larger amplitudes and weaker damping rates. This phenomenon is of high relevance to space weather forecasting and provides a possible explanation for magnetopause oscillations observed following the large interplanetary shocks of August 1972 and March 1991., Comment: 9 pages, 3 figures, 1 table. Accepted as a Geophysical Research Letter on 09 July 2021

We compare mid-IR and ALMA far-IR images of 12 nearby Seyferts selected from GATOS. The mid-IR unresolved emission contributes more than 60% of the nuclear emission in most galaxies. By contrast, the ALMA 870micron continuum emission is mostly resolved and typically along the torus equatorial direction (Paper I, Garcia-Burillo et al. 2021). The Eddington ratios and nuclear hydrogen column densities NH of half the sample are favorable to launching polar and/or equatorial dusty winds, according to simulations. Six show mid-IR extended emission in the polar direction as traced by the NLR and perpendicular to the ALMA emission. In a few, the nuclear NH might be too high to uplift large quantities of dusty material along the polar direction. Five galaxies have low NH and/or Eddington ratios and thus polar dusty winds are not likely. We generate new CAT3D-WIND disk-wind model images. At low wind-to-disk cloud ratios the far-IR model images have disk- and ring-like morphologies. The X-shape associated with dusty winds is seen better in the far-IR at intermediate-high inclinations for the extended-wind configurations. In most models, the mid-IR emission comes from the inner part of the disk/cone. Extended bi-conical and one-sided polar mid-IR emission is seen in extended-wind configurations and high wind-to-disk cloud ratios. When convolved to our resolution, the model images reproduce qualitative aspects of the observed morphologies. Low-intermediate wind-to-disk ratios are required to account for the large fractions of unresolved mid-IR emission. This work and Paper I provide observational support for the torus+wind scenario. The wind component is more relevant at high Eddington ratios and/or AGN luminosities, and polar dust emission is predicted at NH of up to $10^{24}$cm$^{-2}$. The torus/disk component, on the other hand, prevails at low luminosities and/or Eddington ratios. (Abridged), Comment: 27 pages, 27 figures. Accepted for publication in A&A

This paper presents a deep learning-based de-homogenization method for structural compliance minimization. By using a convolutional neural network to parameterize the mapping from a set of lamination parameters on a coarse mesh to a one-scale design on a fine mesh, we avoid solving the least square problems associated with traditional de-homogenization approaches and save time correspondingly. To train the neural network, a two-step custom loss function has been developed which ensures a periodic output field that follows the local lamination orientations. A key feature of the proposed method is that the training is carried out without any use of or reference to the underlying structural optimization problem, which renders the proposed method robust and insensitive wrt. domain size, boundary conditions, and loading. A post-processing procedure utilizing a distance transform on the output field skeleton is used to project the desired lamination widths onto the output field while ensuring a predefined minimum length-scale and volume fraction. To demonstrate that the deep learning approach has excellent generalization properties, numerical examples are shown for several different load and boundary conditions. For an appropriate choice of parameters, the de-homogenized designs perform within $7-25\%$ of the homogenization-based solution at a fraction of the computational cost. With several options for further improvements, the scheme may provide the basis for future interactive high-resolution topology optimization., Comment: 28 pages, 16 figures

We present the first results of the Galaxy Activity, Torus and Outflow Survey (GATOS), a project aimed at understanding the properties of the dusty molecular tori and their connection to the host galaxy in nearby Seyfert galaxies. Our project expands the range of AGN luminosities and Eddington ratios covered by previous surveys of Seyferts conducted by ALMA and allows us to study the gas feeding and feedback cycle in a combined sample of 19 Seyferts. We used ALMA to obtain new images of the emission of molecular gas and dust using the CO(3-2) and HCO+(4-3) lines as well as their underlying continuum emission at 870 microns with high spatial resolutions (0.1'' ~ 7 - 13 pc) in the CND of 10 nearby (D < 28 Mpc) Seyfert galaxies. Our new ALMA observations detect 870 micron continuum and CO line emission from spatially resolved disks located around the AGN in all the sources. The bulk of the continuum flux can be accounted for by thermal emission from dust in the majority of the targets. For most of the sources the disks show a preponderant orientation perpendicular to the AGN wind axes, as expected for dusty molecular tori. The median diameters and molecular gas masses of the tori are ~ 42 pc, and ~ 6 x 10**5 Msun, respectively. We find a positive correlation between the line-of-sight gas column densities responsible for the absorption of X-rays and the molecular gas column densities derived from CO towards the AGN in our sources. The radial distributions of molecular gas in the CND of our combined sample show signs of nuclear-scale molecular gas deficits. We also detect molecular outflows in the sources that show the most extreme nuclear-scale gas deficits in our sample. These observations find for the first time supporting evidence that the imprint of AGN feedback is more extreme in higher luminosity and/or higher Eddington ratio Seyfert galaxies., Comment: 41 pages, new version revised and accepted by A&A on the 22nd of June 2021

Societal biases are a major issue in school students' access to and interaction with science. Schools engagement programmes in science from universities, like independent research projects, which could try and tackle these problems are, however, often inequitable. We evaluate these concerns applied to one such programme, `Physics Research in School Environments' (PRiSE), which features projects in space science, astronomy, and particle physics. Comparing the schools involved with PRiSE to those of other similar schemes and UK national statistics, we find that PRiSE has engaged a much more diverse set of schools with significantly more disadvantaged groups than is typical. While drop-off occurs within the protracted programme, we find no evidence of systematic biases present. The majority of schools that complete projects return for multiple years of the programme, with this repeated buy-in from schools again being unpatterned by typical societal inequalities. Therefore, schools' ability to succeed at independent research projects appears independent of background within the PRiSE framework. Qualitative feedback from teachers show that the diversity and equity of the programme, which they attribute to the level of support offered through PRiSE's framework, is valued and they have highlighted further ways of making the projects potentially even more accessible. Researcher-involvement, uncommon in many other programmes, along with teacher engagement and communication are found to be key elements to success in independent research projects overall.

Physics in schools is distinctly different from, and struggles to capture the excitement of, university research-level work. Initiatives where students engage in independent research linked to cutting-edge physics within their school over several months might help mitigate this, potentially facilitating the uptake of science in higher education. However, how such initiatives are best supported remains unclear and understudied. This paper evaluates a provision framework, `Physics Research in School Environments' (PRiSE), using survey data from participating 14-18 year-old students and their teachers to understand their experience of the programme. The results show that PRiSE appears to provide much more positive experiences than typical university outreach initiatives due to the nature of the opportunities afforded over several months, which schools would not be able to provide without external input. The intensive support offered is deemed necessary, with all elements appearing equally important. Based on additional feedback from independent researchers and engagement professionals, we also suggest the framework could be adopted at other institutions and applied to their own areas of scientific research, something which has already started to occur.

Using 6 years of evaluation data we assess the medium- and long-term impacts upon a diverse range of students, teachers, and schools from participating in a programme of protracted university-mentored projects based in cutting-edge space science, astronomy, and particle physics research. After having completed their 6-month-long projects, the 14-18 year-old school students report having substantially increased in confidence relating to relevant scientific topics and methods as well as having developed numerous skills, outcomes which are corroborated by teachers. There is evidence that the projects helped increase students' aspirations towards physics, whereas science aspirations (generally high to begin with) were typically maintained or confirmed through their involvement. Longitudinal evaluation 3 years later has revealed that these projects have been lasting experiences for students which they have benefited and drawn upon in their subsequent university education. Data on students' destinations suggests that their involvement in research projects has made them more likely to undertake physics and STEM degrees than would otherwise be expected. Cases of co-created novel physics research resulting from PRiSE also has seemed to have a powerful effect, not only on the student co-authors but participating students from other schools also. Teachers have also been positively affected through participating, with the programme having influenced their own knowledge, skills, and pedagogy, as well as having advantageous effects felt across their wider schools. These impacts suggest that similar `research in schools' initiatives may have a role to play in aiding the increased uptake and diversity of physics and/or STEM in higher education as well as meaningfully enhancing the STEM environment within schools.

The discovery of topological order has revolutionized the understanding of quantum matter in modern physics and provided the theoretical foundation for many quantum error correcting codes. Realizing topologically ordered states has proven to be extremely challenging in both condensed matter and synthetic quantum systems. Here, we prepare the ground state of the toric code Hamiltonian using an efficient quantum circuit on a superconducting quantum processor. We measure a topological entanglement entropy near the expected value of $\ln2$, and simulate anyon interferometry to extract the braiding statistics of the emergent excitations. Furthermore, we investigate key aspects of the surface code, including logical state injection and the decay of the non-local order parameter. Our results demonstrate the potential for quantum processors to provide key insights into topological quantum matter and quantum error correction., Comment: 6 pages 4 figures, plus supplementary materials

In this article we provide formulations of energy flux and radiation stress consistent with the scaling regime of the Korteweg-de Vries (KdV) equation. These quantities can be used to describe the shoaling of cnoidal waves approaching a gently sloping beach. The transformation of these waves along the slope can be described using the shoaling equations, a set of three nonlinear equations in three unknowns: the wave height H, the set-down and the elliptic parameter m. We define a numerical algorithm for the efficient solution of the shoaling equations, and we verify our shoaling formulation by comparing with experimental data from two sets of experiments as well as shoaling curves obtained in previous works., Comment: 19 pages, 7 figures

Background SAMtools and BCFtools are widely used programs for processing and analysing high-throughput sequencing data. Findings The first version appeared online twelve years ago and has been maintained and further developed ever since, with many new features and improvements added over the years. The SAMtools and BCFtools packages represent a unique collection of tools that have been used in numerous other software projects and countless genomic pipelines. Conclusion Both SAMtools and BCFtools are freely available on GitHub under the permissive MIT licence, free for both non-commercial and commercial use. Both packages have been installed over a million times via Bioconda. The source code and documentation are available from http://www.htslib.org.

The optimization and scheduling of scientific observations done with instrumentation supported by adaptive optics could greatly benefit from the forecast of PSF figures of merit (FWHM, Strehl Ratio, Encircle Energy and contrast), that depend on the AO instrument, the scientific target and turbulence conditions during the observing night. In this contribution we explore the the possibility to forecast a few among the most useful PSF figures of merit (SR and FWHM). To achieve this goal, we use the optical turbulence forecasted by the mesoscale atmospheric model Astro-Meso-NH on a short timescale as an input for PSF simulation software developed and tailored for specific AO instruments. A preliminary validation will be performed by comparing the results with on-sky measured PSF figures of merit obtained on specific targets using the SCAO systems SOUL (FLAO upgrade) feeding the camera LUCI at LBT and SAXO, the extreme SCAO system feeding the high resolution SPHERE instrument at VLT. This study will pave the way to the implementation of an operational forecasts of such a figure of merits on the base of existing operational forecast system of the atmosphere (turbulence and atmospheric parameters). In this contribution we focus our attention on the forecast of the PSF on-axis., Comment: 9 pages, 3 figures, 2 tables

A promising approach to study condensed-matter systems is to simulate them on an engineered quantum platform. However, achieving the accuracy needed to outperform classical methods has been an outstanding challenge. Here, using eighteen superconducting qubits, we provide an experimental blueprint for an accurate condensed-matter simulator and demonstrate how to probe fundamental electronic properties. We benchmark the underlying method by reconstructing the single-particle band-structure of a one-dimensional wire. We demonstrate nearly complete mitigation of decoherence and readout errors and arrive at an accuracy in measuring energy eigenvalues of this wire with an error of ~0.01 rad, whereas typical energy scales are of order 1 rad. Insight into this unprecedented algorithm fidelity is gained by highlighting robust properties of a Fourier transform, including the ability to resolve eigenenergies with a statistical uncertainty of 1e-4 rad. Furthermore, we synthesize magnetic flux and disordered local potentials, two key tenets of a condensed-matter system. When sweeping the magnetic flux, we observe avoided level crossings in the spectrum, a detailed fingerprint of the spatial distribution of local disorder. Combining these methods, we reconstruct electronic properties of the eigenstates where we observe persistent currents and a strong suppression of conductance with added disorder. Our work describes an accurate method for quantum simulation and paves the way to study novel quantum materials with superconducting qubits.

We study the multi-phase feedback processes in the central ~3 kpc of the barred Sy 2 galaxy NGC 5643. We use observations of the cold molecular gas (ALMA CO(2-1)) and ionized gas (MUSE IFU). We study different regions along the outflow zone which extends out to ~2.3 kpc in the same direction (east-west) as the radio jet, as well as nuclear/circumnuclear regions in the host galaxy disk. The deprojected outflowing velocities of the cold molecular gas (median Vcentral~189 km s^-1) are generally lower than those of the outflowing ionized gas, which reach deprojected velocities of up to 750 km s^-1 close to the AGN, and their spatial profiles follow those of the ionized phase. This suggests that the outflowing molecular gas in the galaxy disk is being entrained by the AGN wind. We derive molecular and ionized outflow masses of ~5.2x10^7 Msun and 8.5x10^4 Msun and molecular and ionized outflow mass rates of ~51 Msun yr^-1 and 0.14 Msun yr^-1. Therefore, the molecular phase dominates the outflow mass and outflow mass rate, while the outflow kinetic power and momentum are similar in both phases. However, the wind momentum load for the molecular and ionized outflow phases are ~27-5 and <1, which suggests that the molecular phase is not momentum conserving while the ionized one most certainly is. The molecular gas content (~1.5x10^7 Msun) of the eastern spiral arm is approximately 50-70% of the content of the western one. We interpret this as destruction/clearing of the molecular gas produced by the AGN wind impacting in the host galaxy. The increase of the molecular phase momentum implies that part of the kinetic energy from the AGN wind is transmitted to the molecular outflow. This suggest that in Sy-like AGN such as NGC 5643, the radiative/quasar and the kinetic/radio AGN feedback modes coexist and may shape the host galaxies even at kpc-scales via both positive and (mild) negative feedback., Comment: 26 pages, 21 figures, 5 tables. Astronomy and Astrophysics, Accepted 2020 September 11, in press