In the Western Scheldt estuary, like in many estuaries, safe navigation, flood protection, and ecological targets require a balanced and sustainable sediment management. A thorough understanding of the morphodynamic functioning of the estuary and its response to changes in hydrodynamics (natural sediment transport) and large-scale interventions is imperative. This paper presents a detailed overview of over 65 years of morphological changes and a comprehensive sediment budget of the Western Scheldt estuary that is based on analysis of a unique series of frequent bathymetric surveys and historical data on human–sediment interactions of dredging, dredge disposal and sand mining. Solving the sediment budget reveals an annual sediment import of 2.2 million m3. The highest sediment imports occurred between 1980–1994 and 2005–2020 (2.9 and 3.7 million m3/year). A 1.3 million m3/year net export prevailed between 1994 and 2005. Natural variations in the hydrodynamics (e.g., tidal asymmetry and amplification) and sediment transports cannot explain the derived temporal variations in sediment import rates. Anthropogenic activities play a dominant role. Capital dredging of the main navigation channel has led to channel deepening and significantly increased dredge and disposal volumes. Disposal on tidal flats and in the secondary channel leads to a long-term storage of sand and, consequently, a local increase in bed level and a sand deficit in the remainder of the system that results in increased sediment imports. Large-scale disposal in the western part of the estuary can (temporarily) reverse the sediment exchange between the estuary and its mouth area, as observed between 1994 and 2005. Apparently, large-scale anthropogenic reallocation of sediment by dredging and/or disposal as part of navigation channel improvement, sand mining or nourishment essentially influences the morphological development of the Western Scheldt estuary. This reveals responsibilities as well as opportunities of sediment management for the Western Scheldt and similar estuaries worldwide.

REAP-2 is an interactive dose-response curve estimation tool for Robust and Efficient Assessment of drug Potency. It provides user-friendly dose-response curve estimation for in vitro studies and conducts statistical testing for model comparisons with a redesigned user interface. We also make a major update of the underlying estimation method with penalized beta regression, which demonstrates great reliability and accuracy in dose estimation and uncertainty quantification. In this note, we describe the method and implementation of REAP-2 with a highlight on potency estimation and drug comparison.

Traditional active flutter suppression controllers are designed based on model. However, as the aircraft becomes more and more powerful, the modeling of aeroelastic system becomes difficult and the model-free requirement of controller design becomes more and more urgent. The complexity of industrial processes has brought about massive operational data generated online. Aviation industry development has entered the era of big data. Breaking through the traditional theoretical framework, mining the correlation, evolution and dynamic characteristics of the system from the data is the inevitable choice to meet this demand. In this paper, a data-driven model-free controller is designed, which relies on ridge regression of the input and output variation at each operating point of the closed-loop controlled system to recursively derive the iterative format of the control signals and ensure the numerical stability of the signals. The controller can only use the real-time measurement of the system’s online input and output data for continuous correction, to achieve the purpose of flutter suppression. Then flutter suppression of a three-degree-of-freedom binary wing with a control surface is studied, and the superiority of model-free controller is demonstrated by comparing it with the optimal controller.

Applying a coastal-geoarchaeological approach, we synthesize stratigraphic, sedimentological, mollusk-zooarchaeological, and radiometric datasets from recent excavations and sediment coring at Harbor Key (8MA15)—a shell-terraformed Native mound complex within Tampa Bay, on the central peninsular Gulf Coast of Florida. We significantly revise the chronological understanding of the site and place it among the relatively few early civic-ceremonial centers in the region. Analyses of submound contexts revealed that the early first millennium mound center was constructed atop a platform of sand and ex situ cultural shell deposits that were reworked during ancient storm landfalls around 2000 BP. We situate Harbor Key within a seascape-scale stratigraphic and paleoenvironmental framework and show that the shellworks comprise an artificial barrier protecting the leeward estuary basin (and productive inshore wetlands) from high-energy conditions of the open bay and swells from the Gulf of Mexico. The sedimentary and archaeological records attest to the long-term history of morphodynamic interaction between coastal processes and Indigenous shell terraforming in the region and suggest that early first millennium mound building in Tampa Bay was tied to the recognition and reuse of antecedent shellworks and the persistent management of encompassing cultural seascapes.

The present work discusses the use of a weakly-supervised deep learning algorithm that reduces the cost of labelling pixel-level masks for complex radio galaxies with multiple components. The algorithm is trained on weak class-level labels of radio galaxies to get class activation maps (CAMs). The CAMs are further refined using an inter-pixel relations network (IRNet) to get instance segmentation masks over radio galaxies and the positions of their infrared hosts. We use data from the Australian Square Kilometre Array Pathfinder (ASKAP) telescope, specifically the Evolutionary Map of the Universe (EMU) Pilot Survey, which covered a sky area of 270 square degrees with an RMS sensitivity of 25–35 $\mu$Jy beam$^{-1}$. We demonstrate that weakly-supervised deep learning algorithms can achieve high accuracy in predicting pixel-level information, including masks for the extended radio emission encapsulating all galaxy components and the positions of the infrared host galaxies. We evaluate the performance of our method using mean Average Precision (mAP) across multiple classes at a standard intersection over union (IoU) threshold of 0.5. We show that the model achieves a mAP$_{50}$ of 67.5% and 76.8% for radio masks and infrared host positions, respectively. The network architecture can be found at the following link: https://github.com/Nikhel1/Gal-CAM

Over the past decade, transdiagnostic indicators in relation to neurobiological processes have provided extensive insight into youth’s risk for psychopathology. During development, exposure to childhood trauma and dysregulation (i.e., so-called AAA symptomology: anxiety, aggression, and attention problems) puts individuals at a disproportionate risk for developing psychopathology and altered network-level neural functioning. Evidence for the latter has emerged from resting-state fMRI studies linking mental health symptoms and aberrations in functional networks (e.g., cognitive control (CCN), default mode networks (DMN)) in youth, although few of these investigations have used longitudinal designs. Herein, we leveraged a three-year longitudinal study to identify whether traumatic exposures and concomitant dysregulation trigger changes in the developmental trajectories of resting-state functional networks involved in cognitive control (N = 190; 91 females; time 1 Mage = 11.81). Findings from latent growth curve analyses revealed that greater trauma exposure predicted increasing connectivity between the CCN and DMN across time. Greater levels of dysregulation predicted reductions in within-network connectivity in the CCN. These findings presented in typically developing youth corroborate connectivity patterns reported in clinical populations, suggesting there is predictive utility in using transdiagnostic indicators to forecast alterations in resting-state networks implicated in psychopathology.

In recent years, there has been significant momentum in applying deep learning (DL) to machine health monitoring (MHM). It has been widely claimed that DL methodologies are superior to more traditional techniques in this area. This paper aims to investigate this claim by analysing a real-world dataset of helicopter sensor faults provided by Airbus. Specifically, we will address the problem of machine sensor health unsupervised classification. In a 2019 worldwide competition hosted by Airbus, Fujitsu Systems Europe (FSE) won first prize by achieving an F1-score of 93% using a DL model based on generative adversarial networks (GAN). In another comprehensive study, various modified and existing image encoding methods were compared for the convolutional auto-encoder (CAE) model. The best classification result was achieved using the scalogram as the image encoding method, with an F1-score of 91%. In this paper, we use these two studies as benchmarks to compare with basic statistical analysis methods and the one-class supporting vector machine (SVM). Our comparative study demonstrates that while DL-based techniques have great potential, they are not always superior to traditional methods. We therefore recommend that all future published studies of applying DL methods to MHM include appropriately selected traditional reference methods, wherever possible.

The Australian SKA Pathfinder (ASKAP) is being used to undertake a campaign to rapidly survey the sky in three frequency bands across its operational spectral range. The first pass of the Rapid ASKAP Continuum Survey (RACS) at 887.5 MHz in the low band has already been completed, with images, visibility datasets, and catalogues made available to the wider astronomical community through the CSIRO ASKAP Science Data Archive (CASDA). This work presents details of the second observing pass in the mid band at 1367.5 MHz, RACS-mid, and associated data release comprising images and visibility datasets covering the whole sky south of $\delta_{\text{J2000}}=+49^\circ$. This data release incorporates selective peeling to reduce artefacts around bright sources, as well as accurately modelled primary beam responses. The Stokes I images reach a median noise of 198 $\mu$Jy PSF$^{-1}$ with a declination-dependent angular resolution of 8.1–47.5 arcsec that fills a niche in the existing ecosystem of large-area astronomical surveys. We also supply Stokes V images after application of a widefield leakage correction, with a median noise of 165 $\mu$Jy PSF$^{-1}$. We find the residual leakage of Stokes I into V to be $\lesssim 0.9$–$2.4$% over the survey. This initial RACS-mid data release will be complemented by a future release comprising catalogues of the survey region. As with other RACS data releases, data products from this release will be made available through CASDA.

The myriad applications of artificial intelligence (AI) by the private and public sectors have exploded in the public consciousness in the postpandemic period. However, researchers and businesses have been working on AI technology applications for decades, and in many ways, governments are rushing to catch up. This article presents an argument that the future of AI policy in the United States will be driven in large part by current and future state-level policy experiments. This argument is presented by drawing on scholarship surrounding federalism, regulatory fragmentation, and the effects of fragmentation on business and social equity. The article then presents the case of autonomous vehicle policy in the states to illustrate the degree of current fragmentation and considers the effects of layering new AI policies on top of existing rules surrounding privacy, licensing, and more. Following this consideration of existing research and its application of AI policy, the article presents a research agenda for leveraging state differences to study the effects of AI policy and develop a cohesive framework for governing AI.

Many Aboriginal Australian communities are undergoing language shift from traditional Indigenous languages to contact varieties such as Kriol, an English-lexified Creole. Kriol is reportedly characterised by lexical items with highly variable phonological specifications, and variable implementation of voicing and manner contrasts in obstruents (Sandefur, 1986). A language, such as Kriol, characterised by this unusual degree of variability presents Kriol-acquiring children with a potentially difficult language-learning task, and one which challenges the prevalent theories of acquisition. To examine stop consonant acquisition in this unusual language environment, we present a study of Kriol stop and affricate production, followed by a mispronunciation detection study, with Kriol-speaking children (ages 4-7) from a Northern Territory community where Kriol is the lingua franca. In contrast to previous claims, the results suggest that Kriol-speaking children acquire a stable phonology and lexemes with canonical phonemic specifications, and that English experience would not appear to induce this stability.

In this paper, an adaptive neural output-constrained control algorithm is proposed for a class of non-affine kinetic kill vehicle (KKV) systems. The key point is that the non-affine control law can be designed and the output of the KKV system conform to the output limit with the aid of the proposed method. Due to the aerodynamic moments, the actual control torque is non-affine, which can be addressed by introducing an integral process to the design of the controller. Besides, in order to improve the control precision, a nonlinear mapping is put forward so that the output constraint can be transformed to the constraint of the introduced dynamic signal that can be simply achieved. From the simulation results it can be concluded that the states of the KKV system can track the desired trajectories in spite of different working conditions and the control precision is higher compared with other control methods.

With the advent of deep, all-sky radio surveys, the need for ancillary data to make the most of the new, high-quality radio data from surveys like the Evolutionary Map of the Universe (EMU), GaLactic and Extragalactic All-sky Murchison Widefield Array survey eXtended, Very Large Array Sky Survey, and LOFAR Two-metre Sky Survey is growing rapidly. Radio surveys produce significant numbers of Active Galactic Nuclei (AGNs) and have a significantly higher average redshift when compared with optical and infrared all-sky surveys. Thus, traditional methods of estimating redshift are challenged, with spectroscopic surveys not reaching the redshift depth of radio surveys, and AGNs making it difficult for template fitting methods to accurately model the source. Machine Learning (ML) methods have been used, but efforts have typically been directed towards optically selected samples, or samples at significantly lower redshift than expected from upcoming radio surveys. This work compiles and homogenises a radio-selected dataset from both the northern hemisphere (making use of Sloan Digital Sky Survey optical photometry) and southern hemisphere (making use of Dark Energy Survey optical photometry). We then test commonly used ML algorithms such as k-Nearest Neighbours (kNN), Random Forest, ANNz, and GPz on this monolithic radio-selected sample. We show that kNN has the lowest percentage of catastrophic outliers, providing the best match for the majority of science cases in the EMU survey. We note that the wider redshift range of the combined dataset used allows for estimation of sources up to $z = 3$ before random scatter begins to dominate. When binning the data into redshift bins and treating the problem as a classification problem, we are able to correctly identify $\approx$76% of the highest redshift sources—sources at redshift $z > 2.51$—as being in either the highest bin ($z > 2.51$) or second highest ($z = 2.25$).