ORCID Profile
0000-0002-8698-7605
Current Organisations
Northumbria University
,
GSI Helmholtzzentrum für Schwerionenforschung GmbH
Does something not look right? The information on this page has been harvested from data sources that may not be up to date. We continue to work with information providers to improve coverage and quality. To report an issue, use the Feedback Form.
Publisher: MDPI AG
Date: 10-07-2022
DOI: 10.3390/S22145161
Abstract: Underwater fish monitoring is the one of the most challenging problems for efficiently feeding and harvesting fish, while still being environmentally friendly. The proposed 2D computer vision method is aimed at non-intrusively estimating the weight of Tilapia fish in turbid water environments. Additionally, the proposed method avoids the issue of using high-cost stereo cameras and instead uses only a low-cost video camera to observe the underwater life through a single channel recording. An in-house curated Tilapia-image dataset and Tilapia-file dataset with various ages of Tilapia are used. The proposed method consists of a Tilapia detection step and Tilapia weight-estimation step. A Mask Recurrent-Convolutional Neural Network model is first trained for detecting and extracting the image dimensions (i.e., in terms of image pixels) of the fish. Secondly, is the Tilapia weight-estimation step, wherein the proposed method estimates the depth of the fish in the tanks and then converts the Tilapia's extracted image dimensions from pixels to centimeters. Subsequently, the Tilapia's weight is estimated by a trained model based on regression learning. Linear regression, random forest regression, and support vector regression have been developed to determine the best models for weight estimation. The achieved experimental results have demonstrated that the proposed method yields a Mean Absolute Error of 42.54 g, R2 of 0.70, and an average weight error of 30.30 (±23.09) grams in a turbid water environment, respectively, which show the practicality of the proposed framework.
Publisher: Copernicus GmbH
Date: 15-05-2023
DOI: 10.5194/EGUSPHERE-EGU23-15264
Abstract: Through their role in buttressing upstream ice flow, Antarctic ice shelves play an important part in regulating future sea level change. Reduction in ice-shelf buttressing caused by increased ocean-induced melt along their undersides is now understood to be one of the key drivers of ice loss from the Antarctic Ice Sheet. However, despite the importance of this forcing mechanism, most ice-sheet simulations currently rely on simple melt-parametrisations of this ocean-driven process since a fully coupled ice-ocean modelling framework is prohibitively computationally expensive. Here, we provide an alternative approach that can capture the greatly improved physical description of this process provided by large-scale ocean-circulation models over currently employed melt-parameterisations, but with trivial computational expense.& This new method brings together deep learning and physical modelling to develop a deep neural network framework, MELTNET, that can emulate ocean model predictions of sub-ice shelf melt rates. We train MELTNET on synthetic geometries, using the NEMO ocean model as a ground-truth in lieu of observations to provide melt rates both for training and to evaluate the performance of the trained network. We show that MELTNET can accurately predict melt rates for a wide range of complex synthetic geometries, with a normalized root mean squared error of 0.11m/yr compared to the ocean model. MELTNET calculates melt rates several orders of magnitude faster than the ocean model and outperforms more traditional parameterisations for 96% of geometries tested. Furthermore, we find MELTNET's melt rate estimates show sensitivity to established physical relationships such as changes in thermal forcing and ice shelf slope. This study demonstrates the potential for a deep learning framework to calculate melt rates with almost no computational expense, that could in the future be used in conjunction with an ice sheet model to provide predictions for large-scale ice sheet models.
Publisher: IEEE
Date: 11-2017
Publisher: EDP Sciences
Date: 12-2022
DOI: 10.1051/0004-6361/202244134
Abstract: We present the first results of a comprehensive supernova (SN) radiative-transfer (RT) code-comparison initiative (StaNdaRT), where the emission from the same set of standardised test models is simulated by currently used RT codes. We ran a total of ten codes on a set of four benchmark ejecta models of Type Ia SNe. We consider two sub-Chandrasekhar-mass ( M tot = 1.0 M ⊙ ) toy models with analytic density and composition profiles and two Chandrasekhar-mass delayed-detonation models that are outcomes of hydrodynamical simulations. We adopt spherical symmetry for all four models. The results of the different codes, including the light curves, spectra, and the evolution of several physical properties as a function of radius and time are provided in electronic form in a standard format via a public repository. We also include the detailed test model profiles and several Python scripts for accessing and presenting the input and output files. We also provide the code used to generate the toy models studied here. In this paper, we describe the test models, radiative-transfer codes, and output formats in detail, and provide access to the repository. We present ex le results of several key diagnostic features.
Publisher: Oxford University Press (OUP)
Date: 31-12-2019
Abstract: We extend the range of validity of the artis 3D radiative transfer code up to hundreds of days after explosion, when Type Ia supernovae (SNe Ia) are in their nebular phase. To achieve this, we add a non-local thermodynamic equilibrium population and ionization solver, a new multifrequency radiation field model, and a new atomic data set with forbidden transitions. We treat collisions with non-thermal leptons resulting from nuclear decays to account for their contribution to excitation, ionization, and heating. We validate our method with a variety of tests including comparing our synthetic nebular spectra for the well-known one-dimensional W7 model with the results of other studies. As an illustrative application of the code, we present synthetic nebular spectra for the detonation of a sub-Chandrasekhar white dwarf (WD) in which the possible effects of gravitational settling of 22Ne prior to explosion have been explored. Specifically, we compare synthetic nebular spectra for a 1.06 M⊙ WD model obtained when 5.5 Gyr of very efficient settling is assumed to a similar model without settling. We find that this degree of 22Ne settling has only a modest effect on the resulting nebular spectra due to increased 58Ni abundance. Due to the high ionization in sub-Chandrasekhar models, the nebular [Ni ii] emission remains negligible, while the [Ni iii] line strengths are increased and the overall ionization balance is slightly lowered in the model with 22Ne settling. In common with previous studies of sub-Chandrasekhar models at nebular epochs, these models overproduce [Fe iii] emission relative to [Fe ii] in comparison to observations of normal SNe Ia.
Publisher: MDPI AG
Date: 05-07-2021
DOI: 10.3390/S21134620
Abstract: Spatial susceptible landslide prediction is the one of the most challenging research areas which essentially concerns the safety of inhabitants. The novel geographic information web (GIW) application is proposed for dynamically predicting landslide risk in Chiang Rai, Thailand. The automated GIW system is coordinated between machine learning technologies, web technologies, and application programming interfaces (APIs). The new bidirectional long short-term memory (Bi-LSTM) algorithm is presented to forecast landslides. The proposed algorithm consists of 3 major steps, the first of which is the construction of a landslide dataset by using Quantum GIS (QGIS). The second step is to generate the landslide-risk model based on machine learning approaches. Finally, the automated landslide-risk visualization illustrates the likelihood of landslide via Google Maps on the website. Four static factors are considered for landslide-risk prediction, namely, land cover, soil properties, elevation and slope, and a single dynamic factor i.e., precipitation. Data are collected to construct a geospatial landslide database which comprises three historical landslide locations—Phu Chifa at Thoeng District, Ban Pha Duea at Mae Salong Nai, and Mai Salong Nok in Mae Fa Luang District, Chiang Rai, Thailand. Data collection is achieved using QGIS software to interpolate contour, elevation, slope degree and land cover from the Google satellite images, aerial and site survey photographs while the physiographic and rock type are on-site surveyed by experts. The state-of-the-art machine learning models have been trained i.e., linear regression (LR), artificial neural network (ANN), LSTM, and Bi-LSTM. Ablation studies have been conducted to determine the optimal parameters setting for each model. An enhancement method based on two-stage classifications has been presented to improve the landslide prediction of LSTM and Bi-LSTM models. The landslide-risk prediction performances of these models are subsequently evaluated using real-time dataset and it is shown that Bi-LSTM with Random Forest (Bi-LSTM-RF) yields the best prediction performance. Bi-LSTM-RF model has improved the landslide-risk predicting performance over LR, ANNs, LSTM, and Bi-LSTM in terms of the area under the receiver characteristic operator (AUC) scores by 0.42, 0.27, 0.46, and 0.47, respectively. Finally, an automated web GIS has been developed and it consists of software components including the trained models, rainfall API, Google API, and geodatabase. All components have been interfaced together via JavaScript and Node.js tool.
Publisher: Copernicus GmbH
Date: 07-02-2023
Abstract: Abstract. Through their role in buttressing upstream ice flow, Antarctic ice shelves play an important part in regulating future sea-level change. Reduction in ice-shelf buttressing caused by increased ocean-induced melt along their undersides is now understood to be one of the key drivers of ice loss from the Antarctic ice sheet. However, despite the importance of this forcing mechanism, most ice-sheet simulations currently rely on simple melt parameterisations of this ocean-driven process since a fully coupled ice–ocean modelling framework is prohibitively computationally expensive. Here, we provide an alternative approach that is able to capture the greatly improved physical description of this process provided by large-scale ocean-circulation models over currently employed melt parameterisations but with trivial computational expense. This new method brings together deep learning and physical modelling to develop a deep neural network framework, MELTNET, that can emulate ocean model predictions of sub-ice-shelf melt rates. We train MELTNET on synthetic geometries, using the NEMO ocean model as a ground truth in lieu of observations to provide melt rates both for training and for evaluation of the performance of the trained network. We show that MELTNET can accurately predict melt rates for a wide range of complex synthetic geometries, with a normalised root mean squared error of 0.11 m yr−1 compared to the ocean model. MELTNET calculates melt rates several orders of magnitude faster than the ocean model and outperforms more traditional parameterisations for 96 % of geometries tested. Furthermore, we find MELTNET's melt rate estimates show sensitivity to established physical relationships such as changes in thermal forcing and ice-shelf slope. This study demonstrates the potential for a deep learning framework to calculate melt rates with almost no computational expense, which could in the future be used in conjunction with an ice sheet model to provide predictions for large-scale ice sheet models.
Publisher: Oxford University Press (OUP)
Date: 24-02-2023
Abstract: The detection of GW170817 and the accompanying electromagnetic counterpart, AT2017gfo, have provided an important set of observational constraints for theoretical models of neutron star mergers, nucleosynthesis, and radiative transfer for kilonovae. We apply the three-dimensional (3D) Monte Carlo radiative transfer code artis to produce synthetic light curves of the dynamical ejecta from a neutron star merger, which has been modelled with 3D smooth particle hydrodynamics and included neutrino interactions. Nucleosynthesis calculations provide the energy released from radioactive decays of r-process nuclei, and radiation transport is performed using grey opacities given as functions of the electron fraction. We present line-of-sight-dependent bolometric light curves, and find the emission along polar lines of sight to be up to a factor of ∼2 brighter than that along equatorial lines of sight. Instead of a distinct emission peak, our bolometric light curve exhibits a monotonic decline, characterized by a shoulder at the time when the bulk ejecta becomes optically thin. We show approximate band light curves based on radiation temperatures and compare these to the observations of AT2017gfo. We find that the rapidly declining temperatures lead to a blue to red colour evolution similar to that shown by AT2017gfo. We also investigate the impact of an additional, spherically symmetric secular ejecta component, and we find that the early light curve remains nearly unaffected, while after about $1\\,$ d the emission is strongly enhanced and dominated by the secular ejecta, leading to the shift of the shoulder from ∼1–2 to 6–10 d.
Publisher: Oxford University Press (OUP)
Date: 05-04-2022
Abstract: The nebular spectra of Type Ia supernovae (⪆100 d after explosion) consist mainly of emission lines from singly and doubly ionized Fe-group nuclei. However, theoretical models for many scenarios predict that non-thermal ionization leads to multiply ionized species whose recombination photons ionize and deplete Fe+, resulting in negligible [Fe ii] emission. We investigate a method to determine the collisional excitation conditions from [Fe ii] line ratios independently from the ionization state and find that it cannot be applied to highly ionized models due to the influence of recombination cascades on Fe+ level populations. When the ionization state is artificially lowered, the line ratios (and excitation conditions) are too similar to distinguish between explosion scenarios. We investigate changes to the treatment of non-thermal energy deposition as a way to reconcile overionized theoretical models with observations and find that a simple work function approximation provides closer agreement with the data for sub-Mch models than a detailed Spencer–Fano treatment with widely used cross-section data. To quantify the magnitude of additional heating processes that would be required to sufficiently reduce ionization from fast leptons, we artificially boost the rate of energy loss to free electrons. We find that the equivalent of as much as an eight times increase to the plasma loss rate would be needed to reconcile the sub-Mch model with observed spectra. Future studies could distinguish between reductions in the non-thermal ionization rates and increased recombination rates, such as by clumping.
Publisher: Oxford University Press (OUP)
Date: 19-07-2023
Abstract: The double detonation is a widely discussed mechanism to explain Type Ia supernovae from explosions of sub-Chandrasekhar mass white dwarfs. In this scenario, a helium detonation is ignited in a surface helium shell on a carbon/oxygen white dwarf, which leads to a secondary carbon detonation. Explosion simulations predict high abundances of unburnt helium in the ejecta, however, radiative transfer simulations have not been able to fully address whether helium spectral features would form. This is because helium can not be sufficiently excited to form spectral features by thermal processes, but can be excited by collisions with non-thermal electrons, which most studies have neglected. We carry out a full non-local thermodynamic equilibrium radiative transfer simulation for an instance of a double detonation explosion model, and include a non-thermal treatment of fast electrons. We find a clear He i λ10830 feature which is strongest in the first few days after explosion and becomes weaker with time. Initially this feature is blended with the Mg ii λ10927 feature but over time separates to form a secondary feature to the blue wing of the Mg ii λ10927 feature. We compare our simulation to observations of iPTF13ebh, which showed a similar feature to the blue wing of the Mg ii λ10927 feature, previously identified as C i. Our simulation shows a good match to the evolution of this feature and we identify it as high velocity He i λ10830. This suggests that He i λ10830 could be a signature of the double detonation scenario.
Publisher: Oxford University Press (OUP)
Date: 09-11-2022
Abstract: The progenitor systems and explosion mechanism of Type Ia supernovae are still unknown. Currently favoured progenitors include double-degenerate systems consisting of two carbon-oxygen white dwarfs with thin helium shells. In the double-detonation scenario, violent accretion leads to a helium detonation on the more massive primary white dwarf that turns into a carbon detonation in its core and explodes it. We investigate the fate of the secondary white dwarf, focusing on changes of the ejecta and observables of the explosion if the secondary explodes as well rather than survives. We simulate a binary system of a $1.05\\, \\mathrm{M_\\odot }$ and a $0.7\\, \\mathrm{M_\\odot }$ carbon-oxygen white dwarf with $0.03\\, \\mathrm{M_\\odot }$ helium shells each. We follow the system self-consistently from inspiral to ignition, through the explosion, to synthetic observables. We confirm that the primary white dwarf explodes self-consistently. The helium detonation around the secondary white dwarf, however, fails to ignite a carbon detonation. We restart the simulation igniting the carbon detonation in the secondary white dwarf by hand and compare the ejecta and observables of both explosions. We find that the outer ejecta at $v~\\gt ~15\\, 000$ km s−1 are indistinguishable. Light curves and spectra are very similar until $\\sim ~40 \\ \\mathrm{d}$ after explosion and the ejecta are much more spherical than violent merger models. The inner ejecta differ significantly slowing down the decline rate of the bolometric light curve after maximum of the model with a secondary explosion by ∼20 per cent. We expect future synthetic 3D nebular spectra to confirm or rule out either model.
Location: United Kingdom of Great Britain and Northern Ireland
Location: United Kingdom of Great Britain and Northern Ireland
Location: United Kingdom of Great Britain and Northern Ireland
No related grants have been discovered for Wai Lok Woo.