Spatial Refinement Research Articles

Improved real-time three-dimensional stereo matching with local consistency

Existing stereo matching algorithms are unable to meet the real-time and high-performance dual requirements in practical applications. To address this problem, a novel stereo network is proposed, which utilizes the prior of local disparity consistency to improve the performance of real-time disparity estimation. Based on the initial disparity estimation by the light-weight pyramid matching network, novel spatial consistency refinement (SCR) module and time consistency refinement (TCR) module are designed for further disparity refinement. SCR module propagates the neighborhood high-confidence predictions of sparse sampling to unreliable regions for disparity refinement. A single-layer Dynamic Local Filter (DLF) is designed to realize the content-adaptive propagation, which effectively improves the disparity quality without significantly increasing the burden of computation and memory. For real-time disparity estimation of consecutive frames, novel TCR module is further proposed to refine the disparity estimation based on the time local consistency of disparity. The proposed method is evaluated on the Scene Flow and KITTI 2015 datasets with comprehensive experiments. Experimental results demonstrated that our method can achieve high-accuracy disparity estimation and real-time running speed of over 40 FPS, which significantly outperforms the compared networks with similar runtimes.

VITAS: A multi-purpose simulation code for the solution of neutron transport problems based on variational nodal methods

Existing neutronic whole-core simulation codes are problem-oriented. With the advancement of computer technology and the development of complicated micro-nuclear reactors, there is a pressing need to create solution methods with broad applicability and the ability to accommodate more complex, high-fidelity, unstructured geometries. This necessity motivates the development of a neutron transport code that can adapt to different types of nuclear reactors. This work describes the code design and theoretical model of VITAS, a multi-purpose simulation code for handling steady-state and time-dependent neutron transport problems. This code incorporates algorithms and models inside a unified variational nodal framework based on the second-order neutron transport equation. The most recent version of VITAS employs the isotropic scattering approximation and permits arbitrary spatial and angular refinement. VITAS is compatible with Cartesian geometry, hexagonal geometry, triangular geometry, and Cartesian geometry with finite element subdivisions in the radial direction (Cartesian-FE). The three-dimensional modeling can be performed by axially extruding the two-dimensional mesh. Users can choose the mesh type and expansion order on-demand for optimal computational efficiency.In this paper, VITAS is validated using benchmark problems, i.e., the TAKEDA3 benchmark and the transient LMW benchmark with Cartesian homogenized geometry, the C5G7 and C5G7-TD exercises with Cartesian-FE geometry, the TAKEDA4 benchmark and the Buckner and Stewart benchmark with hexagonal-z homogenized geometry, and the Dodds benchmark with cylindrical-z geometry described by triangular-z unstructured meshes. The space-angle truncation is extensively examined to demonstrate the asymptotic convergence behavior with p- and h-refinement. The results indicate that VITAS has the potential to become a versatile toolset for solving a variety of nuclear reactor problems.

SPROUT: spectral sparsification helps restore the spatial structure at single-cell resolution.

Single-cell RNA sequencing thoroughly quantifies the individual cell transcriptomes but renounces the spatial structure. Conversely, recently emerged spatial transcriptomics technologies capture the cellular spatial structure but skimp cell or gene resolutions. Ligand-receptor interactions reveal the potential of cell proximity since they are spatially constrained. Cell–cell affinity values estimated by ligand–receptor interaction can partially represent the structure of cells but falsely include the pseudo affinities between distant or indirectly interacting cells. Here, we develop a software package, SPROUT, to reconstruct the single-cell resolution spatial structure from the transcriptomics data through diminished pseudo ligand–receptor affinities. For spatial data, SPROUT first curates the representative single-cell profiles for each spatial spot from a candidate library, then reduces the pseudo affinities in the intercellular affinity matrix by partial correlation, spectral graph sparsification, and spatial coordinates refinement. SPROUT embeds the estimated interactions into a low-dimensional space with the cross-entropy objective to restore the intercellular structures, which facilitates the discovery of dominant ligand–receptor pairs between neighboring cells at single-cell resolution. SPROUT reconstructed structures achieved shape Pearson correlations ranging from 0.91 to 0.97 on the mouse hippocampus and human organ tumor microenvironment datasets. Furthermore, SPROUT can solely de novo reconstruct the structures at single-cell resolution, i.e., reaching the cell-type proximity correlations of 0.68 and 0.89 between reconstructed and immunohistochemistry-informed spatial structures on a human developing heart dataset and a tumor microenvironment dataset, respectively.

Integrating weather observations and local-climate-zone-based landscape patterns for regional hourly air temperature mapping using machine learning

Air temperature is a crucial variable of urban meteorology and is essential to many urban environments, urban climate and climate-change-related studies. However, due to the limited observational records of air temperature and the complex urban morphology and environment, it might not be easy to map the hourly air temperature with a fine resolution at the surface level within and around cities via conventional methods. Thus, this study employed machine learning (ML) algorithms and meteorological and landscape data to develop hourly air temperature mapping techniques and methods at the 1-km resolution over a multi-year warm seasons period. Guangdong Province, China was selected for the case study. Random forest algorithm was employed for the hourly air temperature mapping. The validation results showed that the hourly air temperature maps exhibit good accuracy from 2008 to 2019, with mean R2, root mean square error (RMSE) and mean absolute error (MAE) values of 0.8001, 1.4821 °C and 1.0872 °C, respectively. The importance assessment of the driving factors showed that meteorological factors, especially relative humidity, contributed the most to the air temperature mapping. Simultaneously, landscape factors also played a non-negligible role. Further analysis revealed that the maps steadily maintained high accuracy at nighttime (20:00–7:00), which is essential for investigating nighttime urban climate conditions, especially the urban heat island effect. Moreover, a correlation existed between the nighttime air temperature changes and urban morphology represented by the local climate zones. Air temperatures tended to fall more slowly in the core of metropolitan areas than in the urban fringe. Using ML, this study reliably improves the spatial refinement of hourly air temperature mapping and reveals the spatially explicit air temperature patterns in and around cities at different times in a day during the warm seasons. Moreover, it provides a novel valuable and reliable dataset for air-temperature-related implementation and studies.

Guided-Pix2Pix+: End-to-end spatial and color refinement network for image dehazing

Inevitable hazy contamination degrades the visibility of images, and the resulting haze removal is one of the essential prerequisites for image processing and computer vision tasks. We proposed an end-to-end dehazing network, referred to as Guided-Pix2Pix, to estimate spatially refined transmission maps and dehaze via the physical scattering equation. The remaining enhancement of color contrast, ill-posed adversarial training, and redundant backbone, however, should be thoroughly investigated, as required in the prospect of Guided-Pix2Pix. In this paper, we inherit the end-to-end structure of Guided-Pix2Pix and accordingly propose Guided-Pix2Pix+ as an update, which concatenates transmission estimation and refinement, physical scattering equation-based dehazing, together with color refinement, to achieve haze removal in a one-stage way. Specifically, we make use of the pretrained instance of EfficientNetB0 to estimate coarse-grained transmission maps and concatenate a guided filter layer to perform spatial refinement for the incoming transmission maps. Restored by the physical scattering equation, color refinement of dehazed proposals is finally performed via the standardization and clipping of pixel intensities. All the operations are differentiable, making it possible to achieve end-to-end, tight training. Furthermore, adversarial and perceptual losses are employed to regulate the performance of our model, giving rise to structurally similar but photo-realistic dehazed proposals. Extensive experiments confirm that our Guided-Pix2Pix+ yields dehazed proposals with fine-grained spatial refinement and relatively effective color contrast, compared to our previous Guided-Pix2Pix, the baseline, and advanced dehazing methods. The source code is currently available at https://github.com/92xianshen/guided-pix2pixplus.

Prisoners of Scale: Downscaling Community Resilience Measurements for Enhanced Use

As improved data availability and disaster resilience knowledge help progress community resilience quantification schemes, spatial refinements of the associated empirical methods become increasingly crucial. Most existing empirically based indicators in the U.S. use county-level data, while qualitatively based schemes are more locally focused. The process of replicating resilience indices at a sub-county level includes a comprehensive study of existing databases, an evaluation of their conceptual relevance in the framework of resilience capitals, and finally, an analysis of the statistical significance and internal consistency of the developed metrics. Using the U.S. Gulf Coast region as a test case, this paper demonstrates the construction of a census tract-level resilience index based on BRIC (Baseline Resilience Indicators for Communities), called TBRIC. The final TBRIC construct gathers 65 variables into six resilience capitals: social, economic, community, institutional, infrastructural, and environmental. The statistical results of tract- and county-level BRIC comparisons highlight levels of divergence and convergence between the two measurement schemes and find higher reliability for the fine-scale results.

Stereo Refinement Dehazing Network

The performance of stereo vision tasks degrades when haze exists in the input stereo image pair. Independently applying single image dehazing algorithm on left and right images is not optimal. To overcome the problem, we propose an effective framework, called SRDNet, for simultaneously dehazing stereo images. The main idea of SRDNet is to make full use of the stereo information from cross views improving dehazing performance. It does not explicitly employ the disparity estimation and the correlation matrix. SRDNet comprises two parts: a weight-sharing coarse dehazing network (WSCDN) and a guided separated refinement network (GSRN). The WSCDN is utilized to predict a coarse dehazed image pair. Then the GSRN is introduced to predict the residues for different views by extracting the fused information of cross views and separating the features of different views with a guided channel and spatial refinement module. The residues are added to the coarse dehazed pair so as to make refinement and remove the remained haze. Experimental results demonstrate that our proposed SRDNet surpasses previous image dehazing methods by a significant margin both quantitatively and qualitatively. Moreover, our SRDNet could be a preprocessing step of the stereo-based 3D object detection and boosts the 3D detection accuracy in hazy scenes.

Pyramid frequency network with spatial attention residual refinement module for monocular depth estimation

Deep-learning-based approaches to depth estimation are rapidly advancing, offering superior performance over existing methods. To estimate the depth in real-world scenarios, depth estimation models require the robustness of various noise environments. In this work, a Pyramid Frequency Network(PFN) with Spatial Attention Residual Refinement Module(SARRM) is proposed to deal with the weak robustness of existing deep-learning methods. To reconstruct depth maps with accurate details, the SARRM constructs a residual fusion method with an attention mechanism to refine the blur depth. The frequency division strategy is designed, and the frequency pyramid network is developed to extract features from multiple frequency bands. With the frequency strategy, PFN achieves better visual accuracy than state-of-the-art methods in both indoor and outdoor scenes on Make3D, KITTI depth, and NYUv2 datasets. Additional experiments on the noisy NYUv2 dataset demonstrate that PFN is more reliable than existing deep-learning methods in high-noise scenes.

Water vapor monitoring with IGS RTS and GPT3/VMF3 functions over Turkey

Precise Point Positioning (PPP) has been proven as a cost-effective and efficient technique for monitoring the precipitable water vapor (PWV), which plays a critical role in weather forecasting. The recent developments of the International Global Navigation Satellite Systems Service (IGS) Real-Time Pilot Project (RTPP) ease the difficulties of real-time determination of PWV. In this study, the quality of the real-time PPP-derived PWVs and their correlation with precipitation estimates from different global products are investigated. The accuracies of the orbits and clocks derived from the IGS02 streams are evaluated and their performance outperforms the predicted part of IGS Ultra-rapid (IGU) orbits and clocks. A year of GNSS observations from 9 stations located in Turkey are processed to estimate the PWV in the real-time. Since there are no stations to collect meteorological data near GNSS stations, the real-time solutions employ empirical Global Pressure and Temperature 3 (GPT3) and its fully consistent discrete Vienna Mapping Function 3 (VMF3) to model the troposphere. The performance of the PWV estimates is assessed by comparing with Radiosonde-derived PWV values and post-processed PPP-derived PWV estimates employing the Center for Orbit Determination in Europe (CODE) final orbits and clocks. At most sites, the root-mean-square (RMS) differences of the estimated PWVs in comparison with the radiosonde-derived PWVs are less than 2.4 mm with correlation coefficients greater than 0.90. The RMS differences between RT and post-processed PPP PWV estimates are found to be less than 1.2 mm. The estimated RT zenith total delays (ZTDs) at two IGS sites agree well with the ZTDs provided by the IGS final products. A method of precipitation monitoring is proposed based on the daily standard deviation (SD) of the PWV estimates. Experimental analyses of the proposal are undertaken by comparing the daily SDs of the PWV estimates and the daily accumulated precipitation estimates from the global precipitation grids produced from rain gauges, satellite images, and reanalyses product sets. The results show that the daily variation of the PWVs and the high precipitation estimates present a good agreement for some days but not always. The PWVs overestimate and underestimate a significant number of events. Spatial refinement of GPT3 grids and gridded precipitation products with higher sampling over Turkey may offer better results.

Effects of Spatial Training Paradigms on Auditory Spatial Refinement in Normal-Hearing Listeners: A Comparative Study.

Background and ObjectivesThis study compared the effectiveness of two spatial training programs using real and virtual sound sources in refining spatial acuity skills in listeners with normal hearing.Subjects and MethodsThe study was conducted on two groups of 10 participants each; groups I and II underwent spatial training using real and virtual sound sources, respectively. The study was conducted in three phases: pre-training, training, and post-training phases. At the pre- and post-training phases, the spatial acuity of the participants was measured using real sound sources through the localization test, and virtual sound sources through the virtual acoustic space identification (VASI) test. The thresholds of interaural time difference (ITD) and interaural level difference (ILD) were also measured. In the training phase, Group I participants underwent localization training using loudspeakers in free field, while participants in Group II were subjected to virtual acoustic space (VAS) training using virtual sound sources from headphones. Both the training methods consisted of 5-8 sessions (20 min each) of systematically presented stimuli graded according to duration and back attenuation (for real source training) or number of VAS locations (for virtual source training).ResultsResults of independent t-scores comparing the spatial learning scores (pre vs. post-training) for each measure showed differences in performance between the two groups. Group II performed better than Group I on the VASI test, while the Group I out-performed Group II on the ITD. Both groups improved equally on the localization test and ILD.ConclusionsBased on the present findings, we recommend the use of VAS training as it has practical implications due to its cost effectiveness, need for minimal equipment, and end user usefulness.

Wind turbine wake computation with the ST-VMS method and isogeometric discretization: Directional preference in spatial refinement

Wind turbine wake computation with the ST-VMS method and isogeometric discretization: Directional preference in spatial refinement

HLocalExp-CM: confidence map by hierarchical local expansion moves for accurate stereo matching

We present a stereo matching approach referred to as HLocalExp-CM by exploiting the hierarchical local contextual information and a confidence map based on a new grid structure. The proposed approach preserves fine depth edges and extracts accurate disparities in weak texture, textureless, and repeated texture regions. The proposed approach adopts a two-stage optimization strategy. In the framework of first stage, a multiresolution cost aggregation is minimized to reduce the search space of the disparity plane of each pixel. The second stage iteratively optimizes the confidence map and a global energy function to progressively improve the disparity accuracy for each pixel. The confidence map is estimated through classifying the pixels into distinctive and ambiguous ones by computing the decreasing rate of the multiresolution cost aggregation and then performs a spatial propagation and plane refinement for the update of the disparity of each pixel, thereby successfully eliminating the ambiguity of nondistinctive pixels. The global energy function based on a pairwise Markov random field uses cross-scale cost aggregation for taking advantage of context information of objects in different scenarios on local grid regions, which is different from the deep learning technique uses convolution layers extracting the context information. The proposed approach is evaluated on Middlebury benchmark V3, and is ranked first based on “bad 2.0 all metric,” a widely used criterion for the evaluation of stereo images, while the eighth place on “bad 2.0 nonocc metric” (recorded on July 24, 2021).

Characterizing and structuring urban GIS data for housing stock energy modelling and retrofitting

Addressing the gap between available urban building data, energy performance evaluation, and spatial distribution of end-users is a growing concern of municipalities to support energy planning and urban retrofitting towards low carbon emission. These days, many existing urban building energy models (UBEM) use the non-geo-specified 3D building stock geometries or the digital twins of cities as their primary input for energy simulation. The digital twins of cities or the built environment is an accurate scale virtual representation (geo-specified geometry and information) of city objects enabled to be paired with dynamic predictive models of their energy performance. The use of built environment digital twins for their energy assessment works well for those cities whose digital twin datasets are readily available for public or private use. In the cities without digital twin datasets, their energy assessment is limited by tools that use 2D building stock footprints as their primary input or apply a top-down urban energy simulation approach. This gap motivated the current research to identify an extended UBEM data preparation workflow focused on generating city digital twins using accessible GIS datasets as well as national and local data sources. The workflow is majorly developed to deal with GIS data processing utilizing multilevel spatial data integration and refinement to fill in the recognized inconsistencies of building databases. Then, procedural 3D city modelling and an automated transformation to a semantic, CityGML format are successfully provided to generate digital twin of the study area in an open data model. The archetype modelling method is also employed to create a set of customized housing energy profiles to support tailored bottom-up UBEM and retrofitting scenarios relying on the simulation platform Insel4Cities, under development at Concordia University. The workflow was practically carried out on a case study, Kelowna city in British Columbia (BC)/Canada, to validate the district heating load. The average simulated heating energy use intensity showed a deviation of less than 2.5% from the BC measured data. As an example of workflow usability, the developed model was also utilized to assess a retrofit scenario, combining a decentralized heat pump and photovoltaic system, to present the district's potential for net-zero action planning.

Modeling outdoor thermal comfort along cycling routes at varying levels of physical accuracy to predict bike ridership in Cambridge, MA

The Universal Thermal Climate Index (UTCI) has been linked to outdoor activity patterns and used to evaluate the effectiveness of urban interventions to improve thermal comfort. This study investigates how simulating the urban environment at increasing levels of physical accuracy impacts UTCI values along three cycling routes in Cambridge, Massachusetts. Baseline UTCI values are estimated using a local weather file, and the following increments in physical accuracy are considered: wind-scaling, shading from buildings, shading and cooling from trees, computational fluid dynamics simulations for wind speeds, and simulated surface temperatures. With bike ridership data from Bluebikes, Boston's bike-sharing program, the relationship between bike ridership patterns and UTCI values along each route is studied. Supervised machine learning models are applied to predict bike ridership based on UTCI and other predictors.UTCI simulation results show that incorporating the various increments of accuracy influences hourly UTCI values at urban areas and exposed areas differently. Incorporating local wind speeds is especially impactful for urban areas. The statistical models trained to predict hourly bike trip counts based on UTCI and other demand and weather predictors achieved a root-mean-squared error of 1.06 trips. 47% of predictions were correct, and an additional 42% of predictions were off by 1 trip.This study demonstrates the importance of spatial refinement in simulating UTCI, and motivates future research into efficient simulation methods or rules-of-thumb for deriving spatial-temporal UTCI values. Future work into building a robust predictive model would motivate the design of thermally comfortable environments for human-powered transportation in cities.

Cell‐based hybrid adaptive mesh refinement algorithm for immersed boundary method

AbstractThis article proposes a hybrid adaptive mesh refinement (AMR) algorithm for the immersed boundary method (IBM) combination, and the AMR code is split into mesh and physics codes to optimize each part individually. The uniform parent grid solver is used for AMR grids by constructing hanging cells for the high‐order extension. The restrictive spatial refinement in a fully threaded tree (FTT) data structure is explored, and a simplified stencil search algorithm for hanging cells construction is introduced, including current and following child AMR level cells, if any. The proposed AMR method was applied to IBM, which offers flexibility in treating complex geometries in the Cartesian grid, leading to algorithmic simplicity and computational efficiency. Local near immersed boundary refinement is proposed to avoid complex and computationally expensive IBM and AMR algorithms near the solid bodies. Finally, a high‐order flux scheme extension at AMR level transition cells and the proposed method's applicability for steady and transient flows are demonstrated. Besides state and flux variables storage, the proposed hybrid AMR method's additional cost is 4.5 words/cell instead of 3.5 words/cell for 2D and 3.75 words/cell instead of 2.75 words/cell for 3D compared to the conventional FTT‐based AMR method. By comparing the uniform grid, the AMR final grid distribution shows that the optimal cell distribution based on flow physics reduces the cells required by more than 40% to resolve the complex flow features in less computational time. The enhanced performance and accuracy of the proposed AMR method in resolving different scale flow features are validated through benchmark problems.

Verification and validation of CFD simulations for semi-submersible floating offshore wind turbine under pitch free-decay motion

In this paper a detailed verification and validation study is performed to evaluate the accuracy of CFD simulations for a semi-submersible floating offshore wind turbine in pitch free-decay motion. The simulations are carried out with both linear and dynamic mooring models. The linear and quadratic damping coefficients of the semi-submersible are derived from the pitch motion data. In the verification study, a least-squares method using the observed order of convergence in combination with a data quality measure for different spatial and temporal refinement is applied to estimate the discretization errors. The numerical uncertainty is quantified by applying a safety factor to the summation of numerical errors. The numerical solutions are validated against the experimental measurements obtained from wave tank test of OC5-DeepCWind semi-submersible. Validation is achieved in the pitch natural period and the linear damping coefficient within the level of validation uncertainty. While the numerical simulation under-predicts the quadratic damping coefficient. A guideline is provided for verification and validation study of a CFD code application for floating marine structures under decay tests.

Improving Collaborative Pathfinding Using Map Abstraction

In this paper we combine recent pathfinding research on spatial abstractions, partial refinement, and space-time reserva- tions to construct new collaborative pathfinding algorithms. We first present an enhanced version of WHCA* and then show how the ideas from WHCA* can be combined with PRA* to form CPRA*. These algorithms are shown to effectively plan trajectories for many objects simultaneously while avoiding collisions, as the original WHCA* does. These new algorithms are not only faster than WHCA* but also use less memory.

Spatial refinement grillage model of box-girder bridges

The objective of this study was to overcome the shortcomings of the current analysis and calculation methods of the spatial bar system model, plane beam grillage model and solid element model in the design of the box-girder bridge. For example, the spatial bar system model lacks refined analysis of spatial effects, such as the shear lag effect, effective distribution width problem, and eccentric load coefficient problem. The similarity of the plane beam grillage model on the principle of equivalence cannot accurately reflect the shear stress distribution of the box-shaped composite beam and the local force on the top and bottom plates. The solid element and shell element models cannot directly calculate the internal force of the cross section. The calculation efficiency is low, and it is difficult to combine with the overall calculation. This paper proposes a spatial grid element analysis model suitable for fine analysis of box-girder bridges, which has the completeness and reliability of the analysis results and makes up for the shortcomings of the current box-girder bridge analysis methods. Firstly, the stiffness matrix of the spatial grid element is deduced, and the analysis program of the combined box girder based on the spatial grid element is written using MATLAB software. Then an actual calculation example is analyzed, and the difference between the calculation result of the spatial grid element and the result of the literature is compared and analyzed. The correctness and applicability of the spatial grid element model is verified.

Prediction of Surface Water Quality by Artificial Neural Network Model Using Probabilistic Weather Forecasting

We developed an artificial neural network (ANN)-based water quality prediction model and evaluated the applicability of the model using regional probability forecasts provided by the Korea Meteorological Administration as the input data of the model. The ANN-based water quality prediction model was constructed by reflecting the actual meteorological observation data and the water quality factors classified using an exploratory factor analysis (EFA) for each unit watershed in Nam River. To apply spatial refinement of meteorological factors for each unit watershed, we used the data of the Sancheong meteorological station for Namgang A and B, and the data of the Jinju meteorological station for Namgang C, D, and E. The predicted water quality variables were dissolved oxygen (DO), biochemical oxygen demand (BOD), chemical oxygen demand (COD), total organic carbon (TOC), total phosphorus (T-P), and suspended solids (SS). The ANN evaluation results reveal that the Namgang E unit watershed has a higher model accuracy than the other unit watersheds. Furthermore, compared with Namgang C and D, Namgang E has a high correlation with water quality due to meteorological effects. The results of this study will help establish a water quality forecasting system based on probabilistic weather forecasting in the long term.

Prediction of heat transfer from a circular cylinder in a cross-flow at a low sub-critical Reynolds number with the Partially-Averaged Navier-Stokes method

The prediction of turbulent heat transfer at a reasonable computational cost is a topical issue in computational fluid dynamics. This study addresses the numerical simulation of heat transfer from a circular cylinder in a cross-flow. Simulations were performed at a low sub-critical Reynolds number (Re = 3900) using the Partially-Averaged Navier-Stokes (PANS). Seven ratios of unresolved to total turbulent kinetic energy fk ranging from 0.15 to 0.7 were considered in order to reveal the influence of physical resolution on the flow and heat transfer statistics. Further, a low Reynolds number approach fε = fk was evaluated for fk ≥ 0.5. Additionally, simulations using the Large Eddy Simulation (LES) method were conducted on the same computational grids. The low physical resolution (fk > 0.5) in the PANS simulation at the coarse spatial resolution led to the earlier transition and later separation of the boundary layer which resulted in the overprediction of heat transfer at the rear stagnation point. The low Re approach (fε = fk) provided similar results as the low physical resolution case (fk = 0.7, fε = 1) and spatial refinement did not influence the predicted results. The results of our study indicated the suitability of the PANS for applications involving heat transfer from a bluff body provided an adequate physical resolution level is set.