Accuracy Of Dataset Research Articles

Quantitative Identification of Lithofacies by Integrating Diverse Geophysical Data in the Late Holocene False River Point Bar Complex, Louisiana, USA

The False River area of the Mississippi River preserves a Holocene example of a meandering river point bar in a continental-scale system. We predict lithofacies quantitatively, using K-means cluster analysis from a nine-well data set that integrates well-log data (electrical conductivity (EC), corrected pressure and horizontal hydraulic permeability), grain size analyses and 14 defined sediment types. Data were collected from an unconsolidated and water saturated environment down to 90 feet (27.4 m). Because of the low number of grain-size analyses we develop a novel “hybrid” approach primarily using Generalized Additive Models, which allow us to create a more extensive and accurate data set. Horizontal permeability values calculated from the corrected pressure match those found (800–21,000 md) in the analogous Cretaceous-age McMurray Formation. We create six lithofacies clusters ranging from fine-grained near the surface to coarse sand at the base, with a vertical resolution of 0.5 m. Envelope Seismic attribute analysis along two profiles (30 and 52 m-long) validates the lithofacies prediction by using well-log data from a subset of four additional wells. Also, the relationship between the Envelope seismic attribute and permeability helps validate the number of lithofacies at two, tested, well locations. Permeability projected along the Envelope seismic attribute profile, away from the well location, aids in interpretation. Our quantitative lithofacies, well data (EC), and projected permeability maps support an interpretation of a buried remnant, channel-like feature, predating (> 4000 y B.P.) the modern point bar system.

MOSAEC-DB: a comprehensive database of experimental metal-organic frameworks with verified chemical accuracy suitable for molecular simulations.

Ongoing developments in computational databases seek to improve the accessibility and breadth of high-throughput screening and materials discovery efforts. Their reliance on experimental crystal structures necessitates significant processing prior to computation in order to resolve any crystallographic disorder or partial occupancies and remove any residual solvent molecules in the case of activated porous materials. Contemporary investigations revealed that deficiencies in the experimental characterization and computational preprocessing methods generated considerable occurrence of structural errors in metal-organic framework (MOF) databases. The MOSAEC MOF database (MOSAEC-DB) tackles these structural reliability concerns through utilization of innovative preprocessing and error analysis protocols applying the concepts of oxidation state and formal charge to exclude erroneous crystal structures. Comprising more than 124k crystal structures, this work maintains the largest and most accurate dataset of experimental MOFs ready for immediate deployment in molecular simulations. The databases' comparative diversity is demonstrated through its enhanced coverage of the periodic table, expansive quantity of structures, and balance of chemical properties relative to existing MOF databases. Chemical and geometric descriptors, as well as DFT electrostatic potential-fitted charges, are included to facilitate subsequent atomistic simulation and machine-learning (ML) studies. Curated subsets-sampled according to their chemical properties and structural uniqueness-are also provided to further enable ML studies in recognition of the strict demand for duplicate structure elimination and dataset diversity in such applications.

Accurate Dehydrogenation Enthalpies Dataset for Liquid Organic Hydrogen Carriers

This contribution presents a comprehensive extension of the QM9 dataset (originally at 133 K molecules) with the calculation of G4MP2 enthalpies for 9,841 molecules, featuring up to nine heavy atoms. We present QM9-LOHC, a (de)hydrogenation dataset of 10,373 reactions, including a minimum of 5.5% weight hydrogen storage capacity in line with the Department of Energy standards for Liquid Organic Hydrogen Carriers (LOHC). By utilizing the accurate quantum chemical method G4MP2 we expand the QM9 database and explore new avenues for the exploration of hydrogen storage technologies (electrochemical LOHCs, alkali metal-LOHCs, and mixtures of LOHCs). The QM9-LOHC dataset, with its focus on reactions that vary only by hydrogen saturation levels, provides a needed data resource for advancing the design and optimization of both conventional and innovative LOHC systems, and high-fidelity data for molecular discovery.

On the Measurement of Laser Lines in 3D Space with Uncertainty Estimation.

Laser-based systems, essential in diverse applications, demand accurate geometric calibration to ensure precise performance. The calibration process of the system requires establishing a reliable relationship between input parameters and the corresponding 3D description of the outgoing laser beams. The quality of the calibration depends on the quality of the dataset of measured laser lines. To address this challenge, we present a stochastic method for measuring the coordinates of these lines, considering both the camera calibration uncertainties and measurement noise inherent in laser dot detection on a detection board. Our approach to composing an accurate dataset of lines utilises a standard webcam and a checkerboard, avoiding the need for specialised hardware. By modelling the uncertainties involved, we provide a probabilistic description of the fitted laser line, enabling quality assessment of the measurement and integration into subsequent algorithms. We also offer insights into the optimal number of board positions and the number of repeated laser dot measurements, which are both the main time-consuming factors in practice. In summary, our proposed method represents a significant advancement in the field of laser-based system calibration, offering a robust and efficient solution.

Gender and Racial Diversity Assumed by Text-to-Image Generators in Microsurgery and Plastic Surgery-related Subspecialities

Since the release of ChatGPT by OpenAI in November 2022, generative artificial intelligence (AI) models have attracted significant attention in various fields, including surgery. These advancements have been particularly notable for creating highly detailed and contextually accurate images from textual prompts. A notable area of clinical application is the representation of surgeon demographics in various specialties, particularly in the context of microsurgery and plastic surgery-related subspecialties. This cross-sectional study, conducted in June 2024, utilized the latest version of the Copilot Creative Mode powered by DALL-E 3 to generate images of surgeons across various plastic surgery subspecialties. Real-world demographic data from the US, Japan, and Zambia were compared with AI-generated images for an accurate representation analysis. Five hundred images (350 from various subspecialties and 150 from geographical sources) were analyzed. The AI model predominantly generated images of male and female surgeons with a statistical underrepresentation of female and Black microsurgeons. Geographical prompts influenced the representation, with an overrepresentation of female (64.0%; p<0.001) and Black (16.0%; p<0.001) plastic surgeons in the US and exclusively Asian surgeons in Japan. Discrepancies were also observed in the depiction of surgical equipment, with the majority of AI-generated microsurgeons inaccurately portrayed using either surgical loupes (46.0%) or optical microscopes (32.0%), not with surgical microscopes (4.0%). This study revealed significant disparities between AI-generated images and actual demographics in the fields of microsurgery and plastic surgery-related subspecialties, highlighting the need for more diverse and accurate training datasets for AI models.

Adoption deep learning approach using realistic synthetic data for enhancing network intrusion detection in intelligent vehicle systems

In the dynamic field of cybersecurity within intelligent vehicle systems, the sophistication of threats necessitates continual advancements in security technologies. Traditional Network Intrusion Detection Systems (NIDS) often fall short in detecting emerging and sophisticated intrusion methods, primarily due to their reliance on static datasets that fail to capture the nuanced dynamics and complexity of modern network intrusions. This study presents a sophisticated simulation for NIDS tailored to intelligent vehicle environments, utilizing the extensive capabilities of Scapy—a robust network manipulation tool—to generate a highly accurate dataset of network traffic reflective of real-world scenarios. We created a diverse dataset involving 100,000 network flows, covering a wide array of benign, malicious, and anomalous traffic patterns, to thoroughly evaluate the detection capabilities of our proposed system. This dataset was analyzed using a deep learning framework employing a Convolutional Neural Network (CNN), which demonstrated outstanding performance metrics: an accuracy of 99.08%, precision of 98.96%, recall of 99.11%, and an F1 score of 99.03%. These metrics showcase the system's enhanced capability to precisely classify various network flows, emphasizing the importance of realistic synthetic data in boosting the training and accuracy of NIDS in intelligent vehicles. The results of this research are significant, marking a step forward towards more flexible and preemptive security measures for intelligent vehicles, and effectively narrowing the gap between simulation-based testing and real-world network environments.

Cold Front Identification Using the DETR Model with Satellite Cloud Imagery

The cloud system characteristics within satellite cloud imagery play a crucial role in the meteorological operational analysis of cold fronts, and integrating satellite cloud imagery into automated frontal identification schemes can provide valuable insights for accurately determining the position and morphology of cold fronts. This study introduces Cloud-DETR, a deep learning identification method that uses the DETR model with satellite cloud imagery, to identify cold fronts from extensive datasets. In the Cloud-DETR method, preprocessed satellite cloud imagery is used to generate training images, which are then put into the DETR model for cold front identification, achieving excellent results. The alignment between the Cloud-DETR cold fronts and weather systems during continuous periods and extreme weather events is assessed. The Cloud-DETR method exhibits high accuracy in both the position and morphology of cold fronts, ensuring stable identification performance. The high matching rate between the Cloud-DETR cold fronts and the manually identified ones in the test set, image dataset and labels from 2017 is verified. This indicates that the Cloud-DETR method can provide an accurate cold fronts dataset. The cold fronts dataset from 2005 to 2023 was obtained using the Cloud-DETR method. It was found that over the past 18 years, the frequency of cold fronts displays distinct seasonal patterns, with the highest occurrences observed during winter, particularly along the mid-latitude storm tracks extending from the east coast of East Asia to the Northwest Pacific. The methodology and findings presented in this study could help advance further research on the characteristics of cold front cloud systems based on long-term datasets.

SnowQM 1.0: a fast R package for bias-correcting spatial fields of snow water equivalent using quantile mapping

Abstract. Snow plays a crucial role in regional climate systems worldwide. It is a key variable in the context of climate change because of its direct feedback to the climate system, while at the same time being very sensitive to climate change. Long-term spatial data on snow cover and snow water equivalent are scarce, due to the lack of satellite data or forcing data to run land surface models back in time. This study presents an R package, SnowQM, designed to correct for the bias in long-term spatial snow water equivalent data compared to a shorter-term and more accurate dataset, using the more accurate data to calibrate the correction. The bias-correction is based on the widely applied quantile mapping approach. A new method of spatial and temporal grouping of the data points is used to calculate the quantile distributions for each pixel. The main functions of the package are written in C++ to achieve high performance. Parallel computing is implemented in the C++ part of the code. In a case study over Switzerland, where a 60-year snow water equivalent climatology is produced at a resolution of 1 d and 1 km, SnowQM reduces the bias in snow water equivalent from −9 to −2 mm in winter and from −41 to −2 mm in spring. We show that the C++ implementation notably outperforms simple R implementation. The limitations of the quantile mapping approach for snow, such as snow creation, are discussed. The proposed spatial data grouping improves the correction in homogeneous terrain, which opens the way for further use with other variables.

Use Self-Training Random Forest for Predicting Winter Wheat Yield

The effectiveness of supervised ML heavily depends on having a large, accurate, and diverse annotated dataset, which poses a challenge in applying ML for yield prediction. To address this issue, we developed a self-training random forest algorithm capable of automatically expanding the annotated dataset. Specifically, we trained a random forest regressor model using a small amount of annotated data. This model was then utilized to generate new annotations, thereby automatically extending the training dataset through self-training. Our experiments involved collecting data from over 30 winter wheat varieties during the 2019–2020 and 2021–2022 growing seasons. The testing results indicated that our model achieved an R2 of 0.84, RMSE of 627.94 kg/ha, and MAE of 516.94 kg/ha in the test dataset, while the validation dataset yielded an R2 of 0.81, RMSE of 692.96 kg/ha, and MAE of 550.62 kg/ha. In comparison, the standard random forest resulted in an R2 of 0.81, RMSE of 681.02 kg/ha, and MAE of 568.97 kg/ha in the test dataset, with validation results of an R2 of 0.79, RMSE of 736.24 kg/ha, and MAE of 585.85 kg/ha. Overall, these results demonstrate that our self-training random forest algorithm is a practical and effective solution for expanding annotated datasets, thereby enhancing the prediction accuracy of ML models in winter wheat yield forecasting.

Context-Aware QoE for Mobility-Driven Applications Through Dynamic Surveys

The integration of outdoor smartphone applications with fitness trackers has introduced new opportunities and challenges for user interaction, particularly in mobility-driven activities. While these innovations offer significant benefits, they also pose challenges due to the many factors influencing the user’s quality of experience. Traditional methods of assessing user experiences, such as offline surveys and static questionnaires, often fail to capture the dynamic nature of outdoor activities. This research proposes a novel Quality of Experience (QoE) methodology for mobile applications to enhance the assessment of user experiences in cycling. Focusing on a use case in Blekinge, Sweden, where residents were encouraged to adopt cycling for daily transportation, we extracted land cover data and developed a server-side workflow for bicycle path segmentation. By incorporating dynamic surveys that adapt to users’ real-time experiences, we aim to generate a more accurate and context-aware dataset. This study makes several key contributions: First, it presents a scalable method for bicycle path segmentation; second, it demonstrates the utility and benefits of land cover data extraction; and finally, it evaluates the effectiveness of QoE influence factors through user surveys based on real-world cyclist feedback. This approach is expected to enhance the planning and development of cycling infrastructure by providing urban planners and stakeholders with valuable user insights using adaptable surveys based on route segmentation.

Indian Summer Monsoon Rainfall Characteristics Derived From Multiple Gridded Precipitation Datasets: A Comparative Assessment

ABSTRACTPrecipitation, a crucial component of the Earth system processes, regulates the spatiotemporal cyclicity of water, energy, and carbon fluxes. Accurate precipitation datasets leverage the understanding of precipitation dynamics and are vital for hydro‐climatological studies. South Asian monsoon is a complex, multi‐scale interacting, synoptic, and ocean–land–atmosphere coupled system, contributing to significant spatial and temporal variability in summer monsoonal rainfall across India. This study evaluates four types of gridded (observational, satellite, reanalysis, and hybrid) precipitation products in their ability to replicate Indian Summer Monsoonal Rainfall (ISMR) characteristics using the India Meteorological Department (IMD) 0.25° gridded data as the baseline. A comparative assessment is performed in this study that uses several continuous and interval‐based performance measures to evaluate the overall rainfall magnitude detectability and time‐matched capturing of rainfall events. A new metric, rank score, is developed by aggregating multiple measures to find the best product. The analyses based on several performance measures indicate that MSWEP is the best dataset (rank one) that closely approximates the occurrence and magnitude of IMD‐based rainfall events, while APHRODITE, CHIRPS, and IMDAA are ranked as the next best set of products. PGF is ranked the lowest among all products evaluated and is not recommended for applications. Nonetheless, APHRODITE suffers from strong negative biases, while the reanalysis (IMDAA, ERA5‐Land, PGF) datasets show significant positive biases. Among the products evaluated, APHRODITE, ERA5‐Land, and IMDAA have shown a limited ability to detect excess, normal, and deficit monsoon years, respectively. In general, the performance of satellite‐based data products is superior to that of reanalysis datasets in accurately characterising the monsoon years. ERA5‐Land is noted to be the best‐performing dataset among the reanalysis products. The comprehensive comparative assessment carried out in this study benefits the selection and use of appropriate gridded precipitation products for hydroclimatic modelling, climate variability, and change studies.

A regression-based parametric model for radiative flux density distribution considering shadowing and blocking effects

In solar power tower system, the Radiative Flux Density Distribution (RFDD) on the receiver surface reflected by a heliostat is influenced by various factors, referred to as scene parameters. The previous analytical models simplify the complex optical modeling process, thus neglecting the comprehensive impacts of the scene parameters, resulting in simulation errors. In this paper, a regression-based parametric model, namely Neural Elliptical Gaussian (NEG), is proposed to address this issue. The NEG model comprehensively considers the impacts of various scene parameters on the RFDD, including heliostat size, slant distance of heliostat, incident angle of sunlight, sunlight distribution parameter, slope error, etc. The relationship between the scene parameters and the RFDD is established using a neural network. Additionally, the overlooked shadowing and blocking effects in the conventional analytical models and data-driven methods are addressed by introducing the flux spot centroid offset in the NEG model. Since the NEG model is established based on statistical regression using a sampled and more accurate flux spot dataset, it shows more accurate flux spot prediction ability. Experimental results show that, for most scenarios, the root mean squared error is less than 0.35%, and the total energy error and peak value error are less than 5%.

Development of a Multi-Source Satellite Fusion Method for XCH4 Product Generation in Oil and Gas Production Areas

Methane (CH4) is the second-largest greenhouse gas contributing to global climate warming. As of 2022, methane emissions from the oil and gas industry amounted to 3.586 million tons, representing 13.24% of total methane emissions and ranking second among all methane emission sources. To effectively control methane emissions in oilfield regions, this study proposes a multi-source remote sensing data fusion method based on the concept of data fusion, targeting high-emission areas such as oil and gas fields. The aim is to construct an XCH4 remote sensing dataset that meets the requirements for high resolution, wide coverage, and high accuracy. Initially, XCH4 data products from the GOSAT satellite and the TROPOMI sensor are matched both spatially and temporally. Subsequently, variables such as longitude, latitude, aerosol optical depth, surface albedo, digital elevation model (DEM), and month are incorporated. Using a local random forest (LRF) model for fusion, the resulting product combines the high accuracy of GOSAT data with the wide coverage of TROPOMI data. On this basis, ΔXCH4 is derived using GF-5. Combined with the GFEI prior emission inventory, the high-precision fusion dataset output by the LRF model is redistributed grid by grid in oilfield areas, producing a 1 km resolution XCH4 grid product, thereby constructing a high-precision, high-resolution dataset for oilfield regions. Finally, the challenges that emerged from the study were discussed and summarized, and it was envisioned that, in the future, with the advancement of satellite technology and algorithms, it would be possible to obtain more accurate and high-resolution datasets of methane concentration and apply such datasets to a wide range of fields, with the expectation that significant contributions could be made to reducing methane emissions and combating climate change.

What would environmental impact assessment look like if we started from scratch today? Designing better EIA for developed neoliberal nations

ABSTRACT After more than five decades of practice, environmental impact assessment (EIA) has failed to convince sceptics that it represents value for money. It increasingly overlaps with constantly emerging sustainability requirements. The completed assessments are extremely long, exceed the cognitive capacities of decision makers to assimilate information, cannot address motivated reasoning, and therefore inevitably lead to trade-offs that threaten the very environmental components EIA was designed to protect. In this paper, for the minority of nations with highly developed existing environmental legislation and management only, we propose three radical approaches that include: (1) the adoption of ‘satisficing’, to deliver a streamlined assessment that is good enough; underpinned by (2) better application of ‘acceptable harmrules’ embedded in existing environmental legislation in many jurisdictions to prevent significant harm to environmental media; and (3) an ‘externalities charge’ on developers (irrespective of whether EIA is required) to force more aggressive scoping through market incentives and to fund a shift towards adaptive environmental assessment and management that manages environmental outcomes. Better environmental protections could be delivered using a far more streamlined EIA process, associated with the creation and maintenance of more accurate and comprehensive datasets that can provide better evidence for emerging artificial intelligence tools.

Analysis of Paddy Field Changes (1989–2021) Using Landsat Images and Flooding-Assisted MLC in an Urbanizing Tropical Watershed, Vientiane, Lao PDR

Paddy fields are essential for food security and sustaining global dietary needs, yet urban expansion often encroaches on agricultural lands. Analyzing paddy fields and land use/land cover changes over time using satellite images provides critical insights for sustainable food production and balanced urban growth. However, mapping the paddy fields in tropical monsoon areas presents challenges due to persistent weather interference, monsoon-submerged fields, and a lack of training data. To address these challenges, this study proposed a flooding-assisted maximum likelihood classification (F-MLC) method. This approach utilizes accurate training datasets from intersecting flooded paddy field maps from the rainy and dry seasons, combined with the Automated Water Extraction Index (AWEI) to distinguish natural water bodies. The F-MLC method offers a robust solution for accurately mapping paddy fields and land use changes in challenging tropical monsoon climates. The classified images for 1989, 2000, 2013, and 2021 were produced and categorized into the following five major classes: urban areas, vegetation, paddy fields, water bodies, and other lands. The paddy field class derived for each year was validated using samples from various sources, contributing to the overall accuracies ranging from 83.6% to 90.4%, with a Kappa coefficient of between 0.80 and 0.88. The study highlights a significant decrease in paddy fields, while urban areas rapidly increased, replacing 23% of paddy fields between 1989 and 2021 in the watershed. This study demonstrates the potential of the F-MLC method for analyzing paddy fields and other land use changes over time in the tropical watershed. These findings underscore the urgent need for robust policy measures to protect paddy fields by clearly defining urban expansion boundaries, prioritizing paddy field preservation, and integrating these green spaces into urban development plans. Such measures are vital for ensuring a sustainable local food supply, promoting balanced urban growth, and maintaining ecological balance within the watershed.

Innovative reconstruction and evaluation of forest refinement datasets by combining multi-source data: A case study of Guangdong Province

The acquisition and monitoring of forest cover data are crucial for ecological protection, resource management, and climate change research. However, relying on a single data source provides limited data accuracy and does not adequately capture the forest structure and functional attributes. We combined six commonly land cover datasets and forest age, canopy height, above-ground biomass, and tree species distribution datasets to reconstruct 30 m spatially accurate forest refinement dataset (FRD) for Guangdong Province. In addition, the distribution characteristics of forest structure and function were evaluated using forest morphological spatial pattern analysis. The results show that the overall accuracy of FRD of the Guangdong Province in 2020 reached 86.07 %. Forest types in the Guangdong Province were mainly dominated by evergreen needle-leaf forests. Tsuga chinensis, Red cedar, and Pinus sylvestris were more commonly planted. Older and taller trees were found in northern and eastern Guangdong. In addition, forest above-ground biomass (AGB) was larger in the coastal areas of northern and western Guangdong. The core and perforation had the oldest age and the highest tree height, and the islet had the lowest for all forest structure and function indicators. Based on multi-source datasets, this study contributes to a better understanding of the attributes characterizing the structure and function of forests. The refined dataset and research framework will effectively enhance forest management efficiency and policy making, as well as provide case references for research on climate change response, forest conservation and biodiversity assessment.

CMMI &amp; Lean-Based Metrics Governance Model to Ensure Data Accuracy &amp; Consistency for Ai/Ml Applications for Sustainable Operations in Mining Industry

In today’s competitive business environment, even small improvements in productivity and efficiency can have a huge impact on company’s profitability. Whether it is through reduced lead time/waste, minimized downtime or improved quality, operation managers are looking for any advantage at every stage. Lean management is a long-term operational discipline that methodically seeks to enhance efficiency and quality by eliminating wastage. The concept is being used in varied industries and helped by improved productivity, safety &amp; better management. Lean in being used by few mining companies also at a small scale, which has helped them in limited way. AI/ML is another important technology used by many organizations. Many global mining companies have also used it for improving operational efficiency, which have helped miners in a limited ways in improving the productivity and safety. Mining companies have to deal with tons of data set generated from heavy machinery and plants, which have to be stored, processed, and stored. There are many challenges in the data set generated/used by AI/ML-application for mining operations. One of the key challenges in AI/ML is accessibility and availability of accurate &amp; consistent data. To ensure consistency of data there is a need for a governance structure which will guarantee the availability of required accurate data set for AI/ML system. One possible solution could be, CMMI’s metrics processes used by IT companies and have helped in improving quality. CMMI’s Metrics related processes are being used extensively by the organizations for better monitoring &amp; control of key measures. This paper explores the possibility of usage of CMMI’s metrics practices with lean implementation in mining for ensuring availability of accurate data sets for AI/ML applications. The proposed framework will put a governance model in place to ensure accuracy &amp; consistency of data to be used for AI/ML-applications in Mining.

Real-Time Leaves Segmentation in RGB Images with Deep Learning in a Single-Board Computer

Abstract. This work proposed and evaluated methods for real-time leaf segmentation using a single-board computer. The main aim was to explore the state-of-the-art techniques based on the YOLO algorithm for real-time operation. For this purpose, the available variants of YOLOv8 and YOLOv9 were evaluated, and a semi-automatic labelling method based on the Segment Anything Model (SAM) algorithm was used. Given the need to delimit the leaf contour for labelling, it was possible to create a larger and more accurate dataset compared to the purely manual procedure. In addition, the cost-benefit of the applied algorithms and methods were assessed, considering the computational demand required, as well as the accuracy, recall, and precision delivered by these techniques. In this study, both quantitative analysis of the trained architectures' metrics and qualitative examination through direct observation of images were conducted to identify crucial aspects. The experiments were conducted with a post-processed dataset and the suitability for real-time applications was based on the elapsed time for segmentation. We concluded that the YOLOv8n architecture is the best one among those tested, presenting a precision and recall of 0.9064 and 0.7233, respectively. This architecture represents the best cost-benefit ratio between computational cost and real-time performance, being able to perform segmentation in 310 ms with the NVIDIA Jetson Nano board. Furthermore, when computational cost is not a problem or even when segmentation time can be higher, the YOLO8m network may be recommended when the recall metric is more important than precision. This network presented a precision and recall of 0.8556 and 0.7726, respectively, and presented a better performance in segmenting leaves located in more complex parts of the image and with a higher recall.

Local temperature impact of urban heat mitigation strategy based on WRF integrating urban canopy parameters and local climate zones

Urbanization has led to significant alterations in surface properties, contributing to surging urban heatwave events. The detrimental impact of urban heatwaves on public health highlights the urgency for proactive evaluations of diverse heat mitigation strategies. This study assesses the impacts of cool roof (CR), green roof (GR) and rooftop photovoltaic panel (PV) strategies on urban air temperatures within Guangzhou and Foshan agglomeration, utilizing the Weather Research and Forecasting with Urban Canopy Model (WRF-UCM). A comparative analysis of heat exchanges is conducted between two types of urban morphology datasets: local climate zone maps with categorized urban canopy parameters (LCZ-UCPs) and gridded urban canopy parameters from real building databases (gridded-UCPs). The results indicate that the discrepancies in the average temperature between LCZ-UCP and gridded-UCP is almost negligible, whereas increase to 0.39 °C and 0.54 °C for the daily maximum and minimum temperature, respectively. CR provides the most substantial cooling at an average of 0.44 °C, followed by GR and PV. Scenarios with the two types of morphology datasets reveal varying mitigation efficiencies. CRs with LCZ-UCPs show more pronounced temperature reductions, whereas GRs and PVs with gridded-UCPs in central urban regions demonstrate stronger cooling effects. CRs and PVs cool temperature by decreasing sensible heat flux, whereas GR is largely influenced by enhanced evapotranspiration especially with grass plantation. These findings elucidate the differing efficiencies of the three mitigation strategies and highlight the representation discrepancies in climatic simulations brought by the two urban canopy datasets. This emphasizes the importance of accurate urban morphological datasets in evaluating mitigation strategies in WRF-UCM, thus providing practicable insights for urban planning and climate policy-making.

Evaluation of Automated Object-Detection Algorithms for Koala Detection in Infrared Aerial Imagery.

Effective detection techniques are important for wildlife monitoring and conservation applications and are especially helpful for species that live in complex environments, such as arboreal animals like koalas (Phascolarctos cinereus). The implementation of infrared cameras and drones has demonstrated encouraging outcomes, regardless of whether the detection was performed by human observers or automated algorithms. In the case of koala detection in eucalyptus plantations, there is a risk to spotters during forestry operations. In addition, fatigue and tedium associated with the difficult and repetitive task of checking every tree means automated detection options are particularly desirable. However, obtaining high detection rates with minimal false alarms remains a challenging task, particularly when there is low contrast between the animals and their surroundings. Koalas are also small and often partially or fully occluded by canopy, tree stems, or branches, or the background is highly complex. Biologically inspired vision systems are known for their superior ability in suppressing clutter and enhancing the contrast of dim objects of interest against their surroundings. This paper introduces a biologically inspired detection algorithm to locate koalas in eucalyptus plantations and evaluates its performance against ten other detection techniques, including both image processing and neural-network-based approaches. The nature of koala occlusion by canopy cover in these plantations was also examined using a combination of simulated and real data. The results show that the biologically inspired approach significantly outperformed the competing neural-network- and computer-vision-based approaches by over 27%. The analysis of simulated and real data shows that koala occlusion by tree stems and canopy can have a significant impact on the potential detection of koalas, with koalas being fully occluded in up to 40% of images in which koalas were known to be present. Our analysis shows the koala's heat signature is more likely to be occluded when it is close to the centre of the image (i.e., it is directly under a drone) and less likely to be occluded off the zenith. This has implications for flight considerations. This paper also describes a new accurate ground-truth dataset of aerial high-dynamic-range infrared imagery containing instances of koala heat signatures. This dataset is made publicly available to support the research community.