136986 research outputs found
Sort by
Publisher Correction: Parameters for one health genomic surveillance of Escherichia coli from Australia (Nature Communications, (2025), 16, 1, (17), 10.1038/s41467-024-55103-2)
No full text10.1038/s41467-025-58600-
Dewatering of Clay-Rich Mineral Tailings: Impact Quantification and Improvement through Pelleting Flocculation
No full text© 2025 Yuxuan LuoMineral tailings are the solid-liquid waste produced during mineral processing, consisting of fine-grained solids remaining after the extraction of valuable minerals, along with process water and residual chemical reagents. Tailings are typically dewatered prior to deposition in a tailings storage facility (TSF) to reduce environmental risk, recover process water, and enable safe disposal or potential reuse. However, declining ore grades required finer grinding to liberate target minerals, which has also released ultra-fine and clay minerals, particularly kaolin, into the tailings stream. These clay-rich tailings are difficult to dewater due to their high water retention and low permeability, which significantly limits the performance of conventional dewatering technologies.
This thesis presents an alternative approach that simplifies the dewatering of high-clay content tailings. The work includes three main contributions: systematic quantification of the impact of clay on dewatering using compressional rheology, the development of a strengthened pelleting flocculation technique to improve dewaterability, and the design, construction and demonstration of novel batch and continuous pelleting flocculation devices for processing clay-rich suspensions.
To quantify the impact of clay on dewatering, batch sedimentation and pressure filtration were used to measure compressibility and permeability across varying solids concentrations and clay proportions. Simulated tailings were prepared by adding fine kaolin to calcium carbonate and alumina suspensions. The dewatering characteristics were used to parameterise thickening and filtration models, which revealed a significant decline in dewatering efficiency with increasing clay content. Results showed that increasing kaolin content significantly reduced dewatering performance, resulting in lower underflow solids and higher filter cake moisture. Highlighting the operational difficulty of managing clay-rich tailings.
Although flocculants are typically added to improve permeability and enhance settling, their use often compromises compressibility, resulting in a lower extent of dewatering. To overcome these trade-offs, this work introduces pelleting flocculation, a technique that forms compact, strong, pellet-like aggregates directly from solid-liquid suspensions. A combination of oil-in-water emulsion, polymeric flocculant, and controlled mechanical shear was used to produce dense pelletised kaolin aggregates, resulting in a significantly higher solids concentration than those achievable through gravity thickening alone. The pellets formed were strong enough to undergo further densification through air-tumbling in a rotating screen, producing pure kaolin pellets with solid content up to 52% w/w.
Novel batch and continuous vertical screw pelletiser prototypes were developed and tested. These devices formed centimetre-sized pellets directly from kaolin suspensions, achieving solids contents up to 31% w/w. The dewaterability of the pelletised aggregates was characterised and compared with that formed using the conventional flocculation method. The pelletised aggregates demonstrated a significant improvement in both compressibility and permeability, leading to enhanced thickening and filtration performance.
In summary, this work presents an alternative solution for the dewatering of clay-rich tailings. By converting problematic clay-rich tailings into compact, low-moisture pellets, strengthened pelleting flocculation offers a simplified dewatering technique that reduces reliance on thickeners and filters. The approach has potential both as a stand-alone dewatering method and as a pre-treatment step to improve downstream dewatering processes, with benefits for dewatering efficiency and higher water recovery
Australian urban planners’ preparedness to act on climate change
No full textLand use and development patterns have a significant impact on greenhouse gas emissions (GHG), and on managing the risk that climate change poses. Thus, urban planners play a critical role in addressing climate change, working with diverse built environment actors such as landscape architects. However, research indicates that while urban planners know about climate change, their self-perceived skills and competence are limited. This paper seeks to understand the preparedness of Australian urban planners to act on climate change (both mitigating GHG emissions and adapting to climate change impacts). Through in-depth interviews with 23 diverse Australian urban planners, preparedness to act on climate change is explored using Moser and Luers’ AAA climate change preparedness theory: Awareness of climate change; Analytical capacity to address climate change; and Actions taken to address climate change. Most respondents were able to identify climate change risks (awareness). Climate change risks were being assessed (analytical capacity) at a minimum through planning policy and tools informed by flood modelling and other risk assessments. In more progressive practice, planners draw upon internal or external climate change expertise beyond the planning system tools. The most frequently stated action taken by respondents to address climate change was the development of policies and strategies within their own organisation – from development of climate adaptation plans by those working in government, to organisational sustainability plans for those in the private sector. Results indicate the urban planning system is at times a facilitator of climate change action. A proportion of respondents were only exposed to climate change information, analytical capacity and actions due to planning tools. A framework of climate change preparedness was developed, demonstrating examples of low to high preparedness observed across respondents. The paper identifies characteristics of urban planning cliamte change front-runners, and suggests ways to progress climate change action through urban planning practice
Artificial Intelligence (AI) and Building Information Modeling (BIM) for Enhanced Property Valuation
No full text© 2025 Peyman JafaryAccurate property valuation is essential in navigating housing sector challenges by promoting fair pricing and stability in real estate markets. Financial institutions rely heavily on valuations to determine loan amounts, loan-to-value ratios and interest rates during mortgage underwriting and refinancing. Besides, reliable valuations support fair property tax calculations, allowing affordable housing units to benefit from appropriate tax reductions. Property valuation also informs policy-making by facilitating targeted interventions and incentives to improve housing affordability.
Traditional valuation methods, which rely on direct inspections and professional judgment, are guided by regulatory and professional standards. However, scholars argue that these time- and cost-intensive practices often fail to deliver accurate, efficient and consistent estimations due to the subjective nature of valuers’ judgments, client influence and variations in valuers’ skills and experience. Among the fundamental approaches to property valuation, the market approach estimates property value by comparing it with recent sales of similar assets, offering practical use of current market data but often leading to inconsistencies due to subjective comparisons and market fluctuations. In contrast, the cost approach, particularly the Depreciated Replacement Cost (DRC) method, values properties based on intrinsic characteristics, but it fails to account for broader market impact and socioeconomic influences.
Property valuation is inherently data-driven, and scholars have long turned to data-driven methodologies to enhance valuation practices. A key mathematical technique for automating property valuation is developing Automated Valuation Models (AVMs), where a property’s value can be estimated as a dependent variable influenced by a set of independent structural, geographic, socioeconomic, environmental, and legal and planning factors. Early AVMs relied on statistical algorithms, like Ordinary Least Squares (OLS) regression and Multiple Regression Analysis (MRA), forming the basis for Hedonic Pricing Models (HPMs). However, traditional AVMs struggle with complex real estate market dynamics, especially when spatial dependency and non-linear relationships are present. Recent advancements leverage Machine Learning (ML) to capture non-linear relationships, significantly improving the performance of AVMs. Yet, despite these advancements, AVMs remain vulnerable to market fluctuations, a limited number of transactions and periods of high volatility.
Furthermore, property valuation should account for the three-dimensional (3D) nature of buildings, as real-world environments are inherently 3D. With recent advancements in 3D technologies, Building Information Modeling (BIM) is becoming a common approach for designing and maintaining buildings and infrastructures. These 3D models (BIM) can be a source for providing detailed information about each component in a building. BIM can enhance valuation processes by supplying rich, geometrically and semantically accurate insights into a building’s structure. Integrating BIM with the cost approach, particularly its potential in automating Construction Cost (CC) estimation and Life Cycle Cost (LCC) analysis, offers significant advantages for advancing valuation practices. Although recent research has attempted to incorporate 3D BIM into valuation practices, integrating BIM into AVMs remains challenging due to limited BIM data availability for training and testing models and the complexity of extracting relevant features from BIM.
These ongoing challenges underscore the need for more sophisticated, data-driven and integrated methods to improve valuation accuracy and efficiency across various conditions. From one perspective, the market approach and its automated forms, AVMs, require more reliable datasets, consideration of 3D property characteristics through BIM integration and greater specificity to individual property attributes. On the other hand, while the cost approach neglects market influences, it provides substantial benefits, such as capturing intrinsic structural details of real estate assets and establishing strong connections with BIM-driven procedures in CC estimation and LCC analysis. Accordingly, this study aims to develop an integrated property valuation method that strategically combines elements of both the cost and market approaches, leveraging Artificial Intelligence (AI) and BIM.
This research adopts a computational and case study-based research methodology through the Design Science Research Methodology (DSRM), involving literature review, model design, data collection, model development for different components, their integration and validation through real cases. By incorporating AI and BIM at various stages of the cost approach and using ML for market impact adjustments, the proposed hybrid method comprises the following six key stages:
1. Land valuation: Different AVMs were developed and compared for mass land valuation using ML techniques of Support Vector Regression (SVR), Random Forest (RF), XGBoost and Deep Neural Network (DNN) based on a wide range of influential factors and extensive feature selection and hyperparameter tuning procedures to ensure accuracy and scalability.
2. Reconstruction cost estimation: A BIM-based Quantity Take-Off (QTO) method was integrated with an NLP-based cost-matching model to automate CC estimation. BIM enabled precise and consistent quantity extraction through its 3D geometry and parametric data, supporting a fully digital workflow when integrated with estimation tools like CostX. This approach allowed for reliable quantification of complex elements and consistent classification across trades. Building on this, an NLP-based cost alignment system was developed to automate the matching of QTO descriptions with cost items from industry-standard cost indexes. Using customized models based on spaCy, BERT, Word2Vec and GloVe, the final ensemble model achieved the highest alignment accuracy. This NLP-driven method eliminated manual errors and significantly reduced estimation time.
3. Dynamic depreciation estimation: A BIM-integrated facility management system was proposed to assess depreciation dynamically through maintenance tracking in facility and maintenance management systems to calculate depreciation.
4. ML-based regional market impact assessment: A spatial and ML-driven model was developed to estimate median property values across different sub-markets (suburbs) using market-related factors, such as median price trends, rental yields and clearance rates. Missing data was imputed using spatial ML-based techniques, ensuring comprehensive analysis. This stage systematically quantified the influence of market conditions on property valuation by introducing a market adjustment coefficient derived from regional property market data.
5. Entitlements of different property types were calculated using statistical and optimization techniques, ensuring an accurate distribution of property values in multi-unit developments. This process expressed each unit's value as a proportion of the total value of all units in the scheme determined through the DRC method.
6. Final valuation and validation: Values derived in each stage were integrated into a cohesive valuation model. The estimated property values of the case study were validated against recent transaction data.
The results confirm the hybrid model’s effectiveness and accuracy across its computational components. The ML-based land valuation model, particularly XGBoost, outperformed other techniques, achieving the highest accuracy in estimating land values across the study area. The BIM provided an accurate QTO for the case study, while the ensemble NLP-based cost-matching model successfully assigned cost descriptions from industry-based cost indexes to different building work items in the QTO. The system was tested on a benchmark building with a known Bill of Quantities (BoQs), achieving an 82.96% agreement rate, demonstrating its reliability. For market impact assessment, median property values across different suburbs were estimated using ML techniques and market-related metrics with high accuracy. These values were then used as coefficients to adjust estimates derived from the DRC method. To validate the final valuation, the model’s predictions were compared with the actual transaction prices of a newly constructed high-rise in South Melbourne, Victoria, Australia. The estimated values for one- and two-bedroom apartments fell precisely within observed market transaction ranges, while for three-bedroom apartments, the deviation was just 0.057%. These findings reinforce the model’s robustness and alignment with real-world market trends.
This research contributes to the real estate, property valuation and construction industries by demonstrating the transformative role of AI and BIM in valuation practices and cost estimation procedures. The study highlights how AI-powered automation enhances valuation accuracy and efficiency while BIM-based modeling improves 3D cost estimation and property-specific analysis. The proposed framework bridges the gap between the cost and market approaches, ensuring that valuations remain structurally precise and market-responsive. The modular and scalable nature of the framework allows for integration with government valuation agencies, financial institutions and real estate professionals
The effect and psychological mediators of a redesigned invitation letter to increase bowel cancer screening intention and likelihood: A randomised controlled experiment
No full textBACKGROUND: Participation in the Australian National Bowel Cancer Screening Program (NBCSP) is low. Modifying the invitation letters sent with home bowel cancer screening kits may increase participation. OBJECTIVES: This study compared current and modified versions of the NBCSP invitation letter to assess impact on intention to screen and perceived likelihood of screening and the role of screening barriers, enablers, and recipient characteristics. METHODS: In an online survey, 700 Australians aged 50-74 years were randomly assigned to view either the current NBCSP invitation letter or a redesigned version featuring a deadline, a prompt to place the kit somewhere visible, and simple pictorial instructions. Intention to screen for bowel cancer, perceived likelihood of screening, and screening barriers and enablers were measured before and after letter exposure. Mixed ANOVA and ANCOVA were used to assess pre-post changes in screening outcome variables, the mediating role of changes in screening barriers and enablers, and the moderating effect of age, gender, education, and screening history. RESULTS: Compared to the control letter, the redesigned letter was associated with an absolute increase in perceived screening likelihood of 2.8 %, F(1, 698) = 5.77, p = .017). This effect was mediated by increased screening benefits (b =.03, p = .019), self-efficacy (b =.03, p = .024), and planning to use the kit (b =.05, p = .008). Decreased perceived difficulty and avoidance mediated the effect of the resigned letter only among those who had never screened before (b = -0.04, p = .017), and males (b = 0.03, p = .045), respectively. CONCLUSION: Invitation letters that highlight screening benefits and provide simple, practical guidance to support planning and self-efficacy may increase recipients' screening likelihood. Messaging reducing difficulty and avoidance barriers may be most effective for males and those who have not screened previously
Not just PAH<sub>3.3</sub>: Why galaxies turn red in the near-infrared
No full textWe measured the spectral properties of a sample of 20 galaxies at z ∼ 0.35 selected for having surprisingly red JWST/NIRCAM F200W-F444W colors. Of these, 19 galaxies were observed with JWST/NIRSpec in the PRISM configuration, while the remaining galaxy was observed with the high-resolution gratings. Of the 20 galaxies in our sample, 17 exhibit strong 3.3 μm polycyclic aromatic hydrocarbon (PAH) emission (equivalent width (EW) (PAH3.3) ≥ 0.03 μm). In these galaxies, the strength of the color excess does not depend on environment and correlates with EW(PAH3.3). Nonetheless, the presence of the PAH3.3 alone cannot fully explain the color excess, as an EW of ∼0.1 μm is able to increase the color of galaxies by only 0.13 mag. A contribution from a hot dust component is required to explain the excess. Both the PAH3.3 EW and flux correlate with the Hα EW and flux, suggesting that they are produced by the same mechanism. Five of the galaxies of our sample showing PAH3.3 would be classified as passive based on broadband rest frame colors ((B-V) and/or UVJ diagrams) and are hence “faux passive”. Of these, three galaxies have a significantly lower EW(PAH3.3) given their color and also have low EW(Hα), and we tentatively conclude that this behavior is due to the presence of an active galactic nucleus. The three galaxies with no PAH3.3 in emission have passive spectra, as do the eight galaxies in our sample with normal F200W-F444W colors. We therefore conclude that the PAH3.3 feature is linked to dust-enshrouded star formation. The dust-corrected star formation rate (SFR) from PAH3.3 is a factor of 3.5 higher than the SFR obtained from Hα, suggesting that these galaxies are characterized by significant amounts of dust
Child-Centered AI: Contextualizing Principles and Design in HCI
No full textThis forum features practitioner perspectives on designing technologies for and with communities. We highlight compelling projects and provocative points of view that speak to both community technology practice and the interaction design field as a whole.
--- Sheena Erete, Edito
Early language outcomes of children born with unilateral aural atresia
No full textObjective: Aural atresia is a congenital malformation involving the ear canal. There is limited investigation into the impact of aural atresia and the associated hearing loss on language in the early years of development. Methods: Eight children with unilateral aural atresia were followed longitudinally at 30- and 60-days post hearing device fitting, 12 months, 18 months, 24 months, 30 months, and 36 months of age. Expressive language, auditory development, functional auditory performance, and hearing device use were measured. Results: Two children were delayed in expressive language at 24 months of age. Children's auditory development and functional auditory performance were in the average or above average range. Children wore their hearing devices for an average of 1–2 hours per day. Conclusion: Thirty-three percent of children (n = 2) were performing below the language levels expected for peers without hearing loss. Children's auditory development and functional auditory performance are comparable to children with typical hearing despite their low hearing device use. These results should be interpreted with caution due to the small number of children participating in the study
Enzymatic Degradation of PFAS: Current Status and Ongoing Challenges
No full textPer- and polyfluoroalkyl substances (PFAS) are often considered the quintessential example of industrial chemical pollution – they are toxic and ubiquitous environmental contaminants that are extremely difficult to degrade. There has been a large research focus on the development of effective and renewable degradation technologies. In comparison to traditional pollutant degradation techniques, such as advanced oxidation processes and electrochemistry, degradation of PFAS using extracellular enzymes offers an eco-friendly solution as enzymes are biodegradable, recyclable and have low energy and chemical requirements. This review outlines the current understanding of extracellular enzymatic degradation of PFAS with a focus on reported results and proposed degradation mechanisms. More importantly, this review highlights limitations that hinder the application of enzymes for PFAS degradation and proposes critical future research that is needed to improve the applicability of this promising remediation strategy
Patterns of mixed virus infections: a 3-year study of symptomatic cereal and grass hosts in Australia
No full textContext
Yellow dwarf viruses (YDVs) form a complex of economically important pathogens that can significantly reduce grain yield in cereals. Mixed infections, or infection with two or more YDV species, can be particularly damaging.
Aims
We aimed to examine the proportion of single and multiple virus infections present in symptomatic cereal and grass plants in Victoria, south-eastern Australia.
Methods
Over 3 years (2020–2022), symptomatic cereal and grass plants from within and around cereal fields in Victoria, Australia were individually tested using tissue-blot immunoassay (TBIA) for barley yellow dwarf virus PAV, barley yellow dwarf virus MAV, cereal yellow dwarf virus RPV, wheat streak mosaic virus, and with a generic TBIA test that can detect multiple luteovirus and/or polerovirus species.
Key results
Across 2020–2021, 34% of virus-positive plants were infected with multiple YDV species. The proportion of mixed infections was similar in each individual year. However, higher proportions of wheat (Triticum aestivum, 47%) and wild oat (Avena fatua, 36%) plants were infected with multiple YDV species compared to barley (Hordeum vulgare, 8%) and brome grass (Bromus spp.,17%).
Conclusions
The proportion of virus-positive plants infected with multiple YDV species found was almost four times higher than previously reported in a similar study in Victoria, Australia in 1985. The proportion of plants infected with multiple YDV species varied more with host type than between individual years.
Implications
These findings demonstrate the complex epidemiology of these damaging viruses, and the challenges associated with developing virus-resistant cereal cultivars, while also highlighting the importance of regular surveillance over multiple years