Welcome to Research Commons

Research Commons is the University of Waikato's open access research repository, housing research publications and theses produced by the University's staff and students.

Communities in Research Commons

Select a community to browse its collections.

Now showing 1 - 2 of 2

Recent Submissions

Modelling steroidal hormone transport and evaluating best management practices in New Zealand dairy farming systems
(The University of Waikato, 2026) Majeed, Muhammad Adnan; Lay, Mark C.; Bertram, Danielle Valerie; Glasgow, Graeme D.E.
This doctoral research developed and applied two modelling approaches to simulate and assess estrogen (estrone E1, and 17β-estradiol E2) transport from an intensive dairy farm system located in the Waikato region, New Zealand. The first approach was a numerical unsaturated zone model implemented in MATLAB, coupling Richards’ equation with a Green-Ampt infiltration framework and advection, dispersion, and sorption processes to track estrogen movement through soil. The second approach employed a GIS-based ArcSWAT catchment model to simulate surface runoff, sediment yield, and estrogen transport at the farm watershed scale, including the evaluation of best management practices (BMPs). The novelty of this research lies in the dual-scale modelling of estrogen transport through both soil and surface runoff pathways and the spatially explicit, quantitative assessment of existing best management practices specifically for estrogen mitigation rather than nutrient reduction alone. Soil Model results indicate that estrogens are largely retained and attenuated in the upper soil layers, reducing the risk of leaching to groundwater. The model showed that both E1 and E2 undergo strong sorption near the surface and substantial microbial degradation, with >90% of the applied mass removed within the top ~1 m over a 90-day simulation. Estradiol was more mobile than estrone, with its concentration front advancing deeper (~0.5–0.6 m). Concentrations declined steeply with depth, and values below ~1.5 m were negligible under steady infiltration. Infiltration-driven pulses produced by surface boundary input led to episodic colloid mobilization, which enhanced estrogen transport to mid-depths. Free and attached colloid time series showed synchronous peaks, indicating that mobile colloids can act as vectors for hormone movement. These findings suggest that under normal conditions, leaching risk is low, but during rapid infiltration following effluent application, colloid-facilitated transport may temporarily extend hormone movement deeper into the profile. Overall, the results align with experimental column studies, confirming that estrogen is primarily confined to shallow soil and largely degraded in situ, while emphasizing the importance of managing infiltration timing and flow dynamics. For the dairy farm studied, the ArcSWAT model provided a spatially explicit assessment of estrogen transport via runoff and erosion. Critical source areas (hotspots) for estrogen and sediment loss were identified at the subbasin level. Under baseline (pre-BMP) conditions, eight subbasins were predicted to generate the highest estrogen concentrations in runoff (often >15 ng/L) and severe sediment yields. One subbasin emerged as the most critical, with annual runoff volumes exceeding ~1500 mm and sediment losses on the order of ~3.9 t/ha, concomitant with elevated estrogen loads. Several other subbasins showed overlapping high runoff and erosion, indicating that both dissolved and sediment-bound estrogen transport mechanisms are at play in these areas. The model indicated that surface water contamination is a primary concern, as these high-runoff zones efficiently deliver estrogens to streams (e.g. the Maungatea Stream on the farm’s boundary). To mitigate these exports, the effectiveness of BMP scenarios was evaluated in ArcSWAT, including constructed wetlands, riparian buffer strips, grazing management, and effluent application timing management. All BMPs reduced estrogen and sediment delivery to some degree, but their performance varied. Constructed wetlands placed at critical drainage points showed the greatest overall impact, trapping 50-90% of sediment from upstream areas and removing an estimated 30-70% of estrogen loads via sedimentation, sorption, and microbial degradation in wetland ponds. Riparian buffers were similarly effective: vegetated buffer strips along stream channels filtered runoff, resulting in sediment transport reductions of about 16-80% (average ~42%) in high-erosion subbasins and estrogen concentration reductions on the order of 30-85% in runoff, through enhanced infiltration and filtering of hormone-laden sediment. Improved grazing management (e.g. rotational grazing and reduced stocking rates in wet periods) yielded moderate benefits, with an average ~17.5% decrease in sediment loss (due to better soil cover and less compaction) and commensurate declines in runoff (~12%) and estrogen exports. Effluent management (avoiding manure irrigation during wet weather and optimizing application rates) provided 5-25% lower sediment losses and up to 10-30% reductions in runoff, thereby modestly cutting estrogen runoff concentrations (5-50%). Notably, after implementing an effective riparian buffer scenario, peak estrogen levels in runoff from the worst areas fell from ~19.1 ng/L (baseline) to below 7 ng/L, and peak sediment yields dropped from ~3900 kg/ha to under 1000 kg/ha. These quantitative improvements underscore that targeted BMP adoption can substantially reduce estrogen loading to surface waters. The catchment modelling highlighted where and which interventions yield the greatest water-quality benefits: for instance, combining wetlands and buffers in the most critical subbasins would address both high runoff and erosion, greatly reducing the transport of both dissolved and particle-bound estrogens to streams.
“It almost wanted to hurt someone”: The impact of intentional creepiness on user perceptions
(ACM, 2026-03-16) Turner, Jessica Dawn; Vanderschantz, Nicholas; König, Jemma Lynette; Siddika, Rafeea
The intentional design of robots to evoke creepiness provides a unique lens for studying human perception and willingness to engage. To understand user perceptions and acceptance of robots we developed a robot prototype designed with targeted facial, morphological, and movement features that may be perceived as "creepy". Using the Human-Robot Interaction Evaluation Scale (HRIES) we found that disturbance was moderate towards our intentionally creepy robot with significant participant variation. Furthermore, qualitative results confirmed this polarity, with descriptions ranging from "angry and unfriendly" to "cool and cute". This variability demonstrates that "creepiness" is more subjective than initially anticipated and highlights a key research gap in academic literature with the need for measurement tools which capture negative perceptions in HRI.
Evaluating the effect of intermittent reinforcement on concept learning in canine lung cancer detection
(The University of Waikato, 2026) Ji, Linguo; Edwards, Timothy L.
Dogs have demonstrated the ability to identify a range of human diseases, including lung cancer, through olfactory analysis of biological samples. Although exhaled breath is a non-invasive and accessible sample type, no single volatile organic compound has been reliably identified as a biomarker. Therefore, the process of identifying lung cancer in human implies a process of concept learning, suggesting dogs’ detection of lung cancer may be relying on the subjects' ability to identify a complex, highly variable pattern of volatile organic compounds (VOCs) in exhaled breath. Concept learning appears to be a special form of generalization, with underlying mechanisms in common with those responsible for perceptual concept learning in the visual domain. While intermittent reinforcement is commonly recommended in scent-detection training to simulate and prepare for operational conditions where reinforcement is not always possible, its effects on conceptual generalization remain poorly understood. This study investigated the effects of intermittent reinforcement on canine concept learning in a lung cancer detection task. Five dogs were trained using a fully automated 17-segment carousel apparatus with breath samples collected from 348 patients who visited respiratory clinic, with 115 tested positive and 233 tested negative in lung cancer. A single-subject reversal design was employed; the reinforcement rate for correct indications to positive samples was systematically thinned from 100% down to ranges of 80% and 60%. The findings demonstrated that thinning the reinforcement schedule to a minimum of 60% did not exert a significant disruptive effect on the dogs' diagnostic accuracy. In addition, an exploratory probe test also provided preliminary evidence that the dogs could successfully differentiate between lung-originated cancer and non-lung-originated (NLO) cancer samples.
Accuracy of machine learning models versus "hand crafted" expert systems – A credit scoring case study
(Elsevier, 2009) Ben-David, Arie; Frank, Eibe
Relatively few publications compare machine learning models with expert systems when applied to the same problem domain. Most publications emphasize those cases where the former beat the latter. Is it a realistic picture of the state of the art? Some other findings are presented here. The accuracy of a real world “mind crafted” credit scoring expert system is compared with dozens of machine learning models. The results show that while some machine learning models can surpass the expert system’s accuracy with statistical significance, most models do not. More interestingly, this happened only when the problem was treated as regression. In contrast, no machine learning model showed any statistically significant advantage over the expert system’s accuracy when the same problem was treated as classification. Since the true nature of the class data was ordinal, the latter is the more appropriate setting. It is also shown that the answer to the question is highly dependent on the meter that is being used to define accuracy.
Where confidence fails, training prevails: Effects of behavioural skills training on improving kaimahi confidence when using patient management systems for outreach immunisation services.
(The University of Waikato, 2026) Squire, Danielle; Blackmore, Tania; Carlson, Teah
In New Zealand, immunisation rates among tamariki Māori are consistently lower nationally at all recommended age milestones. Access to clear and timely information can facilitate engagement with whānau Māori. This study evaluated the effectiveness of Behavioural skills training (BST) to improve kaimahi use of a patient management system (PMS) – Indici. A Kaupapa Māori approach was utilised to engage with the organisation. Seven participants were recruited from a local health and social services provider. Consultation supported development of task analyses used for training. BST occurs in four phases: instruction, modelling, rehearsal, and feedback. Data were collected following a multiple baseline design with a single baseline session, BST intervention, and a follow-up session to assess for maintenance. All participants demonstrated improvement in task accuracy following the introduction of BST and maintained task accuracy in a post-training follow-up. Despite improvements across all participants, task adherence was impacted during training. This study highlights that BST is an effective, socially valid tool for training simple skills for accessing information in a PMS such as Indici. However, further research is needed to examine the use of BST for teaching complex computer systems in applied health settings.