Environment Canterbury

Commercial shipping is a cornerstone of global trade. Its impact on the marine environment, however, remains underexplored. This study combines hydroacoustic data, sediment samples, propeller-induced shear stress calculations and vessel tracking information to assess the effects of shipping in one of the busiest maritime regions in the Baltic Sea, the Bay of Kiel. We unveil substantial seafloor erosion, including up to 1.5 m variation in water depths, over 10 years that clearly relates to vessel traffic. By imaging water column disturbance behind passing ships, we trace wake turbulence to the seafloor and show the breakdown of a strongly stratified water column and a possible excitement of internal waves, likely increasing the mixing of oxygen, nutrients, and greenhouse gases. While the environmental consequences of this anthropogenic stressor are unquantified, our findings leave little doubt that they include modifications to marine ecosystems and element budgets on a Baltic-wide scale.

Abstract The detection and monitoring of environmental change is vital for sustainable river management, but historical data are often limited. This study leveraged volunteered geographic information and local ecological knowledge from experienced recreational anglers to identify historical baselines and long-term change in the Rakaia River, a protected catchment under a National Water Conservation Order (WCO) in Aotearoa New Zealand. Oral histories spanning seven decades were recorded using semi-structured interviews with 30 veteran fishers representing 1510 years of combined catchment-specific experience. Inductive thematic analysis and participatory mapping were used to identify physical environment changes and socio-cultural effects associated with recent declines in freshwater fish populations. River environment changes include the loss of natural features such as springs and pools, declines in characteristic native species and habitats, shrinking river mouth la-goon and river plume extents, and interactions with public access. Socio-cultural effects include reduced participation in fishing, community despondence, emotional distress and demographic shifts in fishing hut settlements. Over 20 reported changes involve adverse effects on values that are specifically protected under the WCO, indicative of a policy failure. Declines in multiple indicators of environment health suggest an implementation gap that requires greater attention to environmental monitoring and outcome evaluation yet is hampered by uncertainties around institutional responsibilities. Oral history approaches can help to establish historical baselines for gauging long-term change and enabling adaptive management in this and other data-poor situations.

Grommet insertion is one of the most common surgeries in childhood. While post-operative care has been traditionally provided by Ear, Nose and Throat (ENT) specialists, workforce capacity challenges have led to the delegation of care to alternate workforces, including audiology. This study aims to evaluate safety and effectiveness of audiology-led post-grommet services and to compare costs of audiology-led models with traditional medical and simulated nurse-led models. A retrospective cohort study was conducted in a regional New Zealand public hospital including all 247 children attending an audiology-led post-grommet clinic from 2019 to 2025. Patient outcomes and pathways through post-grommet services were mapped and costs to the healthcare system and patient/families were calculated. 42.5% of children required an ENT assessment following the implementation of an audiology-led post-grommet model. At the first post-operative visit (median 184 days) 57.5% of grommets were nonfunctional. There were no adverse events reported over the six-year study period. Average healthcare system cost per patient was $NZ130.40 for the audiology-led model representing a 12% cost-savings compared to the standard ENT model ($NZ147.85). The simulated nurse-led model was more costly compared to the audiology-led and ENT models. Cost-savings for patients were achieved through salary differences and fewer clinic visits required under the audiology-led model compared to standard care. Audiology-led post-grommet models are a clinically appropriate and cost-efficient approach to optimise service delivery in resource-limited settings.

Greater distance to polling sites negatively impacts voter turnout, with minority communities experiencing the most severe effects. Despite this well-documented relationship, election boards and advocacy groups lack effective tools for selecting polling sites that provide equitable access to voting. We address this gap by developing a method that (1) assesses the suitability of a selection of sites and (2) optimally selects a specified number of sites among a set of potential locations. Our novel method employs the Kolm–Pollak equally distributed equivalent—a new metric from the environmental justice literature that was designed to rank distributions of disamenities, such as air pollution, when equity is a concern. This metric allows us to balance minimizing average distance to polls with improving access to those residents that are farthest from polls. It also enables principled comparisons among competing choices of polling sites and across demographic groups given a set of sites. We demonstrate our methods by analyzing early voting locations in DeKalb County, Georgia during the 2020, 2022, and 2024 elections. In particular, we investigate the distribution of the travel burden across demographic groups, how changes in sites between elections impacts those burdens, and how the early voting sites for each election compares to an optimal selection of sites. We also provide specific insight into where the county should seek to add sites and which sites are most critical for serving the community.

We developed an approach to quantify road infrastructure exposure to fault displacement hazard (FDH) that adopts Fault Displacement Hazard Analysis (FDHA) principles and addresses regional-scale challenges for emergency management and network planning. FDH results from ground deformation in surface-rupturing earthquakes and can damage nearby infrastructure. This study assesses New Zealand's road network exposure and vulnerability to FDH by generating displacement fields around faults based on historical earthquake data and physics-based models. Relative hazard is quantified as the product of 1D strain resolved along roads and normalised fault slip-rate, and was mapped at the regional scale. Vulnerability is assessed using geomorphon land class, or the road's position in the landscape, as a proxy for susceptibility to damage and repairability. Four areas emerged as potentially having higher risk: Wellington, the Eastern North Island, the West Coast, and the Upper South Island. We reviewed the local factors that influenced these measurements of hazard, exposure, and vulnerability. The Wellington Region is particularly vulnerable where roads intersect the Hutt Valley and Wairarapa Faults, and sections of State Highway 6 along the Alpine Fault in the West Coast are at heightened risk. This analysis underscores the need for regional approaches to fault displacement hazard and for targeted mitigation, such as fault avoidance, engineering solutions, and planning for rapid post-event repair. Our findings provide insights into horizontal infrastructure resilience planning for transport and other critical systems in seismically active areas.

Conventional tourism risk-to-life analyses require data that are often unavailable, inaccurate or out-of-date. Thus tourist fatalities can occur where risk analyses have not been completed, as at Whakaari/White Island volcano, New Zealand in 2019, when 22 people on commercial tours died because of a sudden eruption. Even where risk analyses have been carried out (as at Piopiotahi/Milford Sound, New Zealand, a World Heritage Site where thousands of lives are endangered by coseismic landslide-triggered tsunami) unacceptable risks remain because (i) analyses can be misinterpreted, (ii) risk calculations change with time, and (iii) sociopolitical factors influence risk management decision-making. Herein we (i) develop a methodology for reconnaissance ("average") risk-to-life assessment, to indicate where more complete assessments are required, and (ii) test this against (a) the empirical acceptability of risks following a disaster and (b) a series of risk analyses where no disaster has yet occurred. We develop an “average” risk-to-life assessment for Whakaari/White Island, and compare its outcomes with existing risk analyses for Piopiotahi/Milford Sound. It agrees with post-event risk acceptability at Whakaari/White Island, and with pre-event risk analyses at Piopiotahi/Milford Sound. We update the “average” risk analysis for Piopiotahi/Milford Sound by incorporating vulnerabilities of cruise ships and improved hazard data - the societal risk-to-life is unacceptable by several orders of magnitude. The only feasible strategy for achieving acceptable risk is drastically reducing visitor numbers. Socio-economic considerations suggest that this may not be societally practicable, which calls into question the way in which risk is used in planning to reduce tourism fatalities.

Background Rapidly developing pre-fire weather conditions contributing to sudden fire outbreaks can have devastating consequences. Accurate short-term forecasting is important for timely evacuations and effective fire suppression measures. Aims This study aims to introduce a novel machine learning-based approach for forecasting fire potential and to test its performance in the Sunshine Coast region of Queensland, Australia, over a period of 15 years from 2002 to 2017. Methods By analysing real-time data from local weather stations at a sub-hourly temporal resolution, we aimed to identify distinct weather patterns occurring hours to days before fires. We trained random forest machine learning models to classify pre-fire conditions. Key results The models achieved high out-of-sample accuracy, with a 47% higher accuracy than the standard fire danger index for the region. When simulating real forecasting conditions, the model anticipated 75% of the fires (11 out of 15). Conclusions This method provides objective, quantifiable information, enhancing the precision and effectiveness of fire warning systems. Implications The proposed forecasting approach supports decision-makers in implementing timely evacuations and effective fire suppression measures, ultimately reducing the impact of fires.