Australian Urban Research Infrastructure Network

Polynyas, regions of open water enclosed by sea ice, are persistent features near the Antarctic coast as well as in the pack ice. Waves are known to occur within polynyas. If a polynya is sufficiently separated from the “blue” Southern Ocean by pack ice, then it can be considered isolated from Southern Ocean waves. Wave energy in isolated polynyas must be generated locally. During offshore wind conditions, a polynya could provide a long fetch for waves to develop, and the wind-waves may be steep enough to break the ice pack from the inside outward. This is in contrast to the typical focus of wave-induced sea ice break-up from the outside-inward with waves originating from the Southern Ocean. Here, we present our investigation of this inside-out sea-ice erosion mechanism based on buoy measurements of waves in the Vincennes Bay Polynya, East Antarctica. The measurements confirm the presence of energetic locally generated waves, which appear to be sufficiently steep to break the ice at the polynya edge. Further, we evaluate the wave-induced sea-ice break-up potential in this recurring polynya over the past two decades. Our results confirm the importance of locally generated waves in Antarctic polynyas. This highlights the previously overlooked potential of waves to accelerate sea-ice loss from within the pack ice, contributing to the recent Antarctic sea-ice decline.

Public transportation systems play a vital role in modern cities, but they face growing security challenges, particularly related to incidents of violence. Detecting and responding to violence in real time is crucial for ensuring passenger safety and the smooth operation of these transport networks. To address this issue, we propose an advanced artificial intelligence (AI) solution for identifying unsafe behaviours in public transport. The proposed approach employs deep learning action recognition models and utilises technologies like NVIDIA DeepStream SDK, Amazon Web Services (AWS) DirectConnect, local edge computing server, ONNXRuntime and MQTT to accelerate the end-to-end pipeline. The solution captures video streams from remote train stations closed circuit television (CCTV) networks, processes the data in the cloud, applies the action recognition model, and transmits the results to a live web application. A temporal pyramid network (TPN) action recognition model was trained on a newly curated video dataset mixing open-source resources and live simulated trials to identify the unsafe behaviours. The base model was able to achieve a validation accuracy of 93% when trained using open-source dataset samples and was improved to 97% when live simulated dataset was included during the training. The developed AI system was deployed at Wollongong Train Station (NSW, Australia) and showcased impressive accuracy in detecting violence incidents during an 8-week test period, achieving a reliable false-positive (FP) rate of 23%. While the AI correctly identified 30 true-positive incidents, there were 6 cases of false negatives (FNs) where violence incidents were missed during the rainy weather suggesting more data in the training dataset related to bad weather. The AI model’s continuous retraining capability ensures its adaptability to various real-world scenarios, making it a valuable tool for enhancing safety and the overall passenger experience in public transport settings.

Wind-driven rain (WDR) is a primary cause of material degradation in tall buildings that affects the durability and long-term performance. This study investigates the influence of building shape on WDR exposure in tall buildings using computational fluid dynamics (CFD) simulations. Results indicate that building shape influences local flow conditions, which, in turn, influence the trajectory of rain droplets and their impingement on building surfaces. Two specific flow features were found to dictate WDR exposure: the wind-blocking effect and the separation of shear layers at leading edges of the building. Streamlined geometries with small wind-blocking regions experienced higher WDR exposure on windward surfaces. High WDR concentrations also occurred on geometric features protruding into the wind and at locations where shear layers impinge on the building surface. These findings are based on steady Reynolds-Averaged Navier–Stokes (RANS) simulations that do not consider unsteady flow features such as buffeting and vortex shedding. Nonetheless, the study provides valuable insight into the influences of building shape on WDR exposure, which could lead to better weatherproofing of these buildings.

Abstract Tropical cyclones dissipate large amounts of energy into the upper ocean, locally enhancing vertical mixing and cooling the sea surface. In this study, we investigate how the response of the ocean to tropical cyclones is affected by additional mixing from unbroken surface waves. This “Surface Wave Mixing” (SWM) is represented by a novel parameterization, in which the wave orbital motion contributes directly to the production of turbulent kinetic energy. The parameterization is implemented here as a modification to the k‐ε turbulence scheme, used within an ocean model with 1/4° horizontal resolution (MOM5). This model is forced with idealized tropical cyclone wind fields based on observed case studies. Relative to simulations without SMW, the inclusion of SWM leads to surface temperature differences of around 0.5°C near the storm track, typically with warm anomalies on the side with the strongest winds and cool anomalies in other regions. This pattern is explained by an initial wave‐induced deepening of the mixed layer, which can modify the subsequent shear‐induced entrainment and upwelling. The temperature anomalies from SWM could potentially influence tropical cyclone intensity and structure.