John Boland

Understanding the dynamics of high-intensity rainfall is crucial for assessing hydrological risks and informing climate adaptation strategies, particularly in regions vulnerable to climate change and variability, such as South Australia. This study examines the temporal variations in high-intensity rainfall events across four meteorological stations in Greater Adelaide, South Australia, Australia: Adelaide Airport, Kent Town, Hindmarsh, and Parafield. Daily rainfall data obtained from the Scientific Information for Land-Owners (SILO) database were analysed using the Bivariate Truncated Logistic-Geometric (BTLG) model to examine joint relationships between episode duration and peak intensity. Change-point analysis using the nonparametric Pettitt test was conducted to identify the time at which the effect of climate change on the intensity of daily rainfall became evident. The results indicate noticeable changes in the frequency and rainfall episodes, with key shifts around 1985 toward shorter-duration, generally lower-intensity events at several locations with daily data. The joint probability analysis p (peak > 20 mm, duration > 5 days) indicated a reduction in long-duration extreme events. This study highlights emerging trends of concentrated rainfall and their potential effects on urban runoff, flooding, and infrastructure design.

Electricity demand in South Australia is heavily influenced by population growth and temperature variations, yet most planning studies still focus on total demand rather than per capita demand. This research investigates monthly electricity demand in South Australia from 1999 to 2023, accounting for trends in population and temperature. Population data are interpolated from quarterly estimates to a monthly and daily series using quadratic regression. The analysis separates per‑capita demand into long‑term trend and seasonal components through additive time‑series models, comparing the fitted values with the original series to assess model accuracy. The study also employs change-point analysis and bivariate distribution analysis, along with conditional probability, to explore the relationship between duration and peak residuals for hot‑ and cold‑day responses, reflecting the growing impact of air conditioning and rooftop solar on net demand. Findings suggest that, after adjusting for population growth, per‑capita electricity demand in South Australia shows noticeable long‑term trends and an increased sensitivity to extreme heat. Extreme heat episodes are typically short in both periods before and after the change point. However, following the change point, they occur more frequently and exhibit stronger residuals. Prior to the change point, Tmax ≥ 35°C occurred on 325 days, resulting in 40 episodes, with residual peaks generally moderate, reaching a maximum of 0.00989. After the change point, Tmax ≥ 35°C occurred on 275 days, resulting in 47 episodes, with the temperature reaching a maximum of 47.7°C.

This study compares parametric statistical time series models, such as autoregressive moving average (ARMA), with nonparametric artificial neural networks, specifically long short‐term memory (LSTM) models, for univariate forecasting. Two time series are analyzed separately: wind power output from the Clements Gap wind farm and the regional electricity price for South Australia. One‐step‐ahead forecast performance is evaluated using normalized mean bias error (NMBE), normalized mean absolute error (NMAE), and normalized root mean square error (NRMSE). Three LSTM models were examined: a manually tuned model, a structurally equivalent model implemented in a different library, and a model with automated hyperparameter tuning. While LSTM models achieved competitive performance, statistical models often performed equally well or better. For price forecasts, the manually tuned LSTM achieved the lowest NMBE (), while ARMA(3,1) had the best NMAE (0.0253) and AR(6) the best NRMSE (0.220). For wind forecasts, the manually tuned LSTM again performed best overall (NMBE: 0.00034, NMAE: 0.143, NRMSE: 0.225), while the equivalent library model performed worst. These results highlight the need to subject nonparametric LSTM models to more rigorous and systematic evaluation relative to their parametric statistical counterparts.

The Murray Mallee (MM) and Upper Eyre Peninsula (UEP) are two important regions to the South Australian agricultural economy. The UEP is dominated by agricultural land use with a semi‐arid Mediterranean‐type climate, with low, variable annual rainfall. The climate is similar, with significant farming production around the Murray Mallee, adjacent to the Murray River. When investigating rainfall trends in the UEP region, annual rainfall has decreased considerably in the last 73 years, from 1950 to 2023, at all rain monitoring stations. From the 15 stations in the MM, most of the locations' rainfall is decreasing, but three stations, including Callington, Murray Bridge, and Mannum, show an increasing trend in annual rainfall. Monarto Safari Park station had only recent rainfall records; however, it also shows an increasing trend. The records for MM, particularly for gauges where rainfall has seen an increasing trend, seem to correspond with a revegetation program on the Monarto Plateau during the 1970s, and this may have been significant enough to increase rainfall in the surrounding regions, including Callington and Murray Bridge. Each station's annual rainfall, mean minimum temperature, and mean maximum temperature data are analyzed separately using trend analysis with exponential smoothing and three robust change‐point analysis methods in the MM and UEP regions. The significant change points observed in annual rainfall and temperature in each location and the overall reduction in rainfall may be attributed to industrial expansion, climate change, and shifts in land use patterns driven by urbanization, which have likely impacted water resources. This study investigated the relationship between land use and climate change that may impact South Australian rainfall patterns and increases in temperature over the last few decades.

Smart buildings are gaining popularity because they can enhance energy efficiency, lower costs, improve security, and provide a more comfortable and convenient environment for building occupants. A considerable portion of the global energy supply is consumed in the building sector and plays a pivotal role in future decarbonization pathways. To manage energy consumption and improve energy efficiency in smart buildings, developing reliable and accurate energy demand forecasting is crucial and meaningful. However, extending an effective predictive model for the total energy use of appliances at the building level is challenging because of temporal oscillations and complex linear and non-linear patterns. This paper proposes three hybrid ensemble predictive models, incorporating Bagging, Stacking, and Voting mechanisms combined with a fast and effective evolutionary hyper-parameters tuner. The performance of the proposed energy forecasting model was evaluated using a hybrid dataset comprising meteorological parameters, appliance energy use, temperature, humidity, and lighting energy consumption from various sections of a building, collected by 18 sensors located in Stambroek, Mons, Belgium. To provide a comparative framework and investigate the efficiency of the proposed predictive model, 15 popular machine learning (ML) models, including two classic ML models, three NNs, a Decision Tree (DT), a Random Forest (RF), two Deep Learning (DL) and six Ensemble models, were compared. The prediction results indicate that the adaptive evolutionary bagging model surpassed other predictive models in both accuracy and learning error. Notably, it achieved accuracy gains of 12.6%, 13.7%, 12.9%, 27.04%, and 17.4% compared to Extreme Gradient Boosting (XGB), Categorical Boosting (CatBoost), GBM, LGBM, and Random Forest (RF).

Floating hybrid wind-wave systems combine offshore wind platforms with wave energy converters (WECs) to create cost-effective and reliable energy solutions. Adequately designed and tuned WECs are essential to avoid unwanted loads disrupting turbine motion while efficiently harvesting wave energy. These systems diversify energy sources, enhancing energy security and reducing supply risks while providing a more consistent power output by smoothing energy production variability. However, optimising such systems is complex due to the physical and hydrodynamic interactions between components, resulting in a challenging optimisation space. This study uses a 5-MW OC4-DeepCwind semi-submersible platform with three spherical WECs to explore these synergies. To address these challenges, we propose an effective ensemble optimisation (EEA) technique that combines covariance matrix adaptation, novelty search, and discretisation techniques. To evaluate the EEA performance, we used four sea sites located along Australia's southern coast. In this framework, geometry and power take-off parameters are simultaneously optimised to maximise the average power output of the hybrid wind-wave system. Ensemble optimisation methods enhance performance, flexibility, and robustness by identifying the best algorithm or combination of algorithms for a given problem, addressing issues like premature convergence, stagnation, and poor search space exploration. The EEA was benchmarked against 14 advanced optimisation methods, demonstrating superior solution quality and convergence rates. EEA improved total power output by 111%, 95%, and 52% compared to WOA, EO, and AHA, respectively. Additionally, in comparisons with advanced methods, LSHADE, SaNSDE, and SLPSO, EEA achieved absorbed power enhancements of 498%, 638%, and 349% at the Sydney sea site, showcasing its effectiveness in optimising hybrid energy systems.

Sediment transport in urban streams is a critical environmental issue, with significant implications for water quality, ecosystem health, and infrastructure management. Accurately estimating suspended sediment concentration (SSC) is essential for effective long-term environmental management. This study investigates the relationships between streamflow, turbidity, and SSC in Dry Creek, South Australia, to understand sediment transport dynamics in urbanised catchments. We collected grab samples from the field and analysed them in the laboratory. We employed statistical modelling to develop a sediment rating curve (SRC) that provides insights into the sediment transport dynamics in the urban stream. The grab sample measurements showed variations in SSC between 3.2 and 431.8 mg/L, with a median value of 77.3 mg/L. The analysis revealed a strong linear relationship between streamflow and SSC, while turbidity exhibited a two-regime linear relationship, in which the low-turbidity regime demonstrated a stronger linear relationship compared to the high-turbidity regime. This is attributed to the urbanised nature of the catchment, which contributes to a first-flush effect in turbidity. This contributes to sediment hysteresis, resulting in non-proportional turbidity and SSC responses to streamflow changes. The findings demonstrate the effectiveness of a streamflow-based SRC for accurately predicting sediment discharge, explaining 97% of the variability in sediment discharge. The sediment discharge predicted using the SRC indicated a sediment load of 341.8 tonnes per year along the creek. The developed sediment rating curve provides a valuable tool for long-term sediment management in Dry Creek, enabling the assessment of downstream environmental risks. By addressing data limitations, this study contributes to a deeper understanding of sediment transport dynamics in urbanized environments, offering insights for informed decision-making and effective sediment management strategies.