Climate change poses an imminent physical risk to cultural heritage sites and their surrounding landscape through intensifying environmental processes such as damaging wetting and drying cycles that disrupt archaeological preservation conditions, and soil erosion which threatens to expose deposits and alter the archaeological context of sites. In the face of such threats, geospatial techniques such as GIS, remote sensing, and spatial modelling have proved invaluable tools for archaeological research and cultural heritage monitoring. This paper presents the application of secondary multi-source and multi-temporal geospatial data within a processing framework to provide a comprehensive assessment of geophysical risk to the Roman fort of Magna, Carvoran, UK. An investigation into the ancient hydraulic system at Magna was carried out with analysis of vegetation change over time, and spatio-temporal analysis of soil erosion risk at the site. Due to COVID-19 restrictions in place at the time of this study, these analyses were conducted using only secondary data with the aim to guide further archaeological research, and management and monitoring strategies for the stakeholders involved. Results guided inferences about the ancient hydraulic system, providing insights regarding how to better manage the site at Magna in the future. Analysis of soil erosion allowed the identification of hot spot areas, indicating a future increase in rates of erosion at Magna and suggesting a seasonal period of higher risk of degradation to the site. Results have proven that freely available multi-purpose national-scale datasets are sufficient to create meaningful insights into archaeological sites where physical access to the site is inhibited. This infers the potential to carry out preliminary risk assessment to inform future site management practices.

Accurately forecasting air quality could lead to the development of dynamic, data-driven policy-making and improved early warning detection systems. Deep learning has demonstrated the potential to produce highly accurate forecasting models, but it is noted that much literature focuses on narrow datasets and typically considers one geographic area. In this research, three diverse air quality datasets are utilised to evaluate four deep learning algorithms, which are feedforward neural networks, Long Short-Term Memory (LSTM) recurrent neural networks, DeepAR and Temporal Fusion Transformers (TFTs). The study uses these modules to forecast CO, NO₂, O₃, and particulate matter 2.5 and 10 (PM_2.5, PM₁₀) individually, producing a 24 h forecast for a given sensor and pollutant. Each model is optimised using a hyperparameter and a feature selection process, evaluating the utility of exogenous data such as meteorological data, including wind speed and temperature, along with the inclusion of other pollutants. The findings show that the TFT and DeepAR algorithms achieve superior performance over their simpler counterparts, though they may prove challenging in practical applications. It is noted that while some covariates such as CO are important covariates for predicting NO₂ across all three datasets, other parameters such as context length proved inconsistent across the three areas, suggesting that parameters such as context length are location and pollutant specific.

A Service Level Agreement (SLA) is a commitment between a client and provider that assures the quality of service (QoS) a client can expect to receive when purchasing a service. However, evidence of SLA violations in Internet of Things (IoT) service monitoring data can be manipulated by the provider or consumer, resulting in an issue of trust between contracted parties. The following research aims to explore the use of blockchain technology in monitoring IoT systems using smart contracts so that SLA violations captured are irrefutable amongst service providers and clients. The research focuses on the development of a Java library that is capable of generating a smart contract from a given SLA. A smart contract generated by this library is validated through a mock scenario presented in the form of a Remote Patient Monitoring IoT system. In this scenario, the findings demonstrate a 100 percent success rate in capturing all emulated violations.