Dr Ikpe Ibanga

This study investigated the performance of four pilot-scale biofilters for the removal of bioaerosols from waste airstreams in a materials recovery facility (MRF) based in Leeds, UK. A six-stage Andersen sampler was used to measure the concentrations of four groups of bioaerosols (Aspergillus fumigatus, total fungi, total mesophilic bacteria and Gram negative bacteria) in the airstream before and after passing through the biofilters over a period of 11 months. The biofilters achieved average removal efficiency (RE) of 70% (35 to 97%) for A. fumigatus, 71% (35 to 94%) for total fungi, 68% (47 to 86%) for total mesophilic bacteria and 50% (-4 to 85%) for Gram negative bacteria, provided that the inlet concentration was high (103-105 cfu m-3), which is the case for most waste treatment facilities. The performance was highly variable at low inlet concentration with some cases showing an increase in outlet concentrations, suggesting that biofilters had the potential to be net emitters of bioaerosols. The gas phase residence time did not appear to have any statistically significant impact on bioaerosol removal efficiency. Particle size distribution varied between the inlet and outlet air, with the outlet having a greater proportion of smaller sized particles that represent a greater human health risk as they can penetrate deep into the respiratory system where gaseous exchange occurs. However, the outlet concentrations were low and would further be diluted by wind in full scale applications. In conclusion, this study shows that biofilters designed and operated for odour degradation can also achieve significant bioaerosol control in waste gas.

Abstract

The bioaerosol exposure data from the study by Akpeimeh, Fletcher, and Evans (2019) was used to compute the risk of infection from the exposure of dumpsite workers to Aspergillus fumigatus and Escherichia coli O157:H7. A stochastic (Markov Chain) model was used to model the transport of the inhaled dose though the human respiratory system and then integrated into the beta‐Poisson dose–response model to estimate workers risks of respiratory and gastrointestinal (GI) infection. The infection risk was computed based on workers exposure to E. coli O157:H7 at 10–50% pathogen ingestion rate and pathogen‐indicator ratio (P:I) of 1:103 and 1:104, while exposure to A. fumigatus was based solely on the average initial exposure dose. The results showed that after 11 hours of exposure, workers engaged in scavenging, waste sorting, and site monitoring were at risk of respiratory and GI infection in the magnitude of 10−1. However, the risk estimates associated with specific areas of the dumpsite showed that, the risk of GI infection at the active area ranged between 3.23 × 10−3–1.56 × 10−2 and 3.25 × 10−4–1.62 × 10−3; dormant area 2.06 × 10−3–1.01 × 10−2 and 2.09 × 10−4–1.04 × 10−3; entrance 1.85 × 10−3–9.09 × 10−3 and 1.87 × 10−4–9.27 × 10−4; boundary 1.82 × 10−3–8.82 × 10−3 and 2.09 × 10−4–8.94 × 10−4 for P:I = 1:103 and 1:104 respectively, while the risk of respiratory infection risks were in the magnitude of 10−1 for all four locations. The estimated risk of workers developing respiratory and gastrointestinal infections were high for all activities assessed at the dumpsite.

This book will highlight the role of CE in the sustainability field as it is expressed in the various fields and disciplines and its contribution to building a sustainable society by providing a better understanding of the relevant social ...

The relevance of traditional dissertation models is increasingly being challenged by the rapid evolution of generative AI technologies such as ChatGPT and similar tools. This study explores the implications of these technologies for dissertation practice within built environment pedagogy. Specifically, it examines the positive and negative impacts of generative AI on students’ dissertation work and the influence of these technologies on dissertation assessment processes. Dissertation or thesis writing remains a core requirement for degree completion at both undergraduate and postgraduate levels, serving as evidence of students’ academic competence and disciplinary understanding. Within built environment disciplines, the dissertation has long been a key component of pedagogical practice. However, the emergence of generative AI necessitates a critical reassessment of how dissertations are conceptualised, undertaken, and evaluated. This study aims to: 1. Evaluate existing models for assessing dissertations within built environment pedagogy; 2. Examine the potential impacts of generative AI on the quality and authenticity of dissertations; 3. Investigate the influence of generative AI on dissertation assessment practices; and 4. Propose revised assessment models suited to the evolving educational landscape. Using a focus group research approach involving selected undergraduate and postgraduate students, as well as academics in the built environment disciplines, initial findings highlight the urgent need to reconsider and adapt dissertation assessment frameworks in response to the transformative effects of generative AI. The study underscores the need to rethink dissertation supervision and assessment within built environment pedagogy. It calls for the development of ethical and process-oriented assessment frameworks. These insights aim to guide the evolution of dissertation practices that remain both relevant and credible in the AI era. Keywords: Built environment pedagogy, AI, dissertation, assessment

Recent debates on Generative Artificial Intelligence (GenAI) have centred on quantifiable concerns such as computational cost, carbon emissions, and benchmark performance. Yet the most consequential risks may be those that are less visible: the gradual reshaping of human cognition, creativity, and epistemic trust. This paper introduces the concept of AI as cognitive ecology, situating generative systems not merely as tools or agents, but as a pervasive environment in which thought now unfolds. Building on this paradigm, we propose the HORIZON taxonomy of invisible costs: Homogenization, Offloading (deskilling), Resource externalities, Information integrity, Zoomed-in feedback loops, Organisational memory loss, and Normative drift. We illustrate each dimension through conceptual analysis and lightweight audits, and propose new indicators including DAO (Diversity of AI Outputs), CDQ (Cognitive Dependence Quotient), EIS (Epistemic Integrity Score), and RTE (Resource Transparency Equivalent). We argue that sustaining AI innovation requires not only technical and environmental monitoring, but also active stewardship of cognitive ecologies. Published version available in Journal of Artificial Intelligence and AI Ethics (2026).