60419 research outputs found
Sort by
Information from Earth Observation for the Management of Sustainable Land Use and Land Cover in Brazil: An Analysis of User Needs
No full textBrazil has some of the world’s most important forest and natural ecosystem resources and
their sustainability is of global importance. The expansion of agriculture for livestock, the extractive
industries, illegal logging, land conflicts, fire and deforestation are pressures on land use and drivers
of land use change in many regions of Brazil. While different institutions in Brazil have sought to
use Earth Observation (EO) data to support better land use management and conservation projects,
several problems remain at the national and state level in the implementation of EO to support
environmental policies and services provided to Brazilian society. This paper presents the results of
a systematic analysis of the key challenges in using EO data in land management in Brazil and
summarises them in a conceptual model of the factors influencing EO data use for assessing
sustainable land use and land cover in Brazil. The research was based on a series of in‐depth, semistructured
interviews (43) and structured interviews (53) with key stakeholders who make use of
EO data across different locations in Brazil. The major challenges identified in the complex and
multifaceted aspects of using this information were associated with access to, and with the
processing of, raw data into usable information. The analysis also revealed novel insights on a lack
of inter‐institutional communication, adequate office infrastructure and personnel, availability of
the right type of EO data and funding restrictions, political instability and bureaucracy as factors
that limit more effective use of EO data in Brazil at present. We close this analysis by considering
how EO information for the sustainable management of land use and land cover can assist
institutions as they respond to the varied political and economic instabilities affecting
environmental governance and deforestation levels
Locating Female ‘Voices’ in the Minamata Convention on Mercury in Sub-Saharan Africa: The Case of Ghana
No full textCountries that have ratified the Minamata Convention on Mercury, a United Nations-backed international treaty designed to protect human health and the environment from releases of mercury and mercury compounds, are required to produce a National Action Plan (NAP). Each must state, very clearly, how the mercury being used at artisanal and small-scale gold mines will be phased out. In most areas of sub-Saharan Africa, however, devising a comprehensive NAP promises to be an enormous and indeed, challenging, undertaking. Here, the institutional capacity and resources, expertise and at times, commitment needed to capture the level of detail the Minamata Secretariat expects to be included in each NAP are woefully lacking. One of the more challenging tasks ahead, given the shortage of hard data available on the sector’s populations, production and activities, promises to be the design and implementation of appropriate educational, communication and support-related strategies for the ‘vulnerable populations’ who rely on work at artisanal and small-scale gold mines for their incomes. This is especially significant for women, who, despite accounting for at least 50 percent of the region’s artisanal and small-gold mine workforce, mostly carry out the manual work at the lower tiers of the sector’s labour hierarchies. Taking stock of this largely ‘invisible’ work and the circumstances driving individuals to pursue employment in this sector in the first place, this paper reflects critically on the challenges with reducing women’s exposure to mercury at artisanal and small-scale gold mines in sub-Saharan Africa. It draws on findings from ongoing research in Ghana, the location of one of the largest and most dynamic artisanal and small-scale gold mining sectors in the region
Probiotic supplementation increases carbohydrate metabolism in trained male cyclists: a randomized, double-blind, placebo-controlled cross-over trial
No full textWe hypothesised that probiotic supplementation (PRO) increases the absorption and oxidation of
orally ingested maltodextrin during 2h endurance cycling, thereby sparing muscle glycogen for a
subsequent time trial (simulating a road race). Measurements were made of lipid and carbohydrate
oxidation, plasma metabolites and insulin, gastrointestinal permeability, and subjective symptoms of
discomfort. Seven male cyclists were randomized to PRO (bacterial composition given in methods) or
placebo (PLC) for four weeks, separated by a 14-day washout period. After each period, cyclists
consumed a 10% maltodextrin solution (initial 8 mL·kg-1 bolus and 2 mL·kg-1 every 15 min) while
exercising for 2h at 55% Wmax followed by a 100 kJ time trial. PRO resulted in small increases in peak
oxidation rates of the ingested maltodextrin (0.84 ± 0.10 vs 0.77 ± 0.09 g·min-1, P = 0.016), and mean
total carbohydrate oxidation (2.20 ± 0.25 vs 1.87 ± 0.39 g·min-1, P = 0.038), while fat oxidation was
reduced (0.40 ± 0.11 vs 0.55 ± 0.10 g·min-1, P = 0.021) . During PRO small but significant increases were
seen in glucose absorption, plasma glucose and insulin concentration and decreases in NEFA and
glycerol. Differences between markers of GI damage and permeability and time trial performance
were not significant (P > 0.05). In contrast to the hypothesis, PRO led to minimal increases in
absorption and oxidation of the ingested maltodextrin and small reductions in fat oxidation, while
having no effect on subsequent time trial performance
Kinetic and mechanistic aspects of ultrafiltration membrane fouling by nano- and microplastics
No full textThe mechanical and chemical breakdown of plastic litter increases the release of nano- and microplastics, which have the potential to impact the performance of membranes processes used in water treatment plants. In this work, the extent of fouling of a commercial ultrafiltration poly(sulfone) membrane induced by nano- and microplastics ranging from 13 to 690 nm in size was investigated. The cross-flow filtration of the plastic particles over 48 h at a 1 bar pressure reduced the permeate water flux by 38 % compared to pure water filtration. Over 25 % of the nano- and microplastics initially present in the feed were absorbed onto the membrane surface within the 48 h of filtration. Particulate fouling mechanism was sequentially modelled into intermediate and complete pore blockage, followed by cake layer formation. Hydrophobic interactions and surface repulsion forces were found to dictate the adsorption rate of the nano- and microplastics onto the membrane surface. This work opens the understanding of NPs/MPs interactions with water filtration processes and demonstrates the need to develop solutions limiting the impact of NPs/MPs on current treating units
Gated Variational AutoEncoders: Incorporating Weak Supervision to Encourage Disentanglement
No full textVariational AutoEncoders (VAEs) provide a means
to generate representational latent embeddings. Previous research has highlighted the benefits of achieving representations that are disentangled, particularly for downstream tasks. However, there is some debate about how to encourage disentanglement with VAEs, and evidence indicates that existing implementations do not achieve disentanglement consistently. The evaluation of how well a VAE’s latent space has been disentangled is often evaluated against our subjective expectations of which attributes should be disentangled for a given problem. Therefore, by definition, we already have domain knowledge of what should be achieved and yet we use
unsupervised approaches to achieve it. We propose a weakly supervised approach that incorporates any available domain knowledge into the training process to form a Gated-VAE. The process involves partitioning the representational embedding
and gating backpropagation. All partitions are utilised on the forward pass but gradients are backpropagated through different partitions according to selected image/target pairings. The approach can be used to modify existing VAE models such as beta-VAE, InfoVAE and DIP-VAE-II. Experiments demonstrate that using gated backpropagation, latent factors are represented in their intended partition. The approach is applied to images of faces for the purpose of disentangling
head-pose from facial expression. Quantitative metrics
show that using Gated-VAE improves average disentanglement, completeness and informativeness, as compared with un-gated implementations. Qualitative assessment of latent traversals
demonstrate its disentanglement of head-pose from expression, even when only weak/noisy supervision is available
How Will Service Robots Redefine Leadership in Hotel Management? A Delphi Approach
No full textPurpose – Using the Delphi technique, this paper aims to investigate how human resource experts perceive service robots will impact leadership and human resource management in the hospitality industry.
Design/methodology/approach – A three-stage Delphi study with hotel industry human resource experts was conducted to identify the key trends and major challenges that will emerge in the next ten years and how leaders should deal with the challenges brought about by service robot technologies.
Findings – Results show that while service robots are anticipated to increase efficiency and productivity of hotel activities, they may also pose challenges such as high costs, skill deficits, and significant changes to the organizational structure and culture of hotels. Therefore, the anticipated applications and integration of robotic technology will require leaders of the future to carefully consider the balance between the roles of service robots and human employees in the guest experience, and to nurture a work environment that embraces open-mindedness and change.
Originality/value – This is the first type of study to examine hospitality leadership and human resource management in the context of robotized hotels. This study has taken an important step to understand the leadership role in robotized hotels from a human resource perspective, and brings clarity as to how robotic technology can influence leadership in the future workplace
Metrology for Quantitative Imaging and Molecular Radiotherapy
No full textThis thesis investigated quantitative SPECT/CT imaging of lutetium-177 (177Lu): assessing and quantifying variation in reported activity between clinical sites to help inform researchers attempting to compare quantitative imaging and molecular radiotherapy dosimetry performed
at different clinical sites and to provide guidance on which methods and parameters to use in order to minimise activity uncertainty.
The initial work involved assisting in the development of a practical, consensus-based calibration protocol for quantitative SPECT/CT imaging, followed by a pan-European hospital inter-comparison exercise in which participants were required to quantify the activity in a 177Lu source using local clinical protocols. The calibration protocol was also tested at each site. Finally, all calibration, imaging, reconstruction and segmentation methods and parameters used in the local clinical protocols were replicated on a single gamma camera to assess the impact of
each method and parameter on activity quantification and make recommendations as to which methods and parameters to use for quantitative SPECT/CT imaging of 177Lu.
The activities measured by the participants (seven clinical sites) differed from the true activity by up to 70% for an active volume surrounded by a less active shell and by up to 316% for the less active shell. Use of the calibration protocol (five clinical sites) reduced the difference
between true and measured activity from a maximum of 19% to 12% for the active volume but increased the difference for the less active shell, from a maximum of 117% to 235%.
The methods comparison work compared and identified methods and parameters that increased the accuracy of the measured activity volumes and values. In particular, the preferable calibration method was demonstrated to be one that uses objects of a similar size and geometry to the regions of interest for which quantification is required. In situations where this is not feasible, the use of planar Petri dishes performed best as a generic calibration
method. Large volume dispersed sources performing best for quantifying areas of low activity located adjacent to areas of high activity. The use of the 208 keV photopeak only is
recommended, as inclusion of the 113 keV photopeak decreases accuracy of quantitation. Careful use of post-reconstruction filters has been shown to have a beneficial effect on quantitative accuracy. Segmentation of image data should be performed using the CT data rather than SPECT data for volume delineation where possible, with thresholding based on the SPECT data followed by adjustments using the CT data giving improved results for activity quantification.
This work revealed a general tendency to overestimate the accuracy with which volumes and activities can be reported. Uncertainties were underestimated by several of the participants in the inter-comparison exercise and the calibration protocol that was tested also gave underestimated uncertainties.
Finally, the conclusions, recommendations and clinical impact of this work is presented, along with details of further work that could be performed to contribute to the body of research in this area
A multi-source account for affective responses to processing fluency
No full textProcessing fluency has been shown to be flexible metacognitive cue for a range of judgements including truth, familiarity, and trust. Amongst these, affect judgements are of particular interest as 1) affect can be genuinely evoked by fluency, and 2) affect can be used as a cue for other judgements. However, there is disagreement towards the pattern of affective responses arising from fluency. The hedonic marking hypothesis (Winkielman, Schwarz, Fazendeiro, & Reber, 2003) suggests that fluency is fundamentally positive, whilst the fluency amplification account (Albrecht & Carbon, 2014) suggests that the affective response can be positive or negative, depending on (and congruent with) the valence of stimuli being exposed to. Whilst these accounts have been used as competing explanations, this thesis argues that they both contribute to overall affective responses in a novel multi-source account.
This thesis developed a novel set of business scenarios to manipulate fluency (using coherence) and valence (using risk). Evidence from three approaches is presented: 1) Meta-Analysis examining affective responses to fluency, with a sample of 108 publications (k = 591 effect sizes), 2) Five behavioural experiments, and 3) Facial electromyography (fEMG). Across these approaches, neither hedonic marking nor fluency amplification in isolation could account for the full pattern of results. Instead, results were explained by the combined contribution of the two models, as predicted by the multi-source account. The unique findings were uncovered by manipulating stimuli valence, as well as separately measuring positive and negative affect, an approach not previously investigated in the literature, thereby adding methodological, as well as theoretical, contributions to the literature on fluency effects. Implications for future research are to adopt a separate measurement approach to investigate wider judgement domains, whilst practical implications for business assessment and agenda setting are also discussed
Dynamic pseudonym management for privacy preservation in vehicular communication systems.
No full textResearch into the established area of Intelligent Transportation System (ITS) is still evolving and fast-moving fuelled in part by rapid changes in vehicular technologies and mobile communication network. A practical challenge is that many vehicular communication applications and services make use of the Basic Safety Messages that contain the vehicle location and other personal data. A popular way of dealing with this privacy issue is to utilise a pseudonym change scheme to protect the vehicle’s identity and location. However, many such schemes suffer from poor scalability and the certificate management difficulty raises with the number of pseudonyms generated and stored in the ITS system, casting doubt of the economic feasibility of such approaches.
In this thesis, firstly, Cooperative Pseudonym Change Scheme (CPCS) for VCS is proposed for providing high level of anonymity. In this model, the Road Side Units (RSUs) is the key factor by storing generated pseudonym sets from the Trusted Authority (TA), then delivering sets of pseudonyms to vehicles which has similar context. Secondly a Decentralised Pseudonym Management System (DPMS) Using Distributed Ledger Technology is proposed to overcome limits that CPCS and most other pseudonym change schemes of low scalability and high cost of managing large numbers of pseudonym certificates in ITS. This scheme consists of pseudonym distribution and a shuffle operation, allowing the re-distribution and re-usage of existing pseudonyms by different vehicles. The results demonstrate that the proposed scheme achieves a better degree of anonymity and low request delay at a lower cost than existing schemes. Thirdly a Distributed Pseudonym Certificate Revocation Scheme (DPCRS) for VCS is proposed based on the framework that DPMS introduced. The proposed scheme embeds part of the certificate revocation functions within the security and privacy applications, aiming to reduce the communication overhead and shorten the processing time cost. Extensive simulations and analysis show the effectiveness and efficiency of the three proposed schemes
Data sources:
Access Repository Dashboard
Do you manage Surrey Research Insight? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!