Lentil (Lens culinaris Medik.) is an important crop for providing a source of dietary protein globally and is produced over three distinct agroecological zones; Mediterranean, sub-tropical and temperate. Crop production is constrained by abiotic and biotic stresses, including acute high temperature (HT). For Mediterranean-type climates, such as southern Australia, low and unreliable winter rainfall (200 – 600 mm/year) and frequent acute HT events (heat waves) during the reproductive phase, limit production with yield losses up to 70%. Consequently, increasing the adaptation of commercial lentil varieties within contemporary breeding programs, by utilising global germplasm adapted to HT environments, is the critical next step. This project screened 135 lentil genotypes (global landraces and commercial cultivars) for tolerance to acute HT occurring during the reproductive period. Screening was through a combination of late sowing (summer - chronic and acute HT), controlled-environment (acute HT) and subsequent in-season validation (winter - acute HT), where a daily maximum temperature of greater than 30 degrees Celsius was classified as HT tolerant. Genotypes selected for screening were predominately from regions where HT occurs during the reproductive period (i.e. Mediterranean and sub-tropical). Across HT treatments ranging from acute (3 days) to chronic HT during the reproductive period, we observed that HT caused a 48% reduction in grain yield across genotypes screened, which translated to an average reduction of 0.14 and 0.19% per degree (>30 degrees Celsius) for global landraces and commercial cultivars respectively. We identified 15 landraces and the commercial cultivar, Nipper to have a high level of yield stability under HT. Within the 15 landraces identified, AGG 73838 and 73154 consistently exhibited HT tolerance under the multiple screening strategies employed in this study. The additional 13 landraces were identified within the late sowing field screening process. The identification of these 15 HT tolerant landraces provides a valuable source of material that can be immediately utilised by Australian lentil breeding programs. Optimal screening approaches for HT tolerance were also identified in this study, where field-based and controlled environment screening methods were tested and their utility to breeding programs, assessed. To align with current breeding strategies, several requirements were considered important; 1) reliable methods for identifying genetic diversity to HT response within global material, 2) high-throughput potential and (or) ability to measure parameters efficiently, and 3) methods to enable screening of large populations. Through assessing a range of screening methodologies, we determined that a multi-stage screening process integrating late sowing over spring/summer with subsequent winter validation, is required. This process enables the shortlisting of HT tolerant germplasm from a broad range of material and further validation of phenotype response in field conditions. To rank genotypes based on HT tolerance we determined that the indices, stress tolerant index (STI) and high temperature tolerance index (HTTI), which integrate the absolute and relative response of genotypes within the test population, were an effective means of ranking HT tolerance. The final component of this project assessed the combined effects of available soil water, night-time temperature and carbon dioxide concentration on HT response in lentil, which provided insight to the impact of other abiotic constraints associated with climate change. We determined that for HT, lentil response varied with available soil water and soil type. For lentil grown in a sandy loam soil, when HT and high water occurred together, HT caused a significant reduction in yield (33%), whereas under low water the application of HT did not further reduce yield. In contrast, for a clay soil there was a significant reduction in yield due to HT across both high and low water treatments. This highlights the effect of soil type on crop water availability and the potential variable response to acute HT depending on soil type and rainfall patterns of a growing region. For the impact of high night temperature, we observed no effect of night temperature when it occurs in conjunction with high day temperature. For the collective effect of HT and carbon dioxide concentration, elevated carbon dioxide did not alter the pattern of plant response to acute HT. This suggests that for lentil, the effects of HT are unlikely to be exacerbated or reduced under elevated carbon dioxide levels. Ultimately, this project contributes to the Australian lentil industry by identifying landraces with HT tolerance and through the development of a screening methodology that can be adopted by current breeding programs to increase the HT tolerance of future commercial lentil cultivars.

Demand exists for food products formulated specifically for older adult's tastes, constraints and nutritional needs. This thesis undertook research and development of minced meat products for older consumers and used novel biometrics and consumer science to evaluate prototypes. Older adult's food consumption behaviour was investigated (Chapter 3) to understand desirable characteristics and demographic influences. Australians mostly snacked throughout the day and lacked regular mealtimes. Chinese ate regular meals and occasionally snacked. For both groups, texture and flavour were key food choice drivers and eating difficulty with some meat products was evident. Focus of subsequent experiments was on development of convenient and healthy products to enhance older consumers wellbeing. The effects of cooking method, fat concentration and food additives/ingredients on physicochemical/nutritional attributes of minced products fortified with sugarcane fibre were explored in Chapters 4 and 5. Hardness of meatballs decreased with increasing fat content, and also with boiling water as cooking method. Post-cooking nutrient loss was lower when meatballs were oven baked (Chapter 4). Therefore, changes in formulation and cooking method can be used to improve quality traits of minced beef products for elderly consumers. Further development of beef patties formulations (Chapter 5) was performed using a wide range of ingredients, including gums and starch, to achieve varying textures from soft to medium and hard, for subsequent sensory evaluation. Consequentially, potato starch and Australian native ingredients were selected for the creation of desired soft texture and interesting flavour. Conventional and novel sensory techniques were used in Chapters 6 and 7 to evaluate the acceptability of beef patties varying in texture, with or without sauce, between older and younger consumers. Check-All-That-Apply and hedonic tests (Chapter 6) confirmed that both consumer groups preferred the softer patty for all the attributes assessed, regardless of sauce addition. There was an increase in appearance and texture liking of samples for younger consumers, with sauce, but it did not influence liking for older consumers. Biometrics and purchase intent (Chapter 7) were used to investigate differences in facial expressions, heart rate, and willingness-to-buy, between younger and older consumers in response to the same samples used in Chapter 6, and concept products, respectively. A harder texture led to a decrease in the happy facial expression for younger consumers and increased the angry facial expression of older consumers. Young consumers were generally more interested in purchasing healthy and unique products, whilst the traditional/familiar trait of products was the most important for older adults. These Chapters demonstrated clear distinction in sensorial capability between older and younger consumers, with the latter being able to better discriminate samples. In summary, this research demonstrated that meat products can be formulated to provide different textures for various demographics. Products for elderly people's consumption can be created by texture manipulation through varying the formulation and cooking method. Additionally, the combined use of qualitative and quantitative methods with biometrics is recommended in order to obtain reliable responses from older consumers, especially due to their heterogeneity in wants, needs and responses.

Modern zoos are heavily reliant on the maintenance of high standards of animal welfare and the provision of positive visitor experiences to operate successfully as zoo-based conservation organisations. Several factors, both animal and environment related, influence the welfare of zoo animals and visitor experience. However, one major influencing factor has been visitor-animal interactions. Understanding the effects of visitor-animal interactions in zoos is critical because zoo animals can encounter frequent, and sometimes intense, interactions with unfamiliar humans that may compromise both animal welfare and visitor experience. Extensive research over the last several decades has shown that visitors can influence the behaviour and welfare of a range of zoo species. These effects have varied from stressful, enriching, or innocuous. However, the inconsistent findings have limited our understanding of the visitor-animal relationship. Furthermore, limited research has been conducted to identify the effects of visitors on understudied taxa such as penguins. The effects of zoo animals on visitor attitudes and their experience also remains poorly understood. Hence, the aim of this thesis was to identify the effects of regulating visitor-penguin interactions on a rarely studied species, little penguins (Eudyptula minor), and how these interactions affect visitor attitudes and their experience. A combination of animal behaviour and stress physiology as well as visitor behaviour, attitudes and experience were measured, using an experimental approach to examine the zoo visitor-penguin relationship. A total of three zoo sites were involved: Melbourne Zoo (Victoria, Australia), Taronga Zoo (New South Wales, Australia) and Wellington Zoo (Wellington, New Zealand). At each zoo, different aspects of visitor contact were regulated as a result of the differing enclosure designs. At Melbourne Zoo, the effects of regulating visitor viewing proximity and behaviour were examined. It was found that the close viewing proximity of visitors primarily affected little penguins. This was shown by an increase in the number of penguins vigilant, huddling and avoiding the visitor viewing area and a reduction in the number of penguins surface swimming when visitors were in close viewing proximity. However, no effect on little penguin faecal glucocorticoid metabolite (FGM) concentrations was found. There was no detrimental impact of regulating the interactions between visitors and penguins using a physical barrier on visitor attitudes and their experience. Also, a positive relationship between penguin behaviour and visitor attitudes and experience was found. The more visible, active, and close to the visitor viewing area penguins were, the more positive visitor attitudes and experience were at Melbourne Zoo. At Taronga Zoo, the effects of obstructing a single visitor viewing window was examined. Unobstructed visitor contact was found to reduce the presence of little penguins, reduce the number of penguins preening in the water and increase the number of penguins vigilant in the area near the main viewing window when this window was uncovered. There was also a general preference by the penguins for the corner area which was visually hidden from visitors. These results provide evidence of some avoidance of visitors by little penguins at Taronga Zoo. Also, no effect of regulating visitor contact using a visual barrier was found on visitor attitudes and their experience at Taronga Zoo. FGM concentrations were not measured at Taronga Zoo. Lastly, at Wellington Zoo, the effects of the presence and absence of visitors were examined. The presence of visitors was found to reduce the percentage of penguins observed close to the visitor viewing pier which was identified as the main viewing area used by visitors. But there was little effect on other penguin behaviours and FGM concentrations. These results suggest that there was an increase in avoidance of visitors when present by the penguins which suggests close visitor contact may be fear-provoking for these penguins. The results of this thesis demonstrate that zoo visitors may negatively affect zoo-housed little penguins where visitors may be perceived as fear-provoking, as indicated by increases in behaviours indicative of fear. But the magnitude of the penguins’ fear response to visitors varied across zoos. This may be the result of influencing factors including enclosure design, previous experience, genetics, and weather. The most pronounced fear response by little penguins towards visitors was found at Melbourne Zoo, followed by Wellington Zoo and lastly, Taronga Zoo. Penguin fear responses to visitors were most pronounced at Melbourne and Wellington Zoo likely because of the main viewing locations for visitors were generally positioned above the pool. This type of viewing position provided opportunities, if visitors chose, to closely approach and loom over the pool edge which may have been perceived as threatening by the little penguins. In comparison at Taronga Zoo, visitors were positioned below the surface of the pool. Therefore, results from this thesis indicate that the close viewing proximity of visitors, particularly from above, is a common characteristic of visitor contact that little penguins may find fear-provoking. The overall effect of visitors on the welfare of zoo-housed penguins is somewhat unclear. Despite fear responses towards visitors by little penguins were apparent when visitor contact was uncontrolled, there was no effect of the treatments imposed on FGM concentrations. One interpretation of the lack of treatment effects on FGM concentrations at Melbourne Zoo and Wellington Zoo is that the penguin behavioural responses such as avoidance may have been an effective adaptive response in ameliorating the need for a sustained physiological stress response arising from close visitor contact. Nevertheless, close visitor viewing proximity may be fear-provoking for little penguins and can be regulated by positioning visitors further away from the enclosure boundary or by reducing the perceived close proximity of visitors for little penguins. This will result in improvements in penguin welfare and have no detrimental effect on visitor attitudes and experience, which were overall positive and enhanced by penguins displaying more active behaviours and less behaviours indicative of fear. However, it is unclear whether these positive attitudes and experience of visitors affected the way visitors behaved towards little penguins. The results reported in this thesis provide what is believed to be the first comprehensive study investigating the zoo visitor-penguin relationship from both the animal and visitor aspects. This research expands our knowledge on the zoo visitor-penguin relationship, indicating the importance of examining visitor-animal relationships from the animal and visitor perspectives and identifies some practical measures that zoos can utilise to improve the management of visitor-penguin interactions.

Potential for increasing productivity, profit, and owner wealth in small farming in Paraguay. Abstract In Paraguay, the farm's profitability and the introduction of changes to farm systems are challenging to assess due to the lack of a comprehensive method based on standard farm economics budgeting for the whole farm and risk analysis. For its comparative advantage, beef production was selected to test the new method for Paraguay. Beef's annual gross value is around beef USD 1.2 Bn and increasing steadily. In the last 20 years, the national cattle herd has increased from 6,457,329 heads in 1981 to 13.858.584 in 2017 (MAG, 2017). There are 93,935 small farms, each with an average of 9.5 animals (CAN, 2009). Traditionally small farmers have not considered raising cattle as a commercial business but as a store of wealth. Presently, beef is produced as a sideline activity with low stocking rates of up to one animal per hectare feeding on native pasture, taking three to four years to reach slaughter weight. The study's focus is on establishing farm economics methods for the analysis of innovations in small farms in Paraguay. This research tested the new method on the potential gains from changing to a beef production system using improved breeds and pastures to increase the carrying capacity, shorten time to turn-off and increase beef production per farm. The case study research approach is used to analyse increasing beef productivity and production in six small farms in different regions of Paraguay. Each case study farm has been assessed using the Whole Farm Approach to establish the current economic (efficiency and wealth) and financial (liquidity) situation. The improved beef and pasture system investigated has the aim of growing out 200 kg liveweight weaners to slaughter weight of around 400 kg liveweight within a year. Data for the analyses come from technical research and on-ground information, and regional and national price data. Estimation of potential stocking rates was based on animal feed demand for growth and potential pasture growth for the regions. The innovation is judged according to returns and risk. The expected additional profit and wealth and cash flows, and their volatility, are estimated. Discounted cash flow analysis of investing in improving pastures and increasing livestock carrying capacity indicated the innovation is economically sound and financially feasible, provided an extra 2-3 head/ha are finished per year, finishing animals within seven months. The return on marginal investment considering risks was comparable with returns from similarly risky investments in the economy. The novel contribution of this thesis is the review and establishment of modern farm economics methods and analysis to risk assessment for investment in farms in Paraguay.

Pulses, also known as grain legumes, are members of the family Leguminosae and are grown for their edible seeds, containing high amounts of proteins and fibre. The work performed in this thesis is focused on lentil (Lens sp.), which is a self-pollinating, diploid, cool-season grain legume. Lentil production is constrained by multiple biotic and abiotic stresses that reduce growth and grain yield. The development of lentil varieties/cultivars with improved characteristics, including better yield, adaptation, and resistance to biotic and abiotic stresses, is a priority for international breeding programs. Therefore, the thesis investigated advanced genomic and phenomic approaches to characterize lentil germplasm for breeding purposes. Cultivated lentil (Lens culinaris Medik.) has a relatively narrow genetic base. Therefore, characterization of genetic diversity and genomic differentiation of wild gene pools is essential to identify any favorable alleles/genes that can be introduced into elite germplasm. A total of 467 wild and cultivated lentil accessions originating from multiple geographical locations were assessed for understanding genetic and allelic variations using transcriptome sequencing. An enriched single nucleotide polymorphic (SNP) resource (c. 422,101) has been delivered to lentil breeders for mining diverse genotypes for hybridization in future research and breeding. Understanding the relationship between lentil accessions and their geographical origins is also vital for identifying favorable alleles/traits that can be introgressed into the lentil germplasm. However, a weak correlation was observed between the lentil accessions, except for some accessions belong to L. culinaris and L. ervoides. Therefore, the study proposed that identifying lentil accessions with wide genetic distance variations within the same gene pool is more promising for selecting lentil accessions for breeding purposes, which also avoids crossing barriers between different gene pools. Lentil accessions that belong to L. culinaris, L. ervoides and L. nigricans were shown broad genetic distance boundaries. Therefore, these accessions with specific agronomic traits can be used to widen the lentil germplasm for breeding purposes. The genomic differentiation in each lentil species/subspecies was also analyzed using the allele-frequency-based analysis. The major genomic differentiation was observed on Chromosome 1 (Chr1; c. 1.0 Mbp), and results implied that L. nigricans was distantly related to other lentil species/subspecies. A total of five candidate genes were identified on c. 1.0 Mbp physical distance; however, the functionality of these genes in relation to wild and cultivated lentil species/subspecies still needs to be understood. One of the major abiotic stresses affecting gross profit and yield stability in Australian lentil cultivation is soil salinity. Identification of salt-tolerant varieties is the most viable and long-term option to maintain lentil productivity. However, this requires reliable and efficient screening methods. Salt tolerance assessment in lentil is currently based on morpho-physiological characterization and visual score ratings, which are often time-consuming, labor-intense and error-prone. Therefore, a novel high-throughput phenotyping (HTP) approach based on an image-based screen was developed using the LemnaTec 3D scanalyzer system to circumvent the limitations faced by current methods and accelerate the identification of the salt-tolerant varieties. The optimal salt concentration (100 mmol) and growth stages that distinguish salt tolerance levels were identified. Among the multiple phenotypic traits measured, area and color parameters were identified as the most informative traits for salt tolerance in lentil. The significant correlation observed between traditional and image-based screens (r = 0.55; p < 0.0001) demonstrated the accuracy of the developed protocol for salt tolerance in lentil, thereby can replace the conventional phenotyping approach. In addition to the phenotypic approaches, the understanding of the genetic basis of salt tolerance in lentil is important to develop salt-tolerant varieties. Recently, genome-wide association studies (GWAS) have been identified as a powerful tool to dissect the genetic basis of many phenotypic traits in diverse germplasm. Advances in resequencing approaches such as genotyping-by-sequencing (GBS) methods have also enabled the generation of a panel of SNP markers for large genome species like lentil. Two GBS approaches, targeted-capture (tGBS) and transcriptome-based sequencing (GBS-t), were tested to generate high-confidence SNP markers for association study. Among them, tGBS delivered the highest number of SNP markers with uniform distribution across the genome. Genomic regions for salt tolerance in lentil were identified on Chromosome 2 as well as on Chromosome 4. A high-affinity potassium transporter (HKT) gene was identified as the most possible candidate gene for salt tolerance in lentil. Mineral composition analysis performed on salt-treated and control lentil accessions has also been identified; Na+ ions absorbed by tolerant lentil accessions actively re-translocated them into roots or hold within the roots, supporting the candidate gene identified through GWAS. Pedigree analysis performed on salt-tolerant lentil genotypes identified two lentil accessions, ILL7685 and ILL1719, that could have been potential sources of allele contribution to salt tolerance in the lentil population. Overall, the study enriched the genomic and phenomic resources associated with lentil, thereby assisting future lentil research and breeding.

The progeny of primiparous sows (gilts) are recognized as a significant burden on the overall production efficiency of Australian pig herds. Gilt progeny (GP) grow slower and have higher rates of morbidity and mortality compared to the progeny of multiparous sows (sow progeny; SP). Previous research suggests that there is an underlying biological basis for differences between GP and SP. Therefore, the aims of this thesis were to clarify the timeframes of when these differences are in effect by investigating gastrointestinal (GI) development in the pre-weaning period and also to develop late gestational and lactational nutritional interventions to improve GP performance. In doing so, this thesis consists of a series of experiments in which three general hypotheses were tested: (1) GP display poorer production performance compared to SP; (2) GP consistently displayed reduced gastrointestinal (GI) tract development compared to SP and (3) GP production performance could be improved by targeting nutritional interventions towards late gestation and/or lactation. Experiment 1 (Chapter 3) involved the quantification of GI functional development at 4 time points; birth (0h), 24 hours after birth (24h), 1-day pre-weaning (PrW) and 1-day post-weaning (PoW) of GI tissues (stomach, jejunum, ileum and colon). Transepithelial resistance (TER) was measured on all organs of the GI tract while the permeability of macromolecules fluorescein isothiocynate (FD4) and tetramethyl rhodamine isothiocynate (T150) labelled dextrans was measured in the mucosa of the jejunum and ileum. Additional GI function measurements were taken such as tight junction proteins and cytokine abundance in the jejunal and ileal tissues. The main findings were that GP showed early signs of lower GI tract function compared to SP at birth, particularly in terms of lower TER in the stomach tissues of the stomach, jejunum and ileum. However, differences between GP and SP were largely observed from PrW to PoW with GP exhibited poor GI barrier function at weaning reflected the reduced epithelial barrier integrity in response to weaning, mainly in the distal portions of the GI tract (ileum and colon). Differences in GI barrier function between the birth cohort and the weaning cohort suggests that the variation between GP and SP is established in the pre-weaning period. This indicates that the pre-weaning period is critical for optimal GI tract function development for GP and their reduced growth performance compared to SP is likely due to a lack of colostrum/milk consumption. As the development in GI barrier function in the piglet is critical for survival and productive efficiency, we investigated the utility of quantifying intestinal fatty acid binding protein (I-FABP) as a biomarker of GI integrity (chapter 4). Fatty acid binding proteins (FABP’s) are ubiquitous intracellular transport proteins integral to fatty acid metabolism. However, organs of the GI tract have a specific intestinal isoform (I-FABP). The use of FABP’s as biomarkers for organ damage have been proposed, but little is known about the utility if I-FABP as a biomarker in the pig. The outcomes of this experiment showed that plasma I-FABP concentrations were substantially lower in piglets than later in life and this led to significant challenges in detection using existing ELISA based methodologies. While it is interesting that I-FABP is expressed in such low concentrations early in life, to properly understand the utility of I-FABP as an intestinal biomarker will require more sensitive technologies such as mass spectroscopy, which will limit its utility as a biomarker. In order to support growth of GP in the pre-weaning period, various nutritional strategies were investigated and implemented during either late gestation and/or lactation. In experiment 3 (chapter 4), the ketogenic substance 1,3-Butanediol (BD) was fed to gilts and sows from day 90 of gestation to farrowing in order to provide and alternative energy source to GP at farrowing. The main result was that birth litter weights were increased in GP from gilts fed the BD diet and this extended to day 21. Furthermore, the proportion of GP born less than 1.1kg were less from those gilts fed the BD diet, increasing their chances of survival to weaning. Previous research suggests that gilts suffer from high levels of oxidative stress during gestation and possibly lactation compared to multiparous sows resulting in reduced GP growth. Experiment 4 (chapter 5) involved supplementing late gestation and lactation diets with a sugar-cane derived polyphenol mixture (Polygain) to improve GP growth during the pre-weaning period. Furthermore, due to the anti-inflammatory properties of Polygain, weaner diets were supplemented with this mixture to reduce post-weaning inflammation. Polygain was unable to increase GP growth in the pre-weaning period nor did it reduce inflammation measured by the circulation pro-inflammatory cytokine interleukin-1 beta (IL-1b). Interestingly, results from this experiment do suggest that GP exhibit reduce immune function compared to SP reflected in their lower plasma IL-1b concentrations overall compared to SP. Lastly, experiment 5 (chapter 7) investigated the effects of feeding gilts and sows a diet supplemented with lucerne chaff or a diet formulated to have the same estimated fibre, amino acid and energy content as lucerne (SIMLUC) to improve GP performance. These diets were able to increase GP average weight and gilt litter weight at day 21 while a similar effect was not observed in SP. The improvement of D21 litter weights in gilts in response to the LUC and SIMLUC diets fed in late gestation is an important finding which demonstrates the importance of targeted feeding strategies towards gilts to maximize GP performance. In conclusion, results from this thesis indicate that GP are consistently born and weaned lighter and growth slower than SP. They display reduced GI tract development, with GP entering weaning with reduced GI integrity. Furthermore, GP display some signs of reduced immunocompetence compared with SP which may impede their ability to cope with the stressors of weaning. Therefore, nutritional interventions are most likely to work when they support higher milk yields in gilts and support growth of GP in the neonatal and pre-weaning period.

Myanmar is an agricultural country with substantial development potential, having abundance of land, water and labor resources. Despite this, agricultural productivity in Myanmar is low and smallholder farm profits and incomes are amongst the lowest in Asia. Many reports have documented that the underperformance of crops and low yield is mainly due to the low use of fertilizers by smallholders in Myanmar. Fertilizer application is an important strategy in increasing crop yield. However, determining the “Right Rate” at the “Right Time” is challenging to achieve economic, social and environmental goals. Many decision support tools (DSTs) have been developed to support farmers in making better informed fertilizer management decisions. Such tools allow farmers to use fertilizers more efficiently and increase production, reduce risks, save resources and gain more profit with a lower detrimental impact on the environment. DSTs are developed to be a solution and support farmers in Myanmar with their fertilizer decisions. However, there is limited evidence of the acceptability and successful ongoing uptake of DSTs by farmers throughout the world. A review of the literature suggested that agricultural DSTs, in general, have suffered from low levels of use, apparently due to limited acceptability by end-users. Reasons for this include lack of user participation in the initial design and development of the tools, complexity of earlier DST versions, insufficient training and technical support and limited ongoing funding for DST maintenance. Participatory approaches to include farmers as users have been investigated to improve the design and development of DSTs. The complexity of the tools has been simplified into decision support smartphone apps but delivering useful information that represents and relates to farmer decision making remains important. A strong emphasis on the farmer decision-making processes with respect to economics and risks are still required for wider adoption of DSTs. New technologies that integrate socio-economic factors into farmer decision making are likely to be valuable. Thus, this study aims to develop a decision support framework to assist farmers with their fertilizer decisions in Myanmar. Data were obtained from a survey of 600 smallholders in Tatkon, Zeyarthiri and Taungoo Townships in the central Dry Zone of Myanmar. The survey was conducted using CommCare, a mobile application for data collection. Field experience in the design and implementation of using CommCare to collect data for a complex farming system in Myanmar was discussed along with the recommendations which may allow other researchers to conduct a smooth and successful survey using similar mobile data collection process. The survey results provided valuable information on fertilizer use to test whether the reportedly low levels of fertilizer use by Myanmar smallholders are generally true for central Myanmar and compared the timing of fertilizer application practices of smallholders against recommended “Good Management Practices”. There have been concerns over the national data quality with very limited evidence in Myanmar. This is the first substantive study in Myanmar to compare smallholder fertilizer use with national recommended rates and report the fertilizer application times on cereal crops (rice and maize). Smallholders used both urea and compound fertilizers for cereal crops. For rice, the average amount of fertilizer applied was much higher than the reported national-level data of average fertilizer use. With respect to the timing of applications, smallholders were mostly aligned with recommended practices for Nitrogen (N), except that nearly half of the surveyed smallholders were not applying N at that the estimated panicle initiation stage, which is crucial to increase yield. With respect to applying Phosphorus (P), the majority (82%) of smallholders were applying as compound fertilizer throughout the growth stages when earlier application of P is recommended. Smallholder farmers may also be able to reduce the cost of labor by reducing the number of P applications and avoiding late applications. These results have enabled a better understanding of smallholder farmer utilization and management of fertilizers, their experiences and their livelihoods which remain crucial in designing and implementing sustainable interventions. The survey also provided information on farmers’ use of smartphones and agricultural mobile applications (apps) and analyzed factors affecting their use. This was used to assess the potential for farm-based decision support. In Myanmar very little literature relates to perceptions and use of Information and Communication Technologies (ICTs) and no study has considered factors related to the use and adoption of mobile apps in the Myanmar agricultural sector. Given the low uptake of decision support tools/apps by farmers throughout the world, this study provides valuable context for discussing farmer perceptions and what factors need to be considered in developing and scaling up mobile applications. A Probit analysis indicated that when introducing mobile-based tools, focus should be given to younger, more educated farmers growing more specialized crops. Smallholder farmers in Myanmar were found to be optimistic and positive towards agricultural apps, but there were many barriers including limited access to smartphones, lack of digital knowledge, and high cost of internet for effective utilization. However, most surveyed farmers were familiar with information received through Facebook groups. Incorporating useful information and functions from an agricultural mobile app to a Facebook Page could have a more useful and sustainable impact, at the same time eliminating the issues such as lack of awareness, lack of use and long-term servicing and sustainability. This new knowledge is important as many agricultural mobile apps have been developed and are continuing to be developed for decision support to farmers when these apps have not been widely utilized by farmers. Another reason for the limited use is that the tools do not account for complex and heterogeneous smallholder farming systems. Farm typology analysis has facilitated transfer and adoption of appropriate technologies to target groups of farmers with similar circumstances. It is important to consider the heterogeneity of farmer groups if DSTs are to be designed to reflect the needs of targeted end-users. Typology analysis was further applied to the survey dataset to distinguish farmer groups with similar socio-economic characteristics. Six distinct groups of farms were identified. A qualitative participatory research approach with smallholders from each farm type was subsequently conducted to identify potential adopters of a DST for fertilizer management. Only one group of farmers indicated that DSTs could be useful in gaining more information and knowledge. These farmers were more likely to adopt a fertilizer DST than other farmer groups. Farmers who were interested in the technology indicated that Discussion Groups are a better learning environment and preferred a learning-based tool rather than a prescription-based tool. Discussion Support is a collaborative approach between farmers, whereas Decision Support is a more instructive approach providing recommendations which can be misinterpreted. Discussion Support can be used by individual farmers, groups of farmers, or a group of farmers with extension officers enabling discussion of alternative management decisions, asking specific questions and seeking new ideas about adjusting farm decisions to meet smallholder management objectives. Farmers expressed preferences for video clips and infographics integrated into digital platforms, including social media and existing agri apps. Integrating a participatory research framework with typology identification is potentially valuable in assessing new technologies and may have a beneficial role in direct interactions with smallholders. This framework creates an opportunity to implement farmers’ suggestions and opinions, which can be a valuable resource for further information system (IS) and decision support system (DSS) researchers. The proposed decision support framework is to abandon the traditional DST approach, which is to be considered as a “black box”, complex requiring substantial manual data entry requirements which are not attractive to farmers. Instead, this framework suggests ongoing participatory approaches with farmers to gain their trust, integrate their inputs and implement their suggestions, identify specific groups of farmers as early adopters, utilize social media platforms to deliver discussion support outputs which are readily usable in the form of video clips and infographics and develop a discussion support format where farmers can discuss and learn from each other. Considerable care should be taken to ensure that farmers’ socio-economic factors and decision-making contexts are also considered. In the case of smallholders in central Myanmar, there was very little evidence from the field survey in support of the perception that Myanmar farmers are using low levels of fertilizers. The results indicated the farmers appeared to be substantially following good fertilizer management practices and decision support should be focused on other management practices instead of fertilizer.

Animal welfare science has functionally only existed for a little over 50 years. The last three to four decades in particular have seen a relative boom in the rate of expansion, understanding and attention that this discipline has received. Whilst the foundations of animal welfare science mainly focused on identifying and removing negative welfare states in captive animals, modern scientific inquiry is now starting to understand and approach positive welfare states as a crucial part of any sentient being’s experience of life as well. Positive welfare states may include many elements, such as an animal being fit and healthy, experiencing positive moods and affects, and being able to express natural behaviours (or instead, as this thesis will argue, behaviours that are highly motivated and/or highly rewarding). Encouraging, facilitating and maintaining positive animal welfare states in captive zoo animals are a high priority for modern, ethical zoos. However, there is currently a substantial gap in the published literature exploring positive welfare states associated with human (especially visitor) contact in zoo settings. There is limited research that suggests that human-animal interactions in zoos may potentially be rewarding for some animals that are motivated to participate in, and even ‘solicit’, these interactions, from both familiar husbandry providers (zookeepers) and/or unfamiliar zoo visitors. Ethically, zoos operate under a few key theories pertaining to animal welfare, animal rights, and environmental ethics. A few such theories are: Compassionate Conservation, Conservation Welfare, and Duty of Care. These theories take inspiration from multiple philosophical discourses, and many of them co-exist within the zoological and aquaria communities, institutions and (self-regulated) associations. Many individual institutions may favour particular ethical theories over others, and not all zoos are ethically run, nor is their captivity of certain animals justified or adequate. However, many influential voices within zoological associations are creating a robust model for running ‘modern, ethical zoos’. An amalgamation of many theoretical ethical approaches is required to fully articulate why zoos should, and do, continue to exist. The phylogenetic Class Reptilia (reptiles) is now more correctly termed non-avian reptiles, as recent taxonomic amendments have included all extant and extinct animals back to the clade Diapsida, which includes all dinosaurs, and hence all modern birds - Aves. This now makes reptiles a monophyletic group (i.e. with a single common ancestor). Thus, when discussing modern reptiles, such as tortoises, it is proper to distinguish between avian and non-avian reptiles. Currently, none of the non-avian reptile families have been adequately studied in terms of animal behaviour, cognition and animal welfare sciences. These families include: Sub-order Crocodilia – crocodiles, alligators, gharials, & caimans; Order Lepidosauria, which includes both Squamata – lizards and snakes, and Order Rhynchocephalia – Tuatara; and Order Testudines – turtles, tortoises and terrapins. Declarations of sentience (i.e. feelings) made by governments and scientific associations (such as the World Organisation for Animal Health (OIE)) often include reptiles under the assumption of “all vertebrate animals”, but specific declarations of sentience in reptiles are often missing, or overlooked, in scientific writing and welfare-related policies and discussions. Indeed, the historical assumption that reptiles are merely sedentary automata without complex cognitive and sentient capacities persist. Available cognition and sentience research, however, indicates that there is a very solid basis for assuming and declaring that non-avian reptiles do indeed display all relevant capacities to be classified as conscious, aware, and sentient beings. This means that they most likely are consciously aware of their own welfare, and hence their lives and well-being matter to them. This thesis was designed to weave a coherent story connecting animal ethics, states of consciousness, awareness and sentience in non-avian reptiles (that are often overlooked), and experimental research that addresses whether some zoo-housed reptiles perceive human interaction as rewarding, and whether they are indeed motivated to seek these interactions. Three experiments were conducted: An Aldabran Giant Tortoise (Aldabrachelys gigantea) preference test (n=2); a Leopard Tortoise (Stigmochelys pardalis) preference test (n=5); and a zoo visitor survey of behavioural and ethical beliefs about zoo-housed non-avian reptiles, and the acceptability of human-animal interactions and owning wildlife as pets (n=231). Both tortoise preference tests found individual differences between subjects, and I concluded that some individual tortoises do indeed prefer human interaction (shell scratching and neck rubs) over other stimuli in the experimental circumstances. Whether the rewarding component for the tortoises was the interaction with the human or simply a pleasurable outcome was not determined. While these results could not be generalised to all populations of tortoises or reptiles in general, the results showed significant individual preferences amongst the sample populations, indicating that being aware of, and sensitive to, individual animals’ ‘wants’, rather than making decisions at a species level, is warranted for zoo-housed non-avian reptiles. The results of PCAs of visitor survey responses (n=231) found five common ethical beliefs (components) in the sample of zoo visitors, labelled: 1) Human interaction and entertainment priority component; 2) Complicated zoo ethics and animal welfare component; 3) “Wilding”, natural living and anti-captivity sentiments component; 4) Ethical duty of care component; and 5) Animal agency and respect component. There were some significant differences between agreement with Component 2 and respondents’ education level. Furthermore, Wilcoxon’s Signed Rank tests on 3 paired questions (that were answered before and after randomly allocated positive or negative human-animal interaction statements) significantly influenced the re-rated scores to the questions in either the positive or negative direction corresponding to the information provided, showing a significant influence of education (informative statements) on zoo visitor attitudes. Finally, drawing from relevant information of the known capacities of non-avian reptiles, and from the results of the two preference tests conducted herein that indicated some preferences for human interactions by tortoises studied, I investigated an ethical model of human-animal interactions in zoos that may benefit the well-being and positive welfare of both animal and human participants, and proposed some recommendations for improvement of such interactions. These recommendations may be relevant to zoological institutions and their governing associations, and the results of the ethical and experimental chapters within this thesis may help inform evidence-based improvements for non-avian reptile welfare within these institutions.

The processing tomato industry in Victoria, Australia, has experienced a yield decline over the last decade, resulting in losses estimated at 10% per annum. The decline was attributed to the necrosis of lateral and feeder rootlets and the collar region resulting in plant stunting and a reduction in fruit production. Therefore, the hypothesis underlying this study was that the decline is caused by the cumulative effects of damage by a complex of soil-borne root and collar rot pathogens. Surveys of processing tomato crops were undertaken over three consecutive growing seasons between 2016 and 2019 to investigate the pathogens, symptoms and diseases associated with yield decline. Soil-borne fungal and oomycete pathogens were the focus but bacterial pathogens, viruses, nematodes and phytoplasmas were also noted. Systematic isolation from diseased roots and the collar region of plants putatively infected by fungal and oomycete pathogens was undertaken. Identification of isolates was based on cultural morphology, ITS sequencing and in some cases commercial qPCR testing. Fusarium oxysporum and Pythium spp. were the most abundant putative pathogens associated with plants exhibiting poor growth. Other putative pathogenic fungi and oomycetes which were less commonly encountered included Alternaria spp., Colletotrichum coccodes, Fusarium solani, Phytophthora nictotianae, Phytophthora cajani, Plectosphaerella spp., Rhizoctonia solani, Sclerotinia minor and S. sclerotiorum. A novel Fusarium collar and root rot disease of processing tomatoes was discovered during the surveys. The disease was characterised by chocolate-brown streaking in the internal collar and tap root tissue, as well as lateral root rot of stunted tomato plants. Morphological characterisation and multi-loci phylogenetics (ITS, ef1a and Pgx4), were used to identify the causal pathogen as Fusarium oxysporum. The disease was initially thought to resemble Fusarium Crown and Root Rot (FCRR) caused by Fusarium oxysporum f. sp. radicis-lycopersici (Forl), a disease which has not been reported in Australia. However, subsequent pathogenicity and physiological assessments of isolates suggested the disease was caused by a novel Fusarium pathogen. Consequently, this disease was named chocolate streak disease (CSD) to differentiate it from FCRR. Pythium was the second most abundant organism isolated during the surveys. As Pythium is a large genus consisting of species beneficial, neutral and detrimental to plant growth, further investigation was required to understand the impact of Pythium spp. on processing tomato growth and yield. Eleven species of Pythium were identified based on cultural characteristics and phylogenetic analysis using ITS, Cox-1 and Cox-2 gene sequences. None of these Pythium species had been reported previously from processing or table tomatoes in Australia. In addition, this is the first report of P. carolinianum, P. heterothallicum, P. recalcitrans and a new Pythium sp. from field-grown tomato crops globally. Pythium dissotocum was the most abundant and widespread species. Pythium ultimum, P. aphanidermatum and P. irregulare were the most aggressive towards both seedlings and mature plants, causing pre- and post-germination damping-off, severe root rot and stunting. Collectively, the evidence provided by this study supports the hypothesis that a complex of root and collar rot pathogens, particularly F. oxypsporum and Pythium spp., are contributing to the 10% yield loss in Victorian processing tomatoes.

Polyphenols are widely studied due to their antioxidant activities with the capacity of preventing the generation of free radicals and reactive oxygen species and are therefore associated with reduced risks of chronic diseases caused by excessive oxidative stress. Matrix factors, including plant structure and interactions between polyphenols and macromolecules from diets play crucial roles in the bioaccessibility, digestion and absorption of polyphenols. While it is known that structural changes of polyphenols occur during digestion, the influence of the dietary matrix is not well understood. This project presents a stepwise approach, using black carrot as a model plant and both in vivo and in vitro digestive models, to understand digestion of phenol-rich plant foods and its effect on gut microbiome composition as well as long-term plasma antioxidant capacity under influence of plant cell structure, polyphenol-lipid interactions and polyphenol structural biotransformation. The results provided a comprehensive view of the digestion of phenol-rich plant in context of a whole diet and implied the significance of matrix factors in polyphenols bioavailability and colonic microbiome response to phenol-rich plant digestion. There has been a lack of standardised measurement of polyphenol activity and structure that can be generally applied to various fruits and vegetables. Here, improved methods of detection of potential antioxidant activity of food substances were modified and standaridsed, using high-throughput 96-well plate assays, and individual polyphenols were separated and quantified using HPLC-PDA and LC-ESI-QTOF/MS. Sixteen plant foods were tested. Factor analysis (FA) and Pearson’s correlation tests showed high correlations among 2,2’-azinobis-(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS), 1,1-Diphenyl-2-picryl-hydrazyl (DPPH), ferric reducing antioxidant potential (FRAP) and phenolic acids, implying the comparable capabilities of scavenging the DPPH/ABTS free radicals and reducing ferric ions from the antioxidant compounds in the samples. Developing these assays in high-throughput mode allows various fruits and vegetables of different origins to be analysed. Black carrots, together with other anthocyanin-rich fruits and vegetables, showed high antioxidant activities among the tested plant foods. Bioaccessibility of plant phytochemicals to digestive processes is dependent on the integrity of the plant cell but the fate of phenolic compounds is not well described. An in vitro simulated digestive system was used to detect the bioaccessible antioxidant activities and individual polyphenols of black carrots at gastric, small intestinal and colonic digestive stages via the standardised antioxidant assays and LC-ESI-QTOF/MS. Possible degradation pathways of phenolic compounds were proposed. The majority of polyphenols in black carrots were decomposed through anthocyanin and phenolic acid deglycosylation, and phenyl acids and/or benzoic acids pathways during small intestinal and colonic digestions. Using an in vitro system provided standardised interpretation of the biochemical pathways in digesting plant phytochemicals during digestion. The bioavailability of polyphenols is further affected by plant cell structure and dietary compounds. The effect of plant cell structure (raw carrot dices and cooked puree) and dietary lipids (sunflower oil, beef tallow and coconut oil) on polyphenol bioaccessibility and these effects were next studied by investigating the long-term consumption of phenol-rich plants on plasma total antioxidant capacity (TAC) and atherogenic index (AI) with a whole animal (in vivo) pig model. Processing black carrot to smaller particles and inclusion of dietary lipids, especially coconut oil (rich in medium-chain fatty acids (MCFAs)), significantly increased polyphenols bioaccessibility due to attenuated cell wall trapping and increased structural stability of polyphenols enhanced by phenol-micellised lipids interactions. However, the overall bioaccessibility of polyphenols, especially anthocyanins, were low after small intestinal digestion with only around 5% of cooked puree and around 2% of raw diced carrots bioaccessible due to the low stability of anthocyanins at pH 7. The in vivo pig study verified the low bioavailability of polyphenols in black carrots with diets supplemented with black carrots not showing statistically significant increases in plasma TAC in a porcine model after the 4-week length of the study, although it showed the trend with significantly higher plasma TAC in low-fat diet-carrot puree group compared to that of the groups with high-fat diet control and high-fat diet-carrot dices. However, excessive long-term administration of the high-fat diet induced significantly higher plasma AI. These results indicated that although dietary lipids enhance polyphenols bioaccessibility during digestion, consumption of plant material could not counterbalance the negative impacts from long-term excessive intake of lipids. The effect of dietary supplementation of the plant material on the composition of the gut microbiome and production of short-chain fatty acids (SCFAs) were studied both in vitro and in vivo through 16S rRNA sequencing and GC-FID. Presence of dietary black carrots modulated gut microbiome with increased relative abundances in health beneficial bacteria (including the Prevotella, Prevotellaceae NK3B31 group, and Lactobacillus) were observed after 20 h simulated colonic fermentation in the in vitro model. When lipids were included in the dietary matrix, the effect of black carrot was attenuated with more similar microbiome compositions shown between carrot-lipid matrix and the corresponding lipid control groups. In the pig study, however, limited microbial compositional changes were found in pigs with a high- or low-fat diets when combined with pureed or diced carrot. Influence from other polysaccharides in background diets, genetic backgrounds and colonic milieu of the hosts could have contributed to the gut microbiome resilience to the added polyphenols and dietary fibre. Coconut oil (rich in MCFAs) showed much more significant impacts on gut microbiome modulation in vitro compared to sunflower oil and beef tallow (rich in long-chain polyunsaturated fatty acids (PUFAs) and saturated fatty acids (SFAs) respectively) with an increase relative abundance in colonic pathogenic Escherichia. Higher levels of SCFAs were produced by digesta supplemented with raw diced carrots during in both in vitro and in vivo studies. Limited effects on total SCFAs produced were shown when digesta was supplemented with both sunflower oil and beef tallow, but addition of sunflower oil promoted beneficial butyrate production. Coconut oil showed antibacterial activity with significantly reduced SCFAs production. These results contribute to our knowledge of digestion of phenol-rich plants under the influence from complex dietary mixes and the effect on the vast fermentative diversity of the gut microbiome, which could possibly more accurately simulated and reflected the human colonic status responses to the interventions of daily consumed fruits and vegetables. Overall, this project indicated that the impact of polyphenols in the digestive tract should be considered in the context of plant cellular structure and interactions with dietary macromolecules. Destructed cell walls enhanced the absorption of polyphenols but also reduced the fermentation possible from dietary fibre in the large intestine. Bioavailability of polyphenols from anthocyanin-rich plant foods is low due to structural instability of anthocyanins during digestion. Dietary lipids enhance polyphenols bioaccessibility during digestion, but addition of plant material could not counterbalance the negative impacts from long-term excessive intake of lipids when considering a whole animal model. Coupling phenol-rich plant foods with diets containing low amounts (around 5%) of fat could be a good approach to increase the bioavailability of polyphenols.

Abstract Arsenic (As) is a toxic metalloid, which is carcinogenic i.e. cancerous to humans. Besides the drinking water, accumulation of As in food grains through plant uptake, when cultivated in As contaminated soils, is a potential route of human dietary As exposure. This has inspired research into alleviating grain As accumulation, despite there being already existing strategies with major disadvantages such as low efficiency, high costs, and usage being restricted to smaller-scale operations. Therefore, pre-sowing microwave (MW) soil heating and sawdust biochar were used to investigate if they can reduce As concentration in wheat and rice. Microwave is a form of electromagnetic radiation, which can produce heat in the soil by inducing the rotation of the dipoles of polar molecules (e.g. water). Microwave heating depends on the dielectric properties of the soil. Therefore, a study was conducted to determine the dielectric properties of different types of soils with different moisture content. The results showed that the soil moisture was the major contributor to the dielectric behavior of soil since dielectric properties increase as soil moisture increases. Soil types also had an influence as the dielectric properties of sandy soil were much lower than the other soils such as clay and loam soil. To investigate the effect of MW and biochar on wheat and rice grain As concentration, both the wheat and rice soils were spiked with five As concentrations (As-0, As-20, As-40, As-60, and As-80 mg kg-1 soil). In addition to MW, biochar was used to reduce rice grain As accumulation since biochar has been gaining attention for its heavy metal immobilization capacity. After As application, three levels of biochar (BC-0, BC-10, and BC-20 t ha-1 soil) were added only in the rice-growing soil. Then, soils were treated for 0, 3, and 6 minutes (MW-0, MW-3, and MW-6) in an MW chamber to achieve the soil temperature of around room temperature, 60 oC, and 90 oC respectively. The crops were grown in a completely randomized design with four replications in a glasshouse during 2017 (wheat) and 2018 (rice). The results demonstrated that, in both the wheat and rice, MW soil treatments, especially the MW-6, alleviated As phytotoxicity and facilitated less grain total As concentration compared with the MW-0 treatment across all the soil As concentrations. Also, MW treatment significantly reduced the concentration of arsenite [As(III)], the most toxic form of As. Decreased grain As concentration in rice was recorded at BC-10 in lower levels of soil As concentrations (As-20 and As-40) while, a negative impact was observed at BC-20 across all the soil As concentration, compared with BC-0 treatment. Furthermore, rice grain As(III) concentration increased significantly in BC-20 treatment. Thus, MW-6 treatment could be used for the alleviation of grain As concentration in wheat and rice grain, whereas more study is needed for the best biochar application rate. However, understanding the residual effect of MW and biochar is crucial for the sustainability of the treatment. Therefore, the same varieties of wheat and rice were grown in the following year, using the same pots, without the addition of further MW or biochar treatment. The results revealed that, 360 days after MW soil treatments there was still the potential to alleviate grain As concentration in both wheat and rice. A similar result was observed for biochar treatment in the residual year with a positive effect at BC-10 and a negative effect at BC-20 treatment. Furthermore, it is unclear whether MW soil treatment is just a heating effect or if there is some other effect of the electromagnetic wave involved. Therefore, a glasshouse pot study was designed to investigate the effect of MW and conventional electric oven (EO) soil heating on As phytotoxicity alleviation in rice. The soil was spiked with three levels of soil As concentration (As-0, As-40, and As-80 mg kg-1) prior to applying MW and EO heat treatments, to achieve the soil temperature of around 80 - 90 oC. The results showed that, there was no statistically significant difference between MW and EO treatments regarding As phytotoxicity alleviation. However, the positive effect was more in MW treatment than the EO treatment. Significantly less total energy required in the MW to treat the soil than the EO. Besides the As phytotoxicity alleviation, the effect of MW soil heating on soil microorganisms, particularly bacteria, was a topmost concern and investigation was needed to ascertain that MW soil heating does not affect it drastically. Therefore, an experiment was designed to investigate the effect of MW heating (80 - 90 oC) on the soil bacterial community in As contaminated (As-0, As-40, and As-80 mg kg-1 soil) soils. The 16S rRNA bacterial gene copy numbers decreased significantly after MW soil heating but recovered back to its previous number 42 days post treatment. The bacterial diversity also decreased significantly in MW treated soils but did not recover even after 56 days from MW heating. However, there was no noticeable effects of soil As concentration on bacterial community were observed. Furthermore, relative abundance of some beneficial bacteria such as Bacillus and Symbiobacterium were significantly higher in the MW treated soils. Thus, MW soil heating at 80 - 90 oC can potentially be applied for As phytotoxicity alleviation without significantly destroying the ecologically important taxa. Overall, pre-sowing MW soil heating could be applied as a novel technique to alleviate As phytotoxicity in wheat and rice with lower As accumulation in the grain. Thus, application of the MW technology in the As contaminated area like Bangladesh could add another feather in the crown of the As remediation techniques and help to reduce the human health risk through As contaminated food grain. However, further research needed before adopting the MW soil heating technique where different aspects should be explored such as response of MW technology in the field condition, scaling up the MW equipment for field application, cost of MW application in the field at farmers level, long-term effect of MW treatment on soil nutrient dynamics, soil organic matter and soil biota and sustainability of the MW technology in field condition. Also, sawdust biochar could be used in combination with MW soil heating for As phytotoxicity alleviation; however, more study needed to set the appropriate rate of biochar application.

Among a wide range of factors that can affect animals’ wellbeing, stress levels and health status have been identified as relevant factors in production animals. The increasing awareness about animal welfare and the impact that it has on farm productivity has been promoting scientific research and the development of novel and less invasive methods to monitor animals and obtain measurements that can be used as indicators to assess animal wellbeing, including stress and health status. Research presented in this thesis deals with red-green-blue (RGB) and thermal infrared (TIR) imagery, and computer vision techniques, as non-invasive tools to measure physiological changes in cattle and pigs to assist in the assessment of their stress and health. One study was conducted to evaluate the performance of the proposed methods, which were used to analyse RGB and TIR imagery to measure eye-temperature, heart rate (HR) and respiration rate (RR) in cattle. The study was performed in a robotic dairy farm, where TIR and RGB cameras recorded ten dairy cows during six handling procedures, across two consecutive days. Simultaneously, core body temperature, HR and RR were measured using standard methods for comparison with the data obtained from the recorded images using the developed algorithms. A feature tracking algorithm was developed to facilitate the processing of RGB videos, which showed an accuracy between 92% and 95% depending on the area analysed. From the physiological parameters analysed, the highest correlations were observed between eye-temperature and intravaginal temperature (r = 0.8; P<0.01), and between remote RR and the RR obtained from visual observations (r = 0.87; P<0.01). A further two studies were carried out to implement the proposed computer-based methods to remotely measure eye-temperature, heart rate and respiration rate of cattle, and to investigate whether these measures could be used to evaluate the physiological response of cattle to stressful situations and whether they could be used as predictors of beef quality. For these two studies, 215 beef cattle were recorded with RGB and TIR cameras on the farm and at the abattoir to obtain eye-temperature, HR, and RR measurements. Cuts of the respective beef were evaluated by consumers, and the ultimate pH (pHu) and meat colour were obtained from the respective carcasses. It was observed that the physiological variables of cattle were higher at the abattoir compared to the farm. Moreover, eye-temperature obtained on farm and at the abattoir were highly correlated. However, the results of these studies indicated that these measurements had low contribution when predicting beef quality. Finally, two studies were performed to investigate the use of the proposed computer vision methods with RGB and TIR imagery to monitor pigs and detect changes in eye-temperature, ear-base temperature, HR, and RR. The objective was to identify whether these physiological changes could assist in the early detection of pigs affected by respiratory disease. The first of these studies was performed under experimental conditions, where pigs were challenged with Actinobacillus pleuropneumoniae (APP). Images (TIR and RGB) were recorded after this challenge, and the data obtained was then compared between the group of pigs that presented clinical signs of respiratory disease and the group of pigs that were considered healthy. Clear changes in temperature and HR were observed six or more hours before the clinical observations identified sick animals. Conversely, changes in RR were detected in the last period of observations, around the time when clinical signs started to be present. The second of these experiments used different cameras and included improvements to the proposed methods to monitor pigs constantly and in a commercial setting. A total of 48 pigs were monitored between 9 and 20 weeks of age. Eye-temperature, HR and RR measurements were compared between the pigs that were identified as sick and those that were considered healthy. Similarly to what was observed in the previous study, changes in these parameters were identified before the clinical observations indicated signs of illness (up to 2 days before), where the earliest changes were observed in eye-temperature and HR, and the latest changes were observed in RR. This thesis provides evidence that computer vision techniques may be suitable as a non-invasive method for monitoring farm animals. Therefore, it prompts further investigation via controlled studies to continue the development and automatization of these techniques, leading to the improvement of science-based industry-relevant monitoring systems.