An extant model was evaluated to assess its adequacy for nutrient management planning for swine operations in Manitoba with regards to phosphorus (P) and nitrogen (N) excretion and the land base for the optimum spreading of manure based on P requirement of crops. Two dietary treatments were used, control diet formulated to meet the requirement of pigs for nutrients as per the recommendations of NRC (No-phytase) and a diet formulated with P level in the No-phytase diet reduced by an average of 0.1 percentage units and amended with microbial phytase at 500 FTU/kg (Phytase). Data was generated from starter to finisher pigs (10 per dietary treatment) and sows (9 per dietary treatment) to evaluate the model. The model was found to be adequate for the prediction of P outputs from starter to finisher but not sows. Model was found to be inadequate for prediction of N output.

Odour emissions and instantaneous downwind odour plumes were measured on two 3000-sow swine farrowing farms located in a flat area of southern Manitoba, one farm (Farm A) with open earthen manure storage (EMS) and another (Farm B) with negative air pressure (NAP) covered EMS. Three commonly used dispersion models (ISCST3, AUSPLUME, and INPUFF-2) were used to predict downwind odour distributions on the two farms. Dispersion predictions were based on the measured odour emission data for each farm and on-site weather data recorded by a portable weather station. The peak-to-mean ratios of downwind odour intensity were computed from field odour intensity measurements and analysed against averaging time, downwind distance, and atmosphere stability class. It was found that the average odour emission rate from the negative pressure covered earthen manure storage (NAP EMS) was negligible in comparison with the open EMS (0.3 vs 20.3 OU/ s-m2). Downwind odour intensity measured by trained human sniffers on Farm A with covered manure storage was significantly (P<0.05) lower than that on Farm B with open manure storage at 100 and 500 m, but the difference in odour intensity at 1000 m was not significant (P>0.05) between the two farms. A 46% difference in odour emission rate between Farms A and B resulted in a 14% difference in the separation distance for odour annoyance-free between the two farms. When three commonly used dispersion models, namely AUSPLUME, ISCST3, and INPUFF-2, were used to predict downwind odour from the farms, the percentage of agreement between model predictions and field measurements was adequate for downwind distances of 500 and 1000 m, but relatively low for 100 m for all three models. Since the long-distance (>1000 m) predictions are of more practical value, all three models were considered to be adequate in predicting odour downwind from the swine operations. The peak-to-mean ratios of downwind odour intensity were computed from field odour intensity measurements and analysed against averaging time, downwind distance, and atmospheric stability class. The peak-to-mean ratio of field odour intensity increases with averaging time and downwind distance, and unstable atmospheric conditions.

This project explored the feasibility of performing swine influenza surveillance in Manitoba using provincial veterinary diagnostic laboratory data and a farm premises identification registry. Diagnoses of swine influenza using polymerase chain reaction (PCR) were obtained from the veterinary laboratory database and linked with registry data on farm location and characteristics. Statistical and space-time analyses, including the Cuzick and Edwards test, Kulldorff Spatiotemporal scan, the Knox test and the modified CuSum method, were used to determine the time and spatial patterns of swine influenza in Manitoba. Analysis showed that swine influenza was endemic but also seasonal and that the frequency of diagnosis was increasing in time. Swine influenza was clustered in several regions across the province, including the southeast, and was clustered in time, particularly during the later time periods of the study. This study demonstrated that the farm premises identification registry is a crucial component of disease surveillance in animals.

The resource requirements and social behaviour of pigs housed in large groups were examined in a series of experiments. (i) The nipple drinker requirement was addressed by using four treatments (60 pigs, 3 drinkers; 20 pigs, 1 drinker; 60 pigs, 6 drinkers and 20 pigs, 2 drinkers). Drinker provision had no effect on water use, but in a larger group more water was used in less time (p<0.001). The diurnal pattern of drinking, overt aggression and lesion score indicated no difference between treatments. (ii) Pigs, housed on deep straw in groups of 20 or 80, were provided with a low (50<I> </I>kg/m²) or high (32 kg/m²) floor space allowance. Large groups had a lower growth rate. Skin lesions were elevated and immune response was lowered by a low space allowance. (iii) Two feeder space allowances (32.5<I> </I>and 42.5<I> </I>mm/pig) for pigs housed in groups of 20 or 80 were investigated. Food intake was lower in the low feeder allowance treatments and pigs in large groups tended to have a reduced growth rate. (iv) Pigs from the same pen in an unfamiliar arena maintained a similar degree of proximity regardless of origin group size. (v) Pigs from groups of 80 demonstrated reduced aggressiveness (increased latency to fight, decreased rate of aggression) towards unacquainted pigs in an arena, but showed even less aggression towards pen mates. (vi) No evidence of spatial sub-division of the large group into smaller units capable of maintaining a dominance hierarchy was found. Resource provision and group size largely did not interact, but may independently compromise productivity and behaviour. A large group was associated with a reduction in performance. The dominance hierarchy was of less importance in large groups, despite recognition being intact, and sub-grouping behaviour did not provide an alternative strategy for group social organisation.

No description available.

No description available.

No description available.

Typescript. / Thesis (Ph. D.)--University of Hawaii at Manoa, 1977. / Bibliography: leaves [83]-92. / Microfiche. / viii, 92 leaves ill

Thesis (PhD)--University of South Australia, 1998

The intestinal spirochaete Brachyspira hyodysenteriae is the causative agent of swine dysentery (SD), a diarrhoeal disease of pigs which has significant economic impact worldwide. Controlling SD remains problematic, particularly as there is no effective vaccine and there are few definitive diagnostic methods available. In this study, a partial genomic sequence of B. hyodysenteriae was screened in silico. A total of 19 putative open reading frames (ORFs) encoding outer-membrane proteins then were selected and these were subjected to a laboratory screening process. To select potential universal vaccines, a preliminary study was conducted using PCR to determine the distribution of the putative genes in 23 strains of B. hyodysenteriae. A total of 17 of the 19 ORFs were considered to be suitable for further testing as they were found to be present in the majority of strains investigated. After molecular cloning and protein expression and purification, of 19 cloned candidate molecules derived from 17 genes (one large gene was divided into two parts encoding N and C terminal proteins, respectively), 14 were expressed in E. coli and the recombinant proteins were successfully produced. A variety of sera from pigs naturally and experimentally infected with B. hyodysenteriae were tested for reactivity with the 14 recombinant proteins in an immunoblotting assay. Seven molecules from six genes reacted strongly with the tested sera, and therefore were selected and used to immunize mice. All these proteins generated a specific antibody response. Post-immunization sera raised against each recombinant protein had the capacity to agglutinate B. hyodysenteriae cells, and also recognized the cognate proteins of B. hyodysenteriae in cell extracts. Further sequencing analysis demonstrated that these molecules were highly conserved in the genomes of different B.hyodysenteriae strains. Therefore, from the genomic based study, the products of six genes were identified as promising candidates for vaccines or as diagnostic targets. Four genes were expressed on a large scale, the product (NAV-H7, NAV-H17 Cterminal, NAV-H34 and NAV-H42) were combined into one vaccine, and then this preparation was used to immunise pigs that subsequently were challenged with B. hyodysenteriae. These antigens generate systemic and colonic antibody responses, and vaccination tended both to delay the onset of clinical signs and attenuate lesion development. Hence these recombinant proteins showed promise as components for further SD vaccines. Recombinant proteins from the selected genes also were used as antigens in class-specific ELISAs used as serological assays for SD. Three antigens (NAV-H8, NAV-H42 and Bhlp29.7) were selected as good indicators of seroconversion in IgM ELISAs, and these were evaluated further using a large range of serum samples. The NAV-H8 IgM ELISA using a cut-off value 2.5 times the mean value of all negative pigs could be used as a herd test for SD, and both the NAV-H8 and NAV-H42 IgM ELISAs had potential for detecting exposure to B. hyodysenteriae at the pig level.