Search

As is the case for many National Statistics Institutes, the United States Department of Agriculture's (USDA's) National Agricultural Statistics Service (NASS) has observed dwindling survey response rates, and the requests for more information at finer temporal and spatial scales have led to increased response burdens. Non-survey data are becoming increasingly abundant and accessible. Consequently, NASS is exploring the potential to complete some or all of a survey record using non-survey data, which would reduce respondent burden and potentially lead to increased response rates. In this paper, the focus is on a large set of records associated with potential farms, which are operations with undetermined farm status (farm/non-farm) and are referred to here as operations with unknown status (OUS). Although they usually have some agriculture, most OUS records are eventually classified as non-farms. Those OUS that are classified as farms tend to have higher proportions of producers from under-represented groups compared to other records. Determining the probability that an OUS record is a farm is an important step in the imputation process. The OUS records that responded to the 2017 U.S. Census of Agriculture were used to develop models to predict farm status using multiple data sources. Evaluated models include bootstrap random forest (RF), logistic regression (LR), neural network (NN), and support vector machine (SVM). Although the SVM had the best outcomes for three of the five metrics, the sensitivity for identifying farms was the lowest (13.8%). The NN model had a sensitivity of 80.5%, which was substantially higher than the other models, and its specificity of 45.3% was the lowest of all models. Because sensitivity was the primary metric of interest and the NN performed reasonably well on the other metrics, the NN was selected as the preferred model. [ABSTRACT FROM AUTHOR]

As is the case for many National Statistics Institutes, the United States Department of Agriculture's (USDA's) National Agricultural Statistics Service (NASS) has observed dwindling survey response rates, and the requests for more information at finer temporal and spatial scales have led to increased response burdens. Non-survey data are becoming increasingly abundant and accessible. Consequently, NASS is exploring the potential to complete some or all of a survey record using non-survey data, which would reduce respondent burden and potentially lead to increased response rates. In this paper, the focus is on a large set of records associated with potential farms, which are operations with undetermined farm status (farm/non-farm) and are referred to here as operations with unknown status (OUS). Although they usually have some agriculture, most OUS records are eventually classified as non-farms. Those OUS that are classified as farms tend to have higher proportions of producers from under-represented groups compared to other records. Determining the probability that an OUS record is a farm is an important step in the imputation process. The OUS records that responded to the 2017 U.S. Census of Agriculture were used to develop models to predict farm status using multiple data sources. Evaluated models include bootstrap random forest (RF), logistic regression (LR), neural network (NN), and support vector machine (SVM). Although the SVM had the best outcomes for three of the five metrics, the sensitivity for identifying farms was the lowest (13.8%). The NN model had a sensitivity of 80.5%, which was substantially higher than the other models, and its specificity of 45.3% was the lowest of all models. Because sensitivity was the primary metric of interest and the NN performed reasonably well on the other metrics, the NN was selected as the preferred model. [ABSTRACT FROM AUTHOR]

Piglets are born into a relatively cool environment, resulting in body heat loss, especially at low environmental temperatures. Drying and warming reduces this heat loss, but may be less effective at the higher temperatures experienced in farrowing houses in summer. This study compared the effect of warming and drying piglets at birth on post-natal rectal temperature under winter vs. summer conditions. The study was carried out at a commercial facility using a CRD with 96 sows/litters in a 2×3 factorial arrangement of treatments (applied at birth): 1) Piglet Management: Control (no treatment); Warmed (in a heated box for 30 min); Dried+Warmed (dried with desiccant and warmed); 2) Season: Winter (January-March; farrowing room temperature 21.0 ± 1.65°C); Summer (August-September; 25.3 ± 1.67°C). At birth, piglets were weighed; rectal temperature was measured at 0, 10, 20, 30, 45, 60, 120, and 1440 min after birth. Data were analyzed using PROC MIXED of SAS. Litter was the experimental unit; piglet was a subsample of litter. The model included Season, Piglet Management, and the interaction. There were treatment interactions (P < 0.05) for rectal temperatures at all measurement times except at birth. For both Seasons, the Dried+Warmed treatment resulted in greater (P < 0.05) temperatures than Warmed between 20 and 60 min; both treatments had greater (P < 0.05) temperatures than the Control between 10 and 120 min. Rectal temperatures were lower (P < 0.05) in Winter than Summer for the Control and Warmed treatments between 10 and 60 min. However, for the Dried+Warmed treatment, rectal temperatures were similar between Seasons at all times after 10 min. In conclusion, drying and warming piglets at birth was more effective for reducing rectal temperature decline than warming alone and this effect was greater in winter than summer. This research was funded by the National Pork Board. [ABSTRACT FROM AUTHOR]

Piglets often experience hypothermia early after birth. Previous research has suggested that drying piglets and administration of oxygen (a potential treatment for asphyxiation) at birth may increase post-natal rectal temperatures. The objective of this study was to determine the effects of drying and administering oxygen at birth on piglet rectal temperature over the first 24 h after birth. The study, conducted at a commercial facility, used a CRD with 42 sows/litters randomly allotted at start of farrowing to 3 treatments (applied at birth): Control (no drying or oxygenation); Dried (using a cellulose-based desiccant); Dried+Oxygen [dried and placed in a chamber (40% oxygen) for 20 min]. At birth, piglets were weighed and uniquely identified. Rectal temperature was measured at 0, 20, 30, 45, 60, 120, and 1440 min after birth. Data were analyzed using PROC MIXED of SAS. Litter was the experimental unit; piglet was a subsample of litter. The statistical model included effects of treatment, time of measurement, and the interaction. Both the Dried and Dried+Oxygen treatments had greater (P < 0.05) rectal temperatures than the Control between 20 and 120 min. However, the Dried+Oxygen treatment had lower (P < 0.05) rectal temperatures than the Dried treatment between 20 and 60 minutes. Temperatures at 1440 min were lower (P < 0.05) for the Dried+Oxygen than the other treatments; however, differences were small. In conclusion, drying piglets at birth increased rectal temperatures over the first 2 h after birth. The combination of drying piglets at birth and placement in an oxygen chamber for 20 min was less effective at moderating post-natal temperature changes than drying alone. Further research on piglet oxygenation is necessary to understand the reason for these reduced temperatures, and whether this treatment affects pre-weaning mortality. This research was funded by the National Pork Board. [ABSTRACT FROM AUTHOR]

Newborn piglets are susceptible to hypothermia, and associated mortality. Previous research suggested exposing neonatal piglets to elevated oxygen concentrations reduced post-natal body temperature decline. This study evaluated the effects of oxygen administration at birth on piglet temperatures over 24 h after birth. This study used 296 piglets (24 litters) and was carried out in a standard farrowing facility consisting of crates within pens with a heat lamp. A GRBD was used; sow was the blocking factor; piglet the experimental unit. A replicate was 3 littermates of similar BW; piglets were randomly allotted at birth within replicate to treatment: 1) Control [dried at birth (with a desiccant); returned to farrowing pen]; 2) Ambient Oxygen (dried; placed in a chamber at ambient temperature and oxygen concentration for 20 min); 3) 40% Oxygen (dried; placed in a chamber at ambient temperature and 40% oxygen concentration for 20 min). Piglets were weighed at birth. Rectal temperature was measured at 0, 20, 30, 45, 60, 90, 120, 180, 240 and 1440 min after birth. Data were analyzed using PROC MIXED of SAS. The model included fixed effects of treatment, time of measurement (repeated measure), and the interaction and random effects of block and replicate. There was no treatment effect (P > 0.05) on temperatures at birth and between 120 and 1440 min. Compared to the Control, temperatures were lower (P < 0.05) for the 40% Oxygen treatment between 20 and 90 min and for the Ambient Oxygen treatment between 20 and 45 min. Temperatures at 60 and 90 min for the Ambient Oxygen treatment were intermediate to and not different (P > 0.05) from the other treatments. Results suggest that exposing neo-natal piglets to an oxygen-rich environment in a chamber does not reduce post-natal temperature decline. This research was funded by the National Pork Board. [ABSTRACT FROM AUTHOR]

Piglets are susceptible to chilling early after birth, which is a major pre-disposing factor for pre-weaning mortality (PWM). This study evaluated the effect of drying and warming piglets at birth on PWM at a commercial facility. A CRD was used with 802 sows (litters), allotted at start of farrowing to 2 treatments (applied at birth): Control (no treatment); Dried+Warmed (dried with a cellulose-based desiccant and warmed in a heated box for 30 min). Piglets were weighed at birth and weaning; PWM was recorded. The effects of farrowing room temperature on PWM were evaluated by classifying litters as being born under Cool or Warm conditions (< 25° C and = 25°C, respectively). Similarly, effects of birth weight on PWM were evaluated by classifying piglets into 3 Birth Weight Categories (BWC; < 1.0 kg, 1.0-1.5 kg, or > 1.5 kg). PROC GLIMMIX and MIXED of SAS were used to analyze PWM and other data, respectively. Litter was the experimental unit; piglet a subsample of litter. The model included fixed effects of treatment, and farrowing room temperature or BWC as appropriate, and the random effect of piglet within litter. Rectal temperature at 30 min after birth, measured on a sub-sample of 10% of litters, was greater (P < 0.05) for the Dried+Warmed than the Control treatment. There was no effect (P > 0.05) of drying and warming piglets on weaning weight or overall PWM. Additionally, PWM was similar across treatments within each BWC. However, the Dried+Warmed treatment reduced (P < 0.05) PWM compared to the Control under Cool but not Warm farrowing room temperatures. In conclusion, this suggests that drying and warming piglets at birth increases rectal temperature and may be an effective method to reduce piglet PWM under cooler farrowing room temperatures. This research was funded by the National Pork Board. [ABSTRACT FROM AUTHOR]

There is limited understanding of optimum crossfostering methods to use to maximize piglet performance. This study evaluated effects of within-litter birth weight variation after cross-fostering on pre-weaning piglet removals (PR; morbidity and mortality) and ADG. A hierarchical incomplete block design was used (blocking factors day of farrowing and sow parity and structure) with a 3x2 factorial arrangement of treatments: 1) Birth Weight Category (BWC): Light (< 1.0 kg), Medium (1.0-1.5 kg), or Heavy (1.5-2.0 kg); 2) Litter Composition (LC): Uniform (piglets of the same BWC), and Mixed (equal numbers of piglets from each BWC). Piglets were weighed 24 h after birth and allotted to form litters of 15 cross-fostered piglets. The experimental unit was 5 piglets of the same BWC (3 experimental units per litter). A total of 102 litters were allotted to 17 blocks of 6 litters (1 Uniform litter of each BWC; 3 Mixed litters) with 51 replicates (3 replicates/block of 6 litters). Weaning weights and PR were measured. PROC GLIMMIX and MIXED of SAS were used to analyze PR and other data, respectively. Models included BWC, LC, the interaction, and replicate within block. There were treatment interactions (P < 0.05) for all measures except birth weight. There was no effect (P > 0.05) of LC on weaning weight or ADG for Light or Medium piglets; Heavy piglets had greater (P < 0.05) weaning weight and ADG in Mixed than in Uniform litters. PR were greater (P < 0.05) for Light piglets in Mixed than in Uniform litters, and for Heavy piglets in Uniform than in Mixed litters. PR for Medium piglets were similar (P > 0.05) across LC treatments. In conclusion, rearing cross-fostered piglets in Uniform litters reduced PR for Light piglets, but increased PR and reduced ADG of Heavy piglets, with no effect for Medium piglets. [ABSTRACT FROM AUTHOR]