An assessment of Walk-over-Weighing to estimate short-term individual forage intake in sheep

Kalman filter with impulse noised outliers: a robust sequential algorithm to filter data with a large number of outliers

Impulse noised outliers are data points that differ significantly from other observations. They are generally removed from the data set through local regression or the Kalman filter algorithm. However, these methods, or their generalizations, are not well suited when the number of outliers is of the same order as the number of low-noise data (often called nominal measurement). In this article, we propose a new model for impulsed noise outliers. It is based on a hierarchical model and a simple linear Gaussian process as with the Kalman Filter. We present a fast forward-backward algorithm to filter and smooth sequential data and which also detects these outliers. We compare the robustness and efficiency of this algorithm with classical methods. Finally, we apply this method on a real data set from a Walk Over Weighing system admitting around 60 % of outliers. For this application, we further develop an (explicit) EM algorithm to calibrate some algorithm parameters.

Improving Dry Matter Intake Estimates Using Precision Body Weight on Cattle Grazed on Extensive Rangelands

An essential component required for calculating stocking rates for livestock grazing extensive rangeland is dry matter intake (DMI). Animal unit months are used to simplify this calculation for rangeland systems to determine the rate of forage consumption and the cattle grazing duration. However, there is an opportunity to leverage precision technology deployed on rangeland systems to account for the individual animal variation of DMI and subsequent impacts on herd-level decisions regarding stocking rate. Therefore, the objectives of this study were, first, to build a precision system model (PSM) to predict total DMI (kg) and required pasture area (ha) using precision body weight (BW), and second, to evaluate differences in PSM-predicted stocking rates compared to the traditional herd-level method using initial or estimated mid-season BW. A deterministic model was constructed in both Vensim (version 10.1.2) and Program R (version 4.2.3) to incorporate individual precision BW data into a commonly used rangeland equation using %BW to estimate individual DMI, daily herd DMI, and area (ha) required to meet animal DMI requirements throughout specific grazing periods. Using the PSM, differences in outputs were evaluated using three scenarios: (1) initial BW (business as usual); (2) average mid-season BW; and (3) individual precision BW using data from two precision rangeland experiments conducted at the South Dakota State University Cottonwood Field Station. The data from the two experiments were used to develop PSM case studies. The trial data were collected using precision weight data (SmartScale™) collected from replacement heifers (Case study 1, n = 60) and steers (Case study 2, n = 254) grazing native rangeland. In Case study 1 (heifers), Scenario 1 versus Scenario 3 resulted in an additional 73.41 ha required. Results from Case study 2 indicated an average additional 4.4 ha required per pasture when comparing Scenario 3 versus Scenario 1. Sensitivity analyses resulted in a difference between maximum and minimum simulated values of 27,995 and 4265 kg forage consumed, and 122 and 8.9 pasture ha required for Case studies 1 and 2, respectively. Thus, results from the scenarios indicate an opportunity to identify both under- and over-stocking situations using precision DMI estimates, which helps to identify high-leverage precision tools that have practical applications for enhancing animal and plant productivity and environmental sustainability on extensive rangelands.

Short-term responses of meat ewes facing an acute nutritional challenge in early-mid lactation

Simulating a consequence of a climate change event on feed availability, responses of Mediterranean meat ewes facing an acute undernutritional challenge (CHA; i.e., fed only low nutritional value cereal straw) were evaluated at a sensitive physiological stage (i.e., early suckling). Forty Romane ewes were chosen at early-mid pregnancy (around 2 mo) according to parity (20 primiparous, PRIM; 20 multiparous, MULT); feed efficiency genetic line of their sires (residual feed intake [RFI]; efficient, RFI-, n = 10 per parity; inefficient, RFI+, n = 10 per parity); litter size (i.e., bearing twins, diagnosed by ultrasonography); body weight (BW, kg) and body condition score (BCS) (initial BW and BCS [mean ± SD]: 51.6 ± 7.41 kg; 2.5 ± 0.20, respectively; representing flock' averages per parity). Effects on dry matter intake (DMI), ewes' BW and BCS, subcutaneous dorsal fat thickness (DFT), energy metabolism (plasma non-esterified fatty acids [NEFA], β-hydroxybutyrate (β-OHB), glucose, urea, triiodothyronine [T3]), and lambs' growth (BW and average daily gain [ADG]; g/d) were examined before, during and after CHA. Individuals' profiles of the response-recovery to CHA were described using a piecewise mixed-effects model. The fixed effect of parity and genetic line and the random effect of individual (ewe) were considered. A linear mixed-effects model was fitted to explore the effects on lambs' growth. The 2-d straw-only CHA had significant effects on most of the recorded parameters. Meaningful drops and recoveries were observed on ewes' DMI, BW, and DFT with effect on postchallenge levels. BW, BCS, DFT, or DMI were also affected by parity (MULT > PRIM) but not by genetic line. Plasma NEFA, β-OHB, glucose, urea, and T3 responded well to CHA with drops in T3, urea, and glucose levels, whereas NEFA and β-OHB significantly increased after CHA. MULT ewes presented sharper β-OHB recovery from CHA than PRIM (P ≤ 0.05). With this study, we provide tangible and necessary data for an emerging field of research. Our results give new insights into how such a short and abrupt CHA affects some key zootechnical and physiological parameters, and to what extent the impacts of CHA and the ewes' response-recovery are influenced. It also revealed potential between-individual differences in the adaptive capacities of ewes, which require further exploration.

Novel epidemiological model of gastrointestinal nematode infection to assess grazing cattle resilience by integrating host growth, parasite, grass and environmental dynamics

Gastrointestinal nematode (GIN) infections are ubiquitous and often cause morbidity and reduced performance in livestock. Emerging anthelmintic resistance and increasing change in climate patterns require evaluation of alternatives to traditional treatment and management practices. Mathematical models of parasite transmission between hosts and the environment have contributed towards the design of appropriate control strategies in ruminants, but have yet to account for relationships between climate, infection pressure, immunity, resources, and growth. Here, we develop a new epidemiological model of GIN transmission in a herd of grazing cattle, including host tolerance (body weight and feed intake), parasite burden and acquisition of immunity, together with weather-dependent development of parasite free-living stages, and the influence of grass availability on parasite transmission. Dynamic host, parasite and environmental factors drive a variable rate of transmission. Using literature sources, the model was parametrised for Ostertagia ostertagi, the prevailing pathogenic GIN in grazing cattle populations in temperate climates. Model outputs were validated on published empirical studies from first season grazing cattle in northern Europe. These results show satisfactory qualitative and quantitative performance of the model; they also indicate the model may approximate the dynamics of grazing systems under co-infection by O. ostertagi and Cooperia oncophora, a second GIN species common in cattle. In addition, model behaviour was explored under illustrative anthelmintic treatment strategies, considering impacts on parasitological and performance variables. The model has potential for extension to explore altered infection dynamics as a result of management and climate change, and to optimise treatment strategies accordingly. As the first known mechanistic model to combine parasitic and free-living stages of GIN with host feed-intake and growth, it is well suited to predict complex system responses under non-stationary conditions. We discuss the implications, limitations and extensions of the model, and its potential to assist in the development of sustainable parasite control strategies.

What, how, and how much do herbivores eat? The Continuous Bite Monitoring method for assessing forage intake of grazing animals

Determining herbage intake is pivotal for studies on grazing ecology. Direct observation of animals allows describing the interactions of animals with the pastoral environment along the complex grazing process. The objectives of the study were to evaluate the reliability of the continuous bite monitoring (CBM) method in determining herbage intake in grazing sheep compared to the standard double-weighing technique method during 45-min feeding bouts; evaluate the degree of agreement between the two techniques; and to test the effect of different potential sources of variation on the reliability of the CBM. The CBM method has been used to describe the intake behavior of grazing herbivores. In this study, we evaluated a new approach to this method, that is, whether it is a good proxy for determining the intake of grazing animals. Three experiments with grazing sheep were carried out in which we tested for different sources of variations, such as the number of observers, level of detail of bite coding grid, forage species, forage allowance, sward surface height heterogeneity, experiment site, and animal weight, to determine the short-term intake rate (45 min). Observer (Pexp1  = 0.018, Pexp2  = 0.078, and Pexp3  = 0.006), sward surface height (Pexp2  < 0.001), total number of bites observed per grazing session (Pexp2  < 0.001 and Pexp3  < 0.001), and sward depletion (Pexp3  < 0.001) were found to affect the absolute error of intake estimation. The results showed a high correlation and agreement between the two methods in the three experiments, although intake was overestimation by CBM on experiments 2 and 3 (181.38 and 214.24 units, respectively). This outcome indicates the potential of CBM to determining forage intake with the benefit of a greater level of detail on foraging patterns and components of the diet. Furthermore, direct observation is not invasive nor disrupts natural animal behavior.

Short-term liveweight changes of dairy cows measured by stationary and walk-over weighing scales

Monitoring and detecting individual cows' liveweight (LW) and liveweight change (LWC) are important for estimation of nutritional requirements and health management, and could be useful to measure short-term feed intake, water consumption, defecation, and urination. Walk-over weighing (WOW) systems can facilitate measurements of LW for these purposes, providing automated LW recorded at different times of the day. We conducted a field study to (1) quantify the contribution of feed and water intake, as well as urine and feces excretions, to short-term LWC and (2) determine the feasibility of stationary and WOW scales to detect subtle changes in LW as a result of feed and water intake, urination, and defecation. In this experiment, 10 cows walked through a WOW system and then stood individually on a stationary scale collecting weights at 10 and 3.3 Hz, respectively. Cows were offered 4 kg of feed and 10 kg of water on the stationary scale. For each animal, LW before and after eating and drinking was then calculated using different approaches. Liveweight change was calculated as the difference between the initial and final LW before and after eating and drinking for each statistical measure. The weights of feed intake, water consumption, urination, and defecation were measured and used as predictors of LWC. Urine and feces were collected from individual cows while the cow was on the scale, using a container, and weighed separately. The agreement between LWC measured using either stationary or WOW scales was assessed to determine the sensitivity of the scales to detect subtle changes in LW using the coefficient of determination (R²), Lin's concordance correlation coefficient (CCC), and mean bias. The prediction model showed that most of the regression coefficients were not significantly different from +1.0 for feed and water, or -1.0 for urine and feces. The R² and CCC values demonstrated a satisfactory agreement between calculated and stationary LWC and values ranged from 0.60 to 0.92 and 0.71 to 0.94, respectively. A moderate agreement was achieved between calculated and automated LWC with R² and Lin's CCC values of 0.45 to 0.63 and 0.60 to 0.74, respectively. Therefore, results demonstrated that new algorithms and data processing methods need to be continuously explored and improved to obtain accurate measurements of LW to measure changes in LW, especially from WOW scales.

Measuring liveweight changes in lactating dairy ewes with an automated walk-over-weighing system

Monitoring liveweight (LW) is an important part of sound management practices at the individual and flock level (e.g., controlling for nutritional status based on body condition, reproduction, and health-related issues), but it is time consuming and stressful. To our knowledge, no literature has reported on the evaluation of automated weighing systems in dairy sheep as an alternative to conventional static scales. The objective of this research was to evaluate the practical feasibility of using an automated walk-over-weighing (WoW) prototype to measure daily LW changes in dairy ewes without human intervention. We used adult Lacaune dairy ewes in 2 complementary trials conducted indoors. Trial 1 aimed at evaluating the repeatability, precision, and accuracy of LW measures recorded using WoW scales compared with a static scale (the gold standard). Forty-two adult ewes (LW ± standard deviation = 71.3 ± 10.4 kg) were randomly drafted from the main flock and used in a 1-day session. The trial included 3 passages. In each passage, ewes were weighed first on a static scale; once a static position was achieved and LW recorded, they continued the circuit and immediately traversed the WoW scale for an automated LW record. Trial 2 aimed to demonstrate the feasibility of using the WoW device under real-world conditions in a dairy sheep-farming system. The WoW scale was installed in the exit race of the milking parlor and evaluated over 7 wk with adult ewes in mid lactation (n = 93; LW 78.5 ± 8.1 kg). Once the ewes were acclimated to the WoW system, 1 group of ewes (n = 48) continued to receive the same feeding regimen (controls), and the other group (n = 45) underwent a nutritional challenge [challenged; 2 wk of undernutrition and then back to control regimen (refeeding) for 1 wk]. We evaluated the ability of the WoW to detect small changes in LW. We collected LW data (2 weighings per ewe per day) from the WoW after each of the 2 milking sessions (morning and evening). We also obtained LW values by weighing the ewes using a static scale once a week. The automated WoW system showed substantial agreement with the gold standard when assessed using Lin's concordance correlation coefficient and Bland and Altman's method, largely due to high repeatability. The WoW system was adequate for detecting small daily variations in LW during undernutrition and refeeding periods. Misbehaviors resulted in spurious WoW values in trial 2, requiring us to use filtration methods to exclude outlier weights and allow meaningful assessment of small LW changes. The WoW system evaluated here is an alternative to the static scales conventionally used on dairy sheep farms. If sound filtration of raw data is applied, WoW could contribute to the close (daily) monitoring of individual LW without operator intervention (i.e., voluntary weighing) and taking animal welfare into account (i.e., no stress related to the weighing session on static scales).

Advancements in sensor technology and decision support intelligent tools to assist smart livestock farming

Remote monitoring, modern data collection through sensors, rapid data transfer, and vast data storage through the Internet of Things (IoT) have advanced precision livestock farming (PLF) in the last 20 yr. PLF is relevant to many fields of livestock production, including aerial- and satellite-based measurement of pasture’s forage quantity and quality; body weight and composition and physiological assessments; on-animal devices to monitor location, activity, and behaviors in grazing and foraging environments; early detection of lameness and other diseases; milk yield and composition; reproductive measurements and calving diseases; and feed intake and greenhouse gas emissions, to name just a few. There are many possibilities to improve animal production through PLF, but the combination of PLF and computer modeling is necessary to facilitate on-farm applicability. Concept- or knowledge-driven (mechanistic) models are established on scientific knowledge, and they are based on the conceptualization of hypotheses about variable interrelationships. Artificial intelligence (AI), on the other hand, is a data-driven approach that can manipulate and represent the big data accumulated by sensors and IoT. Still, it cannot explicitly explain the underlying assumptions of the intrinsic relationships in the data core because it lacks the wisdom that confers understanding and principles. The lack of wisdom in AI is because everything revolves around numbers. The associations among the numbers are obtained through the “automatized” learning process of mathematical correlations and covariances, not through “human causation” and abstract conceptualization of physiological or production principles. AI starts with comparative analogies to establish concepts and provides memory for future comparisons. Then, the learning process evolves from seeking wisdom through the systematic use of reasoning. AI is a relatively novel concept in many science fields. It may well be “the missing link” to expedite the transition of the traditional maximizing output mentality to a more mindful purpose of optimizing production efficiency while alleviating resource allocation for production. The integration between concept- and data-driven modeling through parallel hybridization of mechanistic and AI models will yield a hybrid intelligent mechanistic model that, along with data collection through PLF, is paramount to transcend the current status of livestock production in achieving sustainability.

Sensing solutions for improving the performance, health and wellbeing of small ruminants

Diversity of production systems and specific socio-economic barriers are key reasons explaining why the implementation of new technologies in small ruminants, despite being needed and beneficial for farmers, is harder than in other livestock species. There are, however, helpful peculiarities where small ruminants are concerned: the compulsory use of electronic identification created a unique scenario in Europe in which all small ruminant breeding stock became searchable by appropriate sensing solutions, and the largest small ruminant population in the world is located in Asia, close to the areas producing new technologies. Notwithstanding, only a few research initiatives and literature reviews have addressed the development of new technologies in small ruminants. This Research Reflection focuses on small ruminants (with emphasis on dairy goats and sheep) and reviews in a non-exhaustive way the basic concepts, the currently available sensor solutions and the structure and elements needed for the implementation of sensor-based husbandry decision support. Finally, some examples of results obtained using several sensor solutions adapted from large animals or newly developed for small ruminants are discussed. Significant room for improvement is recognized and a large number of multiple-sensor solutions are expected to be developed in the relatively near future.