Compilations and updates on residual feed intake in sheep

The increasing global demand for food and the strong effect of climate change have forced animal science to advance regarding new methods of selection in search of more efficient animals in production systems. Feed consumption represents more than 70% of the costs of sheep farms, and more efficient animals can increase the farmers' profitability. One of the main measures of feed efficiency is estimated residual feed intake (RFI), created in 1963 by Robert Koch for estimation in cattle and later adapted for sheep. Animals with negative RFI values (RFI-) are more efficient than animals with positive values (RFI+), with influence on the variables of performance, carcass quality and production of enteric gases. The RFI is the most common and accepted metric of the feed efficiency trait for genetic selection, since it is independent of growth traits, unlike the feed conversion ratio. The purpose of this review article was to present updated literature information on the relationship of RFI estimates with performance measures, molecular markers, greenhouse gas production and feed efficiency, the technical aspects and physiological basis of metabolic in sheep.

Genetic parameters for residual feed intake, methane emissions, and body composition in New Zealand maternal sheep

There is simultaneous interest in improving the feed efficiency of ruminant livestock and reducing methane (CH₄) emissions. The relationship (genetic and phenotypic) between feed efficiency (characterized as residual feed intake: RFI) and greenhouse gases [methane (CH₄) and carbon dioxide (CO₂)] traits in New Zealand (NZ) maternal sheep has not previously been investigated, nor has their relationship with detailed estimates of body composition. To investigate these relationships in NZ maternal sheep, a feed intake facility was established at AgResearch Invermay, Mosgiel, NZ in 2015, comprising automated feeders that record individual feeding events. Individual measures of feed intake, feeding behavior (length and duration of eating events), and gas emissions (estimated using portable accumulation chambers) were generated on 986 growing maternal ewe lambs sourced from three pedigree recorded flocks registered in the Sheep Improvement Limited database (www.sil.co.nz). Additional data were generated from a subset of 591 animals for body composition (estimated using ultrasound and computed tomography scanning). The heritability estimates for RFI, CH₄, and CH₄/(CH₄+CO₂) were 0.42 ± 0.09, 0.32 ± 0.08, and 0.29 ± 0.06, respectively. The heritability estimates for the body composition traits were high for carcass lean and fat traits; for example, the heritability for visceral fat (adjusted for body weight) was 0.93 ± 0.19. The relationship between RFI and CH₄ emissions was complex, and although less feed eaten will lead to a lowered absolute amount of CH₄ emitted, there was a negative phenotypic and genetic correlation between RFI and CH₄/(CH₄+CO₂) _of −0.13 ± 0.03 and −0.41 ± 0.15, respectively. There were also genetic correlations, that were different from zero, between both RFI and CH₄ traits with body composition including a negative correlation between the proportion of visceral fat in the body and RFI (−0.52 ± 0.16) and a positive correlation between the proportion of lean in the body and CH₄ (0.54 ± 0.12). Together the results provide the first accurate estimates of the genetic correlations between RFI, CH₄ emissions, and the body composition (lean and fat) in sheep. These correlations will need to be accounted for in genetic improvement programs.

Eating Time as a Genetic Indicator of Methane Emissions and Feed Efficiency in Australian Maternal Composite Sheep

Previous studies have shown reduced enteric methane emissions (ME) and residual feed intake (RFI) through the application of genomic selection in ruminants. The objective of this study was to evaluate feeding behaviour traits as genetic indicators for ME and RFI in Australian Maternal Composite ewes using data from an automated feed intake facility. The feeding behaviour traits evaluated were the amount of time spent eating per day (eating time; ETD; min/day) and per visit (eating time per event; ETE; min/event), daily number of events (DNE), event feed intake (EFI; g/event) and eating rate (ER; g/min). Genotypes and phenotypes of 445 ewes at three different ages (post-weaning, hogget, and adult) were used to estimate the heritability of ME, RFI, and the feeding behaviour traits using univariate genomic best linear unbiased prediction models. Multivariate models were used to estimate the correlations between these traits and within each trait at different ages. The response to selection was evaluated for ME and RFI with direct selection models and indirect models with ETE as an indicator trait, as this behaviour trait was a promising indicator based on heritability and genetic correlations. Heritabilities were between 0.12 and 0.18 for ME and RFI, and between 0.29 and 0.47 for the eating behaviour traits. In our data, selecting for more efficient animals (low RFI) would lead to higher methane emissions per day and per kg of dry matter intake. Selecting for more ETE also improves feed efficiency but results in more methane per day and per kg dry matter intake. Based on our results, ETE could be evaluated as an indicator trait for ME and RFI under an index approach that allows simultaneous selection for improvement in emissions and feed efficiency. Selecting for ETE may have a tremendous impact on the industry, as it may be easier and cheaper to obtain than feed intake and ME data. As the data were collected using individual feeding units, the findings on this research should be validated under grazing conditions.

Sheep residual feed intake and feeding behaviour: are ‘nibblers’ or ‘binge eaters’ more efficient?

Efficient liveweight gain is an important component of profitable livestock systems. In cattle, studies of residual feed intake (RFI) suggest that there are behavioural differences between efficient and inefficient animals; more efficient animals have less frequent, larger feed events. There is limited understanding of the association between feeding behaviour and feed conversion efficiency (measured as RFI) in sheep. We hypothesised that more efficient sheep would have lower daily number of meals but larger meal size than would less efficient sheep and that feeding behaviour would be repeatable between measurements conducted using the same sheep at different ages. Feeding behaviour was monitored at post-weaning (average 311 days) and hogget (average 533 days) age in a single cohort of maternal composite 2014-born ewes undergoing feed-efficiency testing. Feed intake (kg/day) and daily feeding behaviour (meal number, meal size and eating rate) were recorded by using automated feeders. Feed intake was recorded daily for a minimum of 40 days (mean 41 days), following a 14-day adaptation period. Animals were fed a pelleted hay-based diet, with liveweight measured three times per week. At both ages of measurement, a positive (P < 0.01) linear relationship was established between average daily number of meals and RFI. In these cohorts, as daily number of meals increased, RFI became more positive (less efficient). Meal size (kg DM) was also related to RFI (P < 0.05) when measured at the hogget age, with meal size decreasing with an increasing RFI (less efficient animals had smaller meals). These analyses suggest that sheep feeding behaviour, and in particular daily meal number, is associated with feed efficiency.

Benefits of including methane measurements in selection strategies

Estimates of genetic/phenotypic covariances and economic values for slaughter weight, growth, feed intake and efficiency, and three potential methane traits were compiled to explore the effect of incorporating methane measurements in breeding objectives for cattle and meat sheep. The cost of methane emissions was assumed to be zero (scenario A), A$476/t (based on A$14/t CO equivalent and methane's 100-yr global warming potential [GWP] of 34; scenario B), or A$2,580/t (A$30/t CO equivalent combined with methane's 20-yr GWP of 86; scenario C). Methane traits were methane yield (MY; methane production divided by feed intake based on measurements over 1 d in respiration chambers) or short-term measurements of methane production adjusted for live weight (MPadjWt) in grazing animals, e.g., 40-60 min measurements in portable accumulation chambers (PAC) on 1 or 3 occasions, or measurements for 1 wk using a GreenFeed Emissions Monitor (GEM) on 1 or 3 occasions. Feed costs included the cost of maintaining the breeding herd and growth from weaning to slaughter. Sheep were assumed to be grown and finished on pasture (A$50/t DM). Feed costs for cattle included 365 d on pasture for the breeding herd and averages of 200 d postweaning grow-out on pasture and 100 d feedlot finishing. The greatest benefit of including methane in the breeding objective for both sheep and cattle was as a proxy for feed intake. For cattle, 3 GEM measurements were estimated to increase profit from 1 round of selection in scenario A (no payment for methane) by A$6.24/animal (from A$20.69 to A$26.93) because of reduced feed costs relative to gains in slaughter weight and by A$7.16 and A$12.09/animal, respectively, for scenarios B and C, which have payments for reduced methane emissions. For sheep, the improvements were more modest. Returns from 1 round of selection (no methane measurements) were A$5.06 (scenario A), A$4.85 (scenario B), and A$3.89 (scenario C) compared to A$5.26 (scenario A), A$5.12 (scenario B), and A$4.72 (scenario C) for 1 round of selection with 3 PAC measurements. Including MY in the selection index was less profitable because it did not reduce feed costs relative to weight gain. Consequently, for strategies measuring MY but not MPadjWt (and with no estimate of feed intake in the production environment), proportionately greater emphasis was placed on increasing slaughter weight, and as a result, the decreases in methane emissions per animal and per unit of feed intake were smaller than for strategies that measured MPadjWt.

Comparison of repeated measurements of methane production in sheep over 5 years and a range of measurement protocols

Emissions of 710 ewes at pasture were measured for 1 h (between 09:00-16:30 h) in batches of 15 sheep in portable accumulation chambers (PAC) after an overnight fast continuing until 2 h before measurement, when the sheep had access to baled hay for 1 h. The test was used to identify a group of 104 low emitters (I-Low) and a group of 103 high emitters (I-Hi) for methane emissions adjusted for liveweight (CHawt). The 207 ewes selected at the initial study were remeasured in 5 repeat tests from 2009 through 2014 at another location. The first repeat used the original measurement protocol. Two modified protocols, each used in 2 yr, drafted unfasted sheep on the morning of the test into a yard or holding paddock until measurement. Emissions of the I-Hi sheep were higher (102-112%) than I-Low sheep in all subsequent PAC tests, with statistical significance ( < 0.05) in 3 tests. Tests without overnight fasting were simpler to conduct and had repeatabilities of 51 to 60% compared with 31 and 43% for the initial and first repeat tests, respectively. After habituation to a diet fed at 20 g/kg liveweight, 160 of the 207 sheep were measured in respiration chambers (RC); 10 high (Hi-10) and 10 low (Low-10) sheep were chosen, representing extremes (top and bottom 6.25%) for methane yield (MY; g CH/kg DMI). The Hi-10 group emitted 14% more methane (adjusted for feed intake) in a follow-up RC test, but Low-10 and Hi-10 sheep differed in only 1 of the 5 PAC tests, when Hi-10 sheep emitted less CHawt than Low-10 sheep ( = 0.002) and tended to eat less in the feeding opportunity ( = 0.085). Compared with their weight on good pasture, Low-10 sheep were proportionately lighter than Hi-10 sheep in the relatively poor pasture conditions of the initial test. Sheep identified as low emitters by PAC tests using the initial protocol did not produce less CH (mg/min) when fed a fixed level of intake in RC. Correlations between estimates of an animal's CHawt measured in PAC and CH adjusted for feed intake in RC were quite low ( = 0-19%) and significant ( < 0.05) in only 1 test of unfasted sheep. With moderate repeatability over the 5 yr, PAC tests of CHawt could be a viable way to select for reduced emissions of grazing sheep. As well as exploiting any variation in MY, selecting for reduced CHawt in PAC could result in lower feed intake than expected for the animals' liveweight and might affect the diurnal feeding pattern. Further work is required on these issues.

The impact of selecting for increased ewe fat level on reproduction and its potential to reduce supplementary feeding in a commercial composite flock

In southern Australian grazing systems, energy availability typically exceeds energy requirements. Grazing systems are likely to have higher profitability if ewes can utilise this relatively cheap feed by gaining more condition (muscle and fat) and then mobilising it when feed is expensive. The present paper focusses on the importance of genetic merit for fat on the lifetime productivity of the ewe within a commercial operation. The analysis was carried out on a maternal composite stud flock in Holbrook, New South Wales. Ultrasound fat and muscle depth were measured on 2796 ewes as lambs at post-weaning and as adult ewes, to determine the genetic relationship between young and adult body composition. The hypothesis of the paper is that selection for increased fat at young age will improve body condition of adult ewes, which will lead to improved reproduction and potentially reduced requirements for supplementary feeding. Given that the difference in feed cost between times of abundance and shortage is likely to be larger, the system may become more profitable despite being less efficient (owing to increased feed intake). Our results indicated that selection for scanned post-weaning fat and muscle depth in lambs should lead to increased fat muscle and body condition in the breeding ewe because of strong genetic correlation estimates between the post-weaning and adult traits of 0.68–0.99. The influence of body composition traits on reproduction (number of lambs weaned) within the stud flock was weak. However, it is hypothesised that by genetically increasing post-weaning fat depth in ewes, producers could improve profitability by reducing the demands for supplementary feeding.

Low-methane yield sheep have smaller rumens and shorter rumen retention time

In the present study, following the measurement of methane emissions from 160 mature ewes three times, a subset of twenty ewes was selected for further emission and physiological studies. Ewes were selected on the basis of methane yield (MY; g CH4/kg DM intake) being low (Low MY: >1 sd below the mean; n 10) or high (High MY: >1 sd above the mean; n 10) when fed a blended chaff ration at a fixed feeding level (1·2-fold maintenance energy requirements). The difference between the Low- and High-MY groups observed at the time of selection was maintained (P= 0·001) when remeasured 1-7 months later during digesta kinetics studies. Low MY was associated with a shorter mean retention time of particulate (P< 0·01) and liquid (P< 0·001) digesta, less amounts of rumen particulate contents (P< 0·01) and a smaller rumen volume (P< 0·05), but not apparent DM digestibility (P= 0·27) or urinary allantoin excretion (P= 0·89). Computer tomography scanning of the sheep's rumens after an overnight fast revealed a trend towards the Low-MY sheep having more clearly demarcated rumen gas and liquid phases (P= 0·10). These findings indicate that the selection of ruminants for low MY may have important consequences for an animal's nutritional physiology.

Ruminant enteric methane mitigation: a review

In Australia, agriculture is responsible for ~17% of total greenhouse gas emissions with ruminants being the largest single source. However, agriculture is likely to be shielded from the full impact of any future price on carbon. In this review, strategies for reducing ruminant methane output are considered in relation to rumen ecology and biochemistry, animal breeding and management options at an animal, farm, or national level. Nutritional management strategies have the greatest short-term impact. Methanogenic microorganisms remove H2 produced during fermentation of organic matter in the rumen and hind gut. Cost-effective ways to change the microbial ecology to reduce H2 production, to re-partition H2 into products other than methane, or to promote methanotrophic microbes with the ability to oxidise methane still need to be found. Methods of inhibiting methanogens include: use of antibiotics; promoting viruses/bacteriophages; use of feed additives such as fats and oils, or nitrate salts, or dicarboxylic acids; defaunation; and vaccination against methanogens. Methods of enhancing alternative H2 using microbial species include: inoculating with acetogenic species; feeding highly digestible feed components favouring ‘propionate fermentations’; and modifying rumen conditions. Conditions that sustain acetogen populations in kangaroos and termites, for example, are poorly understood but might be extended to ruminants. Mitigation strategies are not in common use in extensive grazing systems but dietary management or use of growth promotants can reduce methane output per unit of product. New, natural compounds that reduce rumen methane output may yet be found. Smaller but more permanent benefits are possible using genetic approaches. The indirect selection criterion, residual feed intake, when measured on ad libitum grain diets, has limited relevance for grazing cattle. There are few published estimates of genetic parameters for feed intake and methane production. Methane-related single nucleotide polymorphisms have yet to be used commercially. As a breeding objective, the use of methane/kg product rather than methane/head is recommended. Indirect selection via feed intake may be more cost-effective than via direct measurement of methane emissions. Life cycle analyses indicate that intensification is likely to reduce total greenhouse gas output but emissions and sequestration from vegetation and soil need to be addressed. Bio-economic modelling suggests most mitigation options are currently not cost-effective.

Adaptability of pregnant Merino ewes to the cold desert climate in Nevada

Grazing ability is difficult to record in animals under free-ranging conditions without sophisticated methods. Alternatively, grazing ability may be indirectly inferred from changes in BW and production characteristics during the grazing period. The present study investigated the effect of grazing on resource-limited rangelands on BW, wool characteristics, and offspring weaning weights in nine hundred five 5/8, 7/8, and fullblood Merino ewes of 2 to 7 yr of age during a grazing period of approximately 2.5 mo (between January and March). A total of 469 ewes gave birth to a single lamb, 248 to twin lambs, and 188 did not give birth. Body weights were measured and wool samples taken before and after the ewes were allowed to graze freely on the rangelands; absolute change in BW and change in BW as a percentage of initial BW were estimated. On average, grazing on resource-poor rangelands resulted in BW loss, a reduction in fiber diameter and its CV, and increased staple length. Animals with finer wool at the start of the grazing period lost phenotypically (r = -0.07, P < 0.05) and genetically (r = -0.23, P < 0.05) less BW during the grazing period and had a greater probability to carry 1 lamb (or 2) to term (P < 0.05). Animals that lost less BW produced more greasy fleece (r = 0.09, P < 0.01). Body weight change did not significantly influence offspring weaning weights. Change in BW was moderately heritable at h(2) = 0.29; fiber diameter was strongly heritable at h(2) = 0.51. Animals with the least inclusion of Merino genetics lost more BW (P < 0.01) during the grazing period and had a more uniform fiber diameter (P < 0.05) but shorter staples (P < 0.05) and less fleece (P < 0.0001) than animals with a greater level of Merino genetics. Our results indicate that animals with finer wool appeared to be better adapted to the cold Nevada desert. Thus, selection for finer wool may positively influence adaptability to resource-limited cold climate conditions; alternatively, BW change may be selected for directly. Because nutritional deficiencies during pregnancy can have adverse consequences for the offspring, indirect selection for grazing ability would foremost result in healthier ewes that can produce lambs and wool without compromising their welfare.

Heritability of feed intake in grazing Merino ewes and the genetic relationships with production traits

The feed intake of 1279 mature non-pregnant and non-lactating Merino ewes grazing pasture was estimated by faecal marker dilution methods using rumen controlled-release devices containing chromium sesquioxide capsules. The ewes were from two genetic resource flocks (QPLU$ flock at Trangie in NSW and a Western Australia flock at Katanning) that are representative of the major bloodlines and strains in the Australian Merino population. Pedigree information over several generations was used to genetically link other animals from the resource flocks that had additional production data to allow estimation of genetic correlations between feed intake and production traits with greater accuracy. Data were available for over 16 000 records for growth, wool and reproduction traits from the ewes and their relatives, together with carcass and meat quality traits from over 4000 rams that were slaughtered at ~18 months of age. Feed intake of the ewes was significantly affected by the reproductive status of the ewes at their previous lambing and feed intake, expressed as relative digestible dry matter intake (rDDMI), had an estimated heritability of 0.32 ± 0.08. The genetic correlations between rDDMI and growth traits were positive and larger than their standard error for birth (0.24 ± 0.12) and hogget (0.20 ± 0.09) weight, although inclusion of ewe liveweight as a covariate in the model reduced the correlations to close to zero. Generally, the genetic correlations between rDDMI and the wool, reproduction, carcass and meat quality traits were close to zero and smaller than their standard errors, with little effect of including ewe liveweight as a covariate. The results suggest that feed requirements of ewes could be reduced by selection, although ewe weight and growth would also decline unless accounted for in the model. Practical and cost effective methods of measurement of intake under grazing conditions need to be developed.