Proc. 57th Southern Pasture and Forage Crop Improvement Conference, Athens, GA April 23-25, 2002
Measurement of Animal Responses in Grazing Experiments
Glen E. Aiken
USDA-ARS,
Dale Bumpers Small Farms Research Center,
Booneville, AR
Introduction
Funding restraints have caused a steady decline in grazing research over the last couple of decades. Grazing programs focused on improving efficiencies of forage-based livestock production are challenged to overcome the strain of fewer graduate students that are interested in careers in production agriculture, reduced labor pools, and minimal funding from the public sector. Although the situation could be considered by many as bleak, interest by producers in grazing research results has not necessarily waned. There are many retirees from city jobs that are moving back to their rural roots and express an eagerness to seek new technologies. On-going changes in the beef cattle industry will also increase the need for grazing research. One such change is the increased number of cattlemen that retain ownership of their cattle from pasture to feedyard and subsequently market their cattle based on value rather than liveweight. These producers have specific needs and concerns that can be addressed through the development of grazing experiments that follow innovative protocols.
The steady decline in resources devoted to grazing research has produced a situation where “cutting corners” may seem necessary, but consequently would cost both the profession and cattle industry. It is imperative that grazing researchers follow strict standards in the collection of animal data to assure that research results provide both valid and worthwhile conclusions.
Animal Allocation to Pastures
Although cattle performance (i.e., weight gain, milk production, reproductive performance, etc.) is the primary focus of most grazing research, the pasture is always designated as the experimental unit. An assumption of statistical analysis procedures is that experimental errors are random and independent (Steel and Torrie, 1980). Cattle grazing within an experimental pasture cannot be assumed independent because grazing by one can often affect grazing by another.
Evaluating cattle performance as a pasture response and individual cattle as sample units necessitates that pasture means are determined using a sufficient number of cattle that are correctly allocated to pastures. Pasture sizes should be large enough to accommodate at least three tester cattle. Uniformity of tester cattle is critically important, but unfortunately is rarely achieved. Experimental pastures with disproportionate body weights, frame and/or muscle scores, body conditions, and breed types can escalate experimental error. Blocking by body weight for allocation to pastures is a routine practice, but further stratification by other characteristics may be needed. A particular problem is when a small number of testers are
distinctly diverse from other tester cattle, but are in too small a number to be equally allocated to all pastures. These cattle can potentially reduce precision to a level below what is needed to detect treatment effects and, therefore, should be removed from the experiment.
Weighing Errors
Weighing errors represent a major source of experimental error in weight gain data; however, weight measures that follow consistent procedures over subsequent weigh days can alleviate much of the error. Mott (1959) concluded that variation in gut fill can escalate error if weights are not properly measured. Coffey et al. (1997) measured unshrunk body weights at daybreak and at three subsequent 1-h intervals for steers grazing smooth bromegrass. Body weight attributed to gut fill gradually increased over a 3-h time period that started at daybreak, and differences between unshrunk and shrunk (16-h fast from feed and water) body weights also showed to increase over time. Cattle should therefore be weighed at the same time on each weigh day to reduce variation due to gut fill. Weighing as early as possible could also reduce the proportion of body weight that is gut fill.
The researcher has a choice of measuring body weights that are either unshrunk or shrunk by fasting from feed and water for a specified period of time, typically 12 to 14 h. Shrinking can reduce daily fluctuations in weight that are related to gut fill, which can be 12 to 23% of an animals body weight (Hughes, 1976). Mott (1959) concluded that gut fill is influenced by animal genetics (i.e., rumen capacity) and environmental conditions. A concern in relying on unshrunk weights has been that gut fill could mask treatment effects, particularly if the treatments have an influence on dry matter intake. Bransby and Kee (1991) measured shrunk and unshrunk weights for two experiments that evaluated stocking rate effects. Shrunk weights provided slightly lower experimental error and ADG than with unshrunk weights, but shrunk and unshrunk weights were not statistically different. Brown et al. (1993) reported gut fill was higher with cattle grazing tall fescue than those grazing bermudagrass, but there were no statistical differences between shrunk and unshrunk weights.
Unshrunk weights apparently can be measured to provide adequate precision to detect treatment effects. The precision could be seriously compromised, however, if unshrunk weights are not consistently taken at the same time for all weigh days. The impact that gut fill has on average daily gains can also be eliminated by avoiding short-term weight changes. Mott (1959) demonstrated that standard errors and coefficients of variation for average daily gains were reduced as length of grazing period was extended from 28 to 168 d and further concluded that errors with 28-d grazing periods were too excessive to provide reliable data. Regardless if shrunk or unshrunk body weights are collected, weighing errors cannot be completely eliminated but can be minimized if weigh days follow consistent protocols and grazing periods are of sufficient length ($ 56 d).
Variable vs. Fixed Stocking Rates
The most important consideration in preparing protocols for grazing experiments is determining the grazing intensity (or intensities) that will be used or evaluated. A primary reason for this is due to typically strong interrelationships between grazing intensity, forage mass, and animal performance. This was demonstrated by Bransby (1988), who used five bermudagrass hybrids to evaluate relationships between stocking rate and forage mass and between forage mass and average daily gain. Stocking rate and forage mass both interacted with hybrid, but linear slopes for the relationship between stocking rate and forage mass ranged from -1.14 to -2.73 MG/ha and the slopes for the relationship between forage mass and average daily gain ranged from 0.10 to 0.14 kg/d. Thus, stocking rate will affect animal performance through the strong influence that stocking rate has on available forage mass.
The researcher has a choice of evaluating treatments using either variable or fixed stocking rates. Variable stocking can be used under objectives that are established to compare treatments under equal forage masses or allowances. Measurements are collected on tester cattle that reside in the experimental pastures for the entire experimental period. Put-and-take cattle are used to vary the stocking rates to maintain a constant grazing pressure as forage growth and weather patterns change through the experimental period. Fixed stocking rate experiments will establish a range of 3 or 4 stocking rates that remain constant through the experimental period. Forage mass and allowance will therefore vary as forage growth and weather patterns change through the experimental period.
Advantages and Disadvantages
Variable Stocking Rates. An advantage of varying stocking rate is that treatments can be evaluated with a quantity of forage that is optimum for both pastures and animals. Forage is maintained at a quantity that maximizes grazing selectivity and presumably pasture growth. Comparisons in forage quality can be performed between forage species/cultivars under optimum grazing pressures. Another advantage is that observers can develop considerable knowledge of a particular forage regarding growth patterns and responses to weather patterns
A disadvantage of variable stocking rates is that it requires close observation to assure the targeted herbage mass or allowance is maintained. Success in making correct stocking adjustments will be strongly dependent on knowledge and skill of observers. This is critical because deviations from these targets can escalate error and possibly bias results. Another problem with variable stocking experiments is that the scope of the experiment is within a narrow range of managements. Interpretations are restricted to “ideal” grazing conditions and not under- or over-grazing conditions. Also, grazing pressures are subjectively set and maintained through the experimental period.
Fixed Stocking Rates. An advantage of fixed stocking rate experiments is that interactions between experimental variables and stocking rate can be evaluated over a range of low to high herbage allowances. Pasture and animal responses over heavy to light grazing intensities can therefore be ascertained. Another advantage of fixed stocking rate experiments is that the broader scope of grazing managements makes these results more applicable to cattle production systems. Also, once the stocking rates are set the response is random and not subjectively controlled by an observer.
A disadvantage with fixed stocking rates is that applicability of results can be compromised if the intermediate stocking rates are set too high or low and do not represent an actual range of low to high grazing pressures. Another problem is that fluctuations in herbage mass during the experimental period complicates forage quality measurements. A replicated stocking rate experiment can also require a high number of pastures.
Variable or Fixed Stocking Rates?
Determination of whether to use variable or fixed stocking rates will depend primarily on the objectives of the experiment. Variable stocking rates should be used when making comparisons in carrying capacities and animal performance between different treatments grazed under optimum grazing pressures. This approach should be used for evaluations of treatments with objectives to compare animal performance as affected solely by forage quality. Fixed stocking rates should be used with objectives to evaluate the interaction between treatments and stocking rate. Animal performance with a fixed stocking rate experiment is therefore influenced by an interrelationship between forage quality and quantity.
Research Needs
Changes in the cattle industry are causing some shift from the traditional marketing chain that is based on liveweight to one that is based on value. Retained ownership from pasture to packer will establish management goals that target a certain body weight at the conclusion of pasture backgrounding. This “ideal” body weight will be one that provides efficient weight gain in the feedyard and, upon harvest, will yield high retail product and a hot carcass weight that is within an acceptable range. Furthermore, economics of retained ownership will dictate that animal numbers per unit land area will have more relevance than body weight per unit land area (i.e., desirable to maximize number of cattle on feed).
The most beneficial grazing research has typically been done through the combined efforts of forage agronomists and ruminant nutritionists. Future grazing research should include cattle breeders and meat scientists in these collaborations to strengthen the technical value of the work. It is likely that average daily gain will remain as the most meaningful response to pasture treatments, but additional responses might prove worthwhile. Measures of body condition, composition of the gain, or frame and muscle scores could provide useful information when research objectives are to determine carry-over effects from pasture to feedyard and ultimate carcass quality and yield.
References
Bransby, D.I., B.E. Conrad, H.M. Dicks, and J.W. Drane. 1988. Justification for grazing intensity experiments: Analyzing and interpreting grazing data. J. Range. Manage. 41:274-279.
Bransby, D. I., and D. D. Kee. 1991. Influence of shrunk and unshrunk cattle weights on results from stocking rate experiments. p. 186. In Agron. Abstr. Denver, CO.
Brown, M. A., G. E. Aiken, A. H. Brown, Jr., W. G. Jackson, and J. R. Miesner. 1993. Evaluation of overnight shrinkage in reduction of variability of weights and gains of beef cattle. Prof. Anim. Sci. 9:120-126.
Coffey, K. P., F. K. Brazle, J. J. Higgins, and J. L. Moyer. 1997. Effects of gathering time
on weight and shrink of steers grazing smooth bromegrass pastures. Prof. Anim.
Scientist 13:170-175.
Hughes, J. G. 1976. Short-term variation in liveweight and reduction of its effects on weighing. Anim. Breeder Abst. 44:111-118.
Mott, G. O. 1959. Symposium on forage evaluation: IV. Animal variation and measurement of forage quality. Agron. J. 51:223-226.
Steel, R. G. D., and J. H. Torrie. 1980. Principles and procedures of statistics: A biometrical approach. 2nd Ed. McGraw-Hill. New York.