Relationships among variables indexing selective grazing behavior of goats

Abstract

Portable computers presently allow documentation of real-time grazing behavior under field conditions. Researchers are still challenged, however, with the task of selecting proper and easily recorded variables to test their hypotheses. The objective of this research was to examine relationships among variables describing grazing behavior and forage utilization by goats to identify the most easily monitored activity capable of accurately ranking their relative preferences among forages. Relationships were investigated in grassy paddocks and on native shrub/steppe rangeland. Variables recorded for each forage included: total visits, total bites, bites per visit, total grazing time, time per visit, and bite rate in both environments, and number of plants grazed, number of plants regrazed, number of regrazing events, percent utilization, grazed plant utilization, and number of plants utilized in the grassy paddocks. In grassy paddocks, with eight different forages available, total visits, total bites, total time, number of plants grazed, number of plants regrazed, and number of regrazing events were highly correlated ( r ≥ 0.94 and P < 0.01) during both the boot and dormant stages of phonology. Rank correlation coefficients ( r s) among forages with these variables averaged 0.96. Any of these six variables would probably give satisfactory ranking of relative preferences if the animal in question was confined to or elected to forage from a single life form (grass, forb, shrub, or tree). Utilization data were significantly ( P < 0.01) but poorly correlated (mean r = 0.74) with total bites, total visits and total time. Correlations among grazing behavior variables were all statistically significant ( P < 0.01) but much weaker in shrub/steppe vegetation where a variety of life forms were available (mean r = 0.45), Only total time versus total bites and bites per visit versus time per visit exhibited correlations of 0.94 or greater ( P < 0.01). The poorer correlations obtained on rangelands were attributed to a wide array of foraging techniques used by the goats to graze the various grasses, forbs, shrubs, and trees. Total time was the only variable consistently correlated with total bites in both the grassy paddock and shrub/steppe pastures (mean r = 0.94). Monitoring of grazing time devoted to each forage could be more easily accomplished than the more attention demanding process of tallying individual bites. We suggest that adequate rankings of relative forage preference of goats could be obtained by a single observer recording grazing time expended on each forage.