Accuracy of Field Estimates of Deer Food Habits

Abstract

The minutes method of determining food habits was tested with tame mule deer (Odocoileus hemionus) that were simultaneously observed at close range. More use was attributed to shrubs as a class and less use to grasses and forbs than actually occurred. Identifiability of individual species varied with plant size, distance, and observers. Results suggest that the observer must be within 75 feet (23 m) of the deer to identify more than 80 percent of the grazing correctly to species. The feeding method was tested by counting instances of use on a sample of 1-foot2 (30 X 30 cm) plots after the site was grazed by tame deer. Use of shrubs, as a group, was underestimated, grasses overestimated, and forbs underestimated. Error with individual species stemmed largely from improper size and distribution of the feeding-site sample with relation to the distribution of plants and grazing use. J. WILDL. MANAGE. 37(4):556-562 Food habits of wild ungulates are often studied by observing the foods taken as the animals graze, or noting the locations where animals graze and later inspecting the site to see which plants were used. The first of these has been called the minutes or seconds method (Hahn 1945, Buechner 1950) and the second the feeding method (Lovaas 1958). Quantification in the first is in units of time spent grazing on each species, and in the second is in in- stances of use-that is, numbers of plant parts (such as stems) severed by grazing. To determine how much precision is pos- sible with these methods, we compared them with close-range observation of tame mule deer. Quantification was in numbers of bites. Within the limits of the observer's ability to correctly identify plants, it is possible to record kinds of forage used by tame deer essentially without error. With our deer, the observer could approach within inches of the animal's mouth and gently move its head away, if necessary, to see the plants being grazed. Samples of plants that are unidentifiable during grazing can be col- lected for comparison with herbarium speci- mens. We have found the variability be- tween observers in counting bites to be about -1 percent in samples of 1,000 bites. It should be emphasized that none of the three methods provides a direct measure of intake in weight or volume. We have hand- plucked samples of simulated bites to esti- mate weights of forage ingested and have found the results to vary little from relative intake based on bite counts.